I am starting a campaign to lobby the gods of this site to...

« Back

1591 days, 16 hours, 15 minutes ago

I am starting a campaign to lobby the gods of this site to...

I am starting a campaign to lobby the gods of this site to implement Big Beefer's Die Hard ladder system for the whole site.

If you are not familiar with Die Hard ladder system you can see it here: http://planets.nu/_library/diehard/diehard.html

This system is awesome, not only does it rank players by skill it also provides achievements that could be added to player profiles to make them more interesting.

Please support this idea as I feel it would be good for the site as a whole.

Link to vote for it:
http://planets.uservoice.com/forums/136520-general/suggestions/7487350-implement-big-beefer-s-die-hard-ladder-system-for

1591 days, 16 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Vote added, and hope it gets serious consideration.

1591 days, 14 hours, 31 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I gave up 3 votes to support this!

୧〳 ” ʘ̆ ᗜ ʘ̆ ” 〵୨

1591 days, 13 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I keep seeing suggestions for a "Ladder" system, yet when I look at the Ladder link on the DieHard page, what I see doesn't make sense. Perhaps this needs a description of what puts individual players higher/lower in the list.

I think the current system would be improved if the "Achievement Score" were multiplied by a "difficulty multiplier" that was based on the rank of the players in the game, and mineral/native resources available. With a change in the Achievement points needed to be promoted to specific ranks, this would keep Beginner games from creating Admirals. I believe that many of the pieces for this are in place already, making it a relatively simple change.

1591 days, 13 hours, 30 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer the current system already does what you say however it still rewards quantity rather than quality.

I am also not advocating changing the current system - that is a discussion for another time and place.

This is about adding this as a complimentary system to enrich player data / profiles and give "spicing up" to by introducing achievement badges.

In the link I provided click on ladder. The list there list players sorted by average finish and their score and any achievements they have such as wins by race and solo wins.

1591 days, 13 hours, 26 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

We had a lively discussion about 9 months ago when Dotman came up with a rough draft for a ladder system.

The link is: http://planets.nu/#/activity/1427142

1591 days, 13 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Azzazzello,

> @Whisperer the current system already does what you say however it still rewards quantity rather than quality.

If that were the case, winners of the Beginner Melee games would not be promoted to Admiral. As they are, something's wrong with the current system. I expect that when the current system is fixed, it will do a much better job of rewarding Quality play, or at least successful play against high ranking players. Of course, with the ranks so screwed up, it will take a while (several years) to straighten out :(

You have full access to all the player histories through the API. Have you tried to assemble them into a ladder, just to see what it would look like? There's no reason why this couldn't be done off-site. I expect that if the results end up working better than the current system, Joshua & Co. might be interested in it.

1591 days, 13 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer: http://en.wikipedia.org/wiki/Ladder_tournament

I'm not aware of the specifics either... but I was surprised when I first came to Nu and realized it wasn't a game ladder and just a ranking based on accumulation of points (as /\/\ule likes to mention).

Add in the Diehard results (so were not at square zero), and Big Beefer stamp... and you have my votes.

I looked up a thread on it with a few more details: http://planets.nu/discussion/die-hard-ranking-ladder

1591 days, 11 hours, 20 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I rather like the present system, but even I would like to see it somewhat modified. My impression had been that the Difficulty Modifier was somehow already employed in awarding Achievements, and Tenancy (formerly known as Tenacity) is likewise added to the mix.

It would be nice for the Wars to be peopled only by players who regularly won games. The converse, however, must be considered: that any player who has played a great deal must have SOME competence, and that, were it not for a measurable reward for those who play competently and steadily, we'd lose a lot of players.

My impression is that the next set of Wars will be different than those we've seen. It's too bad, in a way, because in Capricorn we've got a serious set of extreme competence. One example is Siberian Snake, who wrote guides in the 90s that are presently archived at Donovan's. Look at his score and you might think he's a poor player; he's certainly not. J-Zan is competent solidity personified; SMN is not without flashes of genius; Othrym is legendarily clever... I could go on.

My point is that the present wars are a stunning vindication of the present system. Leo, you'll notice, is a long way from the end, and Capricorn is still well-balanced over 200 turns from its beginning.

So cry for change, if you will. Me, I'm not convinced we need it.

1591 days, 6 hours, 16 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer: http://planets.nu/_library/2014/6/nuelo.html

@Gnerpkh: Ladder fixes also standard games when more competent games can compete against others.

And it gives rewards to every player finishing the game, it will be crucial to finish high rather than just lay low and die or wait for the shit to be over.

And in my opinion with ladder you don't need tenacity. Without tenacity you'll simply drop when you quit/drop and are assigned to the last position.

Its too bad people don't see the light in the ladder system.

1591 days, 5 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Oh, I dunno; I rather approve of the achievement system, warts and all. Most of the big holes have been knocked out of it, and all that's left is a quantified measure of players that's somewhat representative of a combination of ability, applied luck, and experience.

What's more, the old Tenacity system (rather than the present Tenancy system) served nicely as a measure of a player's willingness to complete his turns and stay in a game; simultaneously, it encouraged the wavering just a little while strongly discouraging serial droppers.

The difficulty with a ladder system in our present structure is that it would attempt to quantify player rankings based on a necessarily small sample in some cases and a rather large one in others. It's not statistically valid to analyze both data sets using the same non-cumulative method; likewise, it would be inappropriate to alter that method to suit cases because it would remove any basis of comparison. While it likely is an ideal analysis tool for Die Hard games, it would generate little of relevance for a large yet shallow population.

We're concerned with precision for the top slots in a race rather than the pure difference between #142 and #143. There are more coarse methods available to us for that.

https://www.youtube.com/watch?v=CD8EtvWW8nw

1591 days, 4 hours, 56 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

I've no interest in top slots of a race; they can always be argued.

I'm interested in fixing the "regular" games and all analysis point that ladder system would do that better than tenacity, difficulty factor and achievements combined.

After all, most players enjoy regular games and even more are playing just those. I would not put effort in fixing Championship games for a small minority but as a side-effect ladder would fix even that :D

1591 days, 4 hours, 40 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If the top slots don't matter, why would we bother with #142, which also would be imprecisely applied? That's kinda where I'm sticking on this.

And it's a serious objection, Ville. My reasoning isn't so weak as to be dismissed by a swipe at my character, as though participating in a championship, two experimental games and a Mentor game were somehow shameful.

I state that any non-accumulation system would be inherently invalid for some pretty good reasons. Now, that doesn't mean I'm right, but it surely does mean that my arguments ought to be addressed rather than me.

1591 days, 4 hours, 37 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Are you sure you did read the http://planets.nu/#/activity/1427142 - discussion about the ladder-example I gave ?

It is not 100% perfect ladder and needs minor tweaking but it is EXTREMELY close.

1591 days, 4 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I read it all at the time (and skimmed it earlier today; nothing new) and decided I would avoid posting what was certain to be an unpopular conclusion when people were building things. I see no reason to discourage initiative -- and in time, if it were kept updated, I feel there would be enough data for some of us to be appropriately rated by a more complex system such as the ELO.

But my objection remains: We're talking a very small sample size per player. No matter how complex the formula, it's going to be virtually impossible to rate individual performance based on such limited data.

It would be possible to rate a player based on their per-game performance, however, analyzing the planet and military score (relative to ship count) curves to generate a richer (albeit simplified) data sample. The difficulty is that it's tough to determine from the outside what a player ought to have done at any given point. What truly is the correct move?

I know people that are excellent at chess when playing Queen's Gambit games but who collapse against a well-played French Defense. Are they good or bad players?

The best answer to that class of question is: It's more complex than that.

1591 days, 4 hours, 24 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

In chess there is only winner and loser. In 11 player games I see a lot more positions. Currently we favor only the winner which, in my opinion, is plainly stupid.

As said, finishing 4th or 5th is not a bad performance. Finishing 10th or 11th is.

1591 days, 4 hours, 20 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

By the way: I would dispute that placing 4th or 5th is "not bad". In my opinion, failure to win is bad; winning (or allying with the winner) is the best measure of success. I maintain that no player should be eligible for a championship game who has never won a game (either of the two victors would be fine by me).

If a rating system were to include in-game rankings of the play quality of the other players, it might be more representative, but there we have bias; after all, the backstabbed might possibly object to the backstabber's play style. Winners are apt to be arrogant and losers bitter, but we cannot reliably analyze subjective data by pre-determining bias.

1591 days, 4 hours, 14 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Winning a game is overrated. There are lots of dropfests and colonizing runs around.

What comes to failing a win I think you are just looking through coalition eyes; every strong player should face a coalition and be brought down. If you fail to do that, you are not worthy.

But that form of gaming is too exhausting for most and eating broken glass throgh rear-end can easily be more fun than running in a coalition.

1591 days, 3 hours, 12 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's a perspective, sure, but it doesn't answer my objection.

1591 days, 3 hours, 4 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

By default ladder manages play quality of other players automatically. If you don't play well, you don't place well. I you don't place well, you'll drop in rankings. Achivement, difficulty factor and/or tenacity have nothing to offer as long as you keep on pressing end turn and don't get eliminated.

Doesn't really matter HOW you play, are you a notorious backstabber or a knight in shining armour, ready to aid damsells in distress.

1591 days, 0 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

As a system, it seems to make good sense. I think that, in time, we ought to adopt that or a similar rating tool.

Nevertheless, I still don't see how it can be valid for any small sample where some players concentrate on a single game and others play at capacity all the time. The former will maintain a consistently low rating; the latter will pass their peak and be artificially low (though higher than the single-game players). The highest and lowest on a ladder are always those players who are in a recent streak, whether a losing or winning. Streaks are statistical inevitabilities and will always be inaccurate; they're also more common than one might think in any random sample. Fully a quarter of all players in a two-option system will have a 100% streak in a three game set; the number is far larger for a one-in-eleven or two-in-eleven mix. Oddly enough, statistical normalization will fail in an ability-modified measurement system, wherein players will defy deviation and consistently finish in similar positions.

While I understand the innate difficulty in a non-weighted achievement system in representing the abilities of non-victorious yet consistent players, this weakness grants no corresponding virtue to an alternative. That's like saying, "Well, the Democrats suck, so I'll vote Republican." In truth, they both suck but in different ways.

That's why I think Dotman's (WAS it Dotman's? Too little time to check) system was interesting, in that it weighted performance relative to the opposition rather than straight planet score. However, a better system would have to involve not a ladder at present (since it's statistically invalid) but an accumulative points-driven score. Logically, we have to end up with something similar to our present scoring.

Instead of a fundamental alteration, instead, I would advise a logarithmically-weighted planetary awards scale with linear multipliers for Daily Grind, Tenacity, and a relative Difficulty Modifier. The log factor would be for placement, such that the difference between eleventh and fourth place would be small but measurable, and it would be modified by high placement for the winner(s). In addition, I'd drastically reduce destruction and capture rewards and add a small bonus for successful ground assault captures relative to the duration of possession preceding the capture.

But that's my preference. I'm sure other people have their own ideas.

1591 days, 0 hours, 39 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

You think it way too complicated.

If you look at the list http://planets.nu/_library/2014/6/nuelo.html there are several players with quality over quantity.

1591 days, 0 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Any complex system demands a complex solution. Not everything is a Gordian knot; that is merely the exception that proves the rule.

1591 days, 0 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Gnerphk, ever thought about a spare-time job as an orator ? :-)

1591 days, 0 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm available, but I'm expensive.

1591 days, 0 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Thats what I always hear ;-)

1591 days, 0 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

:o) A professional doesn't come cheap, mate -- you get what you pay for. And I'm worth every penny.

1590 days, 21 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm with Ville on this. I've fought against apparent "no names" in terms of achievement who were amazing players, and I've fought championship experienced players who seemed quite mediocre.

Some means of rewarding quality players so they rise to the chance to compete in the championship game rather than just "play them 10 times to get to the top rank" would be beneficial.

Also, as to the "sour grapes" theory of a subjective ranking. I can assure you, though we may feel bitter when initially getting our asses handed to us, those of us that are beaten by a superior foe are, for the most part, more than willing to praise their skills, as it means we lost to a superior foe.

1590 days, 20 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Furey no question about it - I have made some amazing friends with people I fought over the years where I did not do so well (J-Zan, Commander Koshki) and in other cases if I did not become friends, I did develop a healthy amount of respect. I think you learn more about a player fighting him than being allied with him.

The sour grapes (I hate that name - it is more of loss of motivation) is something real but has to be dealt on personal level and very little can be done on the site level. I play to win and when it becomes clear that my goal is not achievable is difficult to accept and motivate myself to push through. I gotten better and for example in one of my games where I am losing I made a point to punish the aggressor and take away his win and the game became fun again :)

1590 days, 20 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I still don't know why the two systems can't coexist on the site. Championship games can continue to be based on the achievement leaderboard, but give players a chance to see who's who on the ladder.

Although Planets.Nu added a Mercenary Leaderboard several months ago it's still true that the current system rewards players who focus on one race and stick with it exclusively. In my opinion that's a shame. You can't truly understand your enemy unless you walk in their shoes for a few months, and that can't help but improve your overall skills in Planets.

1590 days, 20 hours, 10 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Tom N:

Exactly. When you walk into a championship or any senior game and find players not being able to build 2nd base before shiplimit or are completely clueless how web mines work (or how dangerous a developed crystal is) it really undermines it a bit.

I do agree with E-mork that in every game there are performers and underperformers but that unfortunately makes it a lottery who happens to be next who and/or gets someone easily to do their bidding.

1590 days, 20 hours, 10 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Consider 2 players...

A plays only plays 3 games and finishes #2 and #3 and #4 - finishing all of them and having a average finish of #3.

B plays 20 games and finishes #1 once (maybe with help from his ally and maybe because of poor competition.) He finishes poorly or drops every other game and has an average finish of something like #8.

I am no number cruncher and haven't analyzed the system in a while but I believe the current system would list B as the "better" player if one assumes the points accumulation points in that direction. But, I would suggest player A is really the better player.

As far as any reference to championship games, while I think that subject may be somewhat related, for many of us is NOT the goal or the reason for this discussion. We just want the ranking/rating system to more accurately reflect RELATIVE skill level and not an just an accumulation of (possibly cherry picked) points. If you also want to select for a Championship game based on that... that's OK too.

1590 days, 20 hours, 6 minutes ago

olegboleg

RE: I am starting a campaign to lobby the gods of this site to...

*subscribe*

1590 days, 20 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> You can't truly understand your enemy unless you walk in their shoes for a few months, and that can't help but improve your overall skill at Planets.

Agreed, but that will show up in the current system as increased planet Achievement and improved position at end of game (more bonus Achievement), and probably more wins.

--------------------------------------

The Ladder may be useful (I have no opinion on this, one way or the other), but it has not been demonstrated that it's better than what we currently have, on the scale of the full site. Without that demonstration, I believe that the chances of seeing Planets.nu implement this is rather small.

Back to my original suggestion - Write the code and test it out on the full site. If it works well, Joshua might consider it. If not, it will have demonstrated that it's not worth his time before he wastes any on it. It's not at all unusual for a complex system (any scoring system for Planets.nu is almost certainly going to be complex) to fail to function properly when the amount of data grows too large.

This is a radical change. Unless you can show that it works better than the system that's currently in place, on the full site instead of the very small sample you're currently using, I don't believe Joshua, or any of the developers, should be spending much time working on it.

It may well be a good idea, and it's been shown to work on a small subset of data. It's time to step up to the plate and demonstrate that it works properly with the full set of data.

1590 days, 19 hours, 51 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

This is no change - I am not advocating for replacement of existing system. There is no reason why two system cannot coexist side by side.

Also there is no question about that the current system rewards quantity over quality. Go to leaderboards and look at top 10 for all the races and click on player profiles. You will see quite a few people who I would not consider 10 best players for that race but again this point is mute as I am not asking to replace existing system.

If nothing else, having an achievement badges or medals on the office profile of the official site would be a huge improvement. Big Beefer system did that beautifully.

1590 days, 19 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> This is no change

I disagree. This is a request that the Planets.nu developers create an entirely new scoring system. Whether or not it replaces the current scoring system is not the issue. The issue is that new code would need to be created (adding complexity to the system), tested and implemented, old games would need to be rolled into the new scoring system, a hook in the game exit code would have to be made, and new pages created on the site. THAT is the change, and it is absolutely NOT trivial.

> Also there is no question about that the current system rewards quantity over quality

Agreed. What you haven't done is demonstrate that your proposed system works any better for the site as a whole.

> this point is mute

Then please mute your comments.

If, on the other hand, you mean the point is MOOT, then you're in error (see my first response).

1590 days, 19 hours, 13 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer - thats a complete load of horseshit. You cannot expect people to do the work of developers but of course they can throw in ideas. Not every player here has a) time b) interest c) knowledge to create real demos through API.

And yes, many players agree that the Holy Triangle of Achiemenvets, Tenacity and Difficulty Factor are not doing a good job.

1590 days, 19 hours, 6 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I suspect (but can't prove) the first person to envision a rocket to the moon was not a rocket scientist. That does not invalidate the original idea.

Nor, as in this case, does a system (whoever codes it) to measure relative skill necessarily invalidate a points accumulation system such as we now have. It WOULD however tell us more about player quality rather than quantity.

I am for it.

1590 days, 18 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The more I hear about a ladder the better it sounds. My only concern is the volume of games needed to get sensible results.

That said, my personal wish for the achievement/tenacity/rank system is that it would
a) reduce bad drops
b) encourage replacements
c) ensure that senior officer games and championship games would have dedicated players who are actually senior.

I think as things currently are a) and b) are addressed rather well. But c) still needs some work.

Utilizing the existing concepts, I'd love to see some limiters on how to achieve in rank and subsequently rank limiters for advanced games.

Something like only getting commodore rank if winning a game and needing commodore for senior officers games. And requiring senior officers game victory for rear-admiral and above, and only allow players from that rank or higher to champs games.

1590 days, 18 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

One benefit of a ladder is the ability to have games with similar player quality, which I'd imagine would do wonders for player retention.

How do we do that now? Join a "Senior Officers" game and play against somebody who has finished 70 games but never built a Starbase?

1590 days, 18 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

In my last senior officers game we had a player who didn't know that taxes could be set higher than 1% on the homeworld. I'm not kidding.

1590 days, 18 hours, 41 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer

For empirical evidence, links to two efforts that tested different ranking methods have already been posted in this thread.

Big Beefer made a test run of Die Hard ranking system applied to all public games. The results were beautiful.

Dotman(?) did the nuelo rankings and while there were still some oddities it was great too.

It has already been proven that better rating systems exist.

1590 days, 18 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

As I have written in the Nu forums a few times over the years, my vote is to have as many good/worthwhile Ranking systems as possible. They are all very interesting and provide some insight. They add to the flavor of the game. And they would all tend to generate interesting discussions to read in one's spare time waiting for the Host to run.

1590 days, 17 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

+1 Bondservant. I enjoy reading the feed when there's interesting ideas and debates about the game we all enjoy. On that front I think we might encourage more involvement from ALL members if there was more concern about the content of posts and less worry about a misspelled word or a missing comma.

P.S. What happened to Ninja Bunny? He (or she) went from several posts to zero! You may not agree with the ideas, but they were certainly entertaining.

1590 days, 17 hours, 30 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ville+Kauppinen,

> Not every player here has a) time b) interest c) knowledge to create real demos through API.

Except that you don't need _every_ player to do this. You only need _one_ such player. Of all the people who support this concept, there should be one with that ability. In fact, most of the API interface code could be taken from the current client, making the task much easier.

> And it is up to developer(s) IF they wish to implement a different scoring system.

Now we're down to an issue of risk, cost and return on investment. How many man-hours will the implementation take? What gets ignored while this idea (never tested at scale) is implemented? What happens if it falls apart under the load of the full Planets.nu user base? What additional load will this new system place on the server? These all need to be considered by Joshua & Co. before they undertake a task such as this. If some of these questions can be answered in a favorable manner, he'll be more likely to perform the task.

> many players agree that the Holy Triangle of Achievements, Tenacity and Difficulty Factor are not doing a good job.

As do I. I'm just not sure that a Ladder will fix the problem. On the other hand, I _am_ sure that a few tweaks of the current system will improve it (probably not "fix" it). I'm also sure that these minor tweaks will have lower risk and cost than implementing the new system being proposed.

Again, if the risk and cost can be reduced, it might be worthwhile for Joshua & Co. to make the investment. The best way to do this is with a full-scale test. While the NUELO test linked above (the only link I found in this thread that comes close to a full-scale test) is a good start on that, it's not there yet. Each player would have to be broken out by race, and rank points would have to be established.

1590 days, 17 hours, 24 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

I seriously doubt that running full database script instead bombarding API will break any site; we've hundreds of thousands SQL-records and creating different shit out of those is quite efficient.

And please, don't act like you own the company or understand something about RoI, risks and/or costs included. We all have our jobs but I like to keep mine separated from my spare-time. Thus I'm not coding anything or testing anything.

1590 days, 17 hours, 18 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Points accumulation will always be points accumulation. Tweeks will not change this. Until the results of all games joined are tabulated and averaged (or some other mathematical process is used) then we will not have a valid measurement of skill. Only, as now happens, a summation of points collected.

Have we missed the point that has already been made that BB and Dotman have both independently demonstrated better systems?

So... why continue on with this argument about who is going to code this or a risk-reward analysis if it has already been done?

1590 days, 16 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

IIRC the way chess handles new players on the ELO system is that their rating is flagged 'provisional' until they've played a certain number of games, and that could be done pretty easily with the system I wrote too; after a certain number of games, the rating is provably statistically valid; of course, larger samples are better, but anyone with over X number of games could be considered as having an accurate rating.

When I wrote that it was a proof of concept kind of thing, and I believe it was just a bunch of python scripts that each did a step in the process. As far as server load, its minimal once it's automated; the system would simply make an API request or 3 once a day, and update the stats. It only needs to know what games have finished since the last time it ran. I didn't move forward at the time because I didn't have anywhere to host it. But impact on server is minimal; my planetary management plugin eclipses any elo system in JSON requests against the server in a single session; its no big deal.

Development is minimal; I believe I wrote that proof of concept in an afternoon. But there's no need for the site people to do it, it can be 3rd partied, thanks to the API.

Interesting discussion. If nothing else, I think a ladder system is interesting to look at, and certainly at least as valid as the achievement system (I would say much more so, but at least will do) as a ranking.

1590 days, 16 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I reckon the only realistic way to push this forward would be just doing it as a third party effort, maybe as a separate page maybe as a plugin. Let it run for a few months and get data and feedback. If it seems solid - nothing prevents Joshua/BB from adding similar functionality to the site.

1590 days, 15 hours, 30 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Dotman,

Thank you for the additional information. The system load was the big unknown. While I could guess that it would be low, it would only be a guess.

@Smn,

That's more or less what I've been trying to write. Thank you for creating some clearer wording for me.

@Mule,

> Points accumulation will always be points accumulation. Tweeks will not change this.

Nor have I ever written that they could/would. I've written that they could improve the situation.

> Have we missed the point that has already been made that BB and Dotman have both independently demonstrated better systems?

I've seen the NUELO results, which look encouraging. I haven't seen the results of a full-scale run with race isolation, which is what Planets.nu would have to implement. Or are you claiming that skill at Planets.nu is independent of the race being played?

@Ville+Kauppinen,

> I fail to see how planets.nu database would be any different or less efficient if designed properly.

Apparently you also fail to see that your workplace is funded at a completely different fiscal level than the Planets.nu site.

"speed's just a question of money. How fast you wanna go?"
-- Grease Rat from Mad Max

> please, don't act like you own the company

I have no financial investment in GMI, and have never meant to imply that I do.

> or understand something about RoI, risks and/or costs included

While I don't have any financial investment in GMI, I've made my living from computers for 35 years (mostly as a programmer), and I run a small consulting business, so I do, in fact, understand quite a bit about risks, costs and ROI for software projects.

> if you like to do some volunteer-work for my daytime job I'm happy to offer some

While I don't know what your day-job is, I doubt I'm qualified, as my Finnish (non-existent) is much worse than your English (quite good; better than many of the people here who claim to be in the US or Canada).

1590 days, 15 hours, 16 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer:

IMO a single race player is less skillful than a multi-race player. But, that is a separate issue I would rather not sidetrack this thread with.

I would defer to the guys doing the work (Dotman / BB / or even you) on how best to collect and number crunch race results against individual player results.

If it were me I would ignore race. We all get to make our own choices on that, just like we get to make our own choices once the game begins. If we think we are more skillful at one race over another we have only to select that race and prove it.

1590 days, 15 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I would REALLY REALLY want to see this as a 1st party rather than a 3rd party implementation.

I want to be able to click on somebodies profile and see their medals / badges for outstanding feats such as solo wins, wins with certain race, etc.

With Big Beefer onboard, considering he wrote the whole system as 3rd party, I am hopeful he takes on this plea.

1590 days, 14 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule said > "If it were me I would ignore race. We all get to make our own choices on that"

Here here! Keep it simple. We can all adjust if we end up seeing a Colonies/Rebel ranked higher than someone who primarily plays Birdmen, EE, or Fascsits.

@Azzazzello said > "I would REALLY REALLY want to see this as a 1st party rather than a 3rd party implementation. I want to be able to click on somebodies profile and see their medals / badges for outstanding feats such as solo wins, wins with certain race, etc."

Although I agree, if a 3rd party demonstration keeps the ball rolling maybe this idea won't drift away for 9 months again.

1590 days, 14 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well put, @Tom+N. Heck, if people really support a 3rd party solution, they'll start by placing a link in their profiles.

I still hold that a ladder-based system is statistically invalid given our presumed data set. It would be more correct if we ranked it differently, using in-game metrics instead of just results. Here's some things that we could track which would be of interest:

- Rather than destruction, let's track the ratio of enemy destruction to personal destruction. So as not to invalidate a win through superior logistics, let's also track enemy destruction relative to fleet size.
- Speaking of logistics, let's not track only planet count but also planets + starbases + ships, in similar fashion to the present (underused) Experience metric.
- In addition, I really think a subjective poll that rates us by our surviving opponents would be valuable.

These are methods of adding meaningful data to a system in order to increase its validity. For more examples of skilled warfare, I refer you to Sun Tzu (who isn't entirely valid in our sub-universe, but that's OK too).

1590 days, 13 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The only way to properly gauge a player is to review their game play history and scan the turns. Were they lucky? Did they get a favorable position? Did their neighbors drop? This is how you learn a player's skill level. The current methods are poor, and I have seen little improvement in the newer proposals. Play a lot and get rewarded. Wins so overwhelm poor finishes, that really all you need to do is play a max set of games over and over to get that ideal combination of position, poor neighbors, and good race to set up a win.

I for one do not see how you can factor out race selection either. No slight intended, but I would be far more threatened by a competent Bird player who manages a lot of strong finishes than any Borg Admiral. Some races should get multipliers based upon finished results just because they are so difficult.

I happen to agree with Gnerphk that the rankings don't really reflect actual skill and need more in game data to be accurate.

1590 days, 13 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If we want a really really simple but reliable method for the value of a victory, to give difficulty multipliers to the value of a victory, the turn ship limit was reached is pretty much the only thing that is fairly reliable and can be automated.

1590 days, 12 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> IMO a single race player is less skillful than a multi-race player.

I absolutely agree. So far, I'm best with the Crystals (the first race I played), but in a game where I'm playing as Robots, Toxis142 is teaching me quite a bit about how to do a better job (Thanks Toxis ... I think ;) ) given adequate resources (he has them, I don't).

> If it were me I would ignore race.

While I agree that keeping it simple would be good, sometimes simple doesn't provide enough information, and I believe that the scoring system being suggested here is an example of that. Let me provide an example to clarify this.

For this example, let's assume that the player has played as two different races, Birds and Colonies. These two races require completely different playing styles to do well. The player has three 1st-place finishes (in an alliance) as Colonies, and two FoFs/eliminations as Birds. While his overall rating in the proposed Ladder system might be fairly high due to the three 1st-place finishes (depending on the skill ranking of the other players in the games), it's obvious that this player would be a poor choice to team up with if they're playing as the Birds.

This is one example of why I believe that the races would need to be broken out in the ladder system. Perhaps an overall score and a per-race score would work best. Of course, for any to be valid, the score would need at least two or three games, and possibly more if most of the opponents are also unranked. It would also be appropriate to age off older games, as they should count for less than recent games.

In the end, each race has to be played differently to get the most out of it. Because of that, there will always be a race where the optimum methods most closely align with the player's skill set and natural tendencies. This will be their best race, once they learn how to play it. It would be useful to know, at a glance, which, of the races they've played, a player is best at. I believe that this will help the player and will create a more enjoyable experience for all.

Also, scoring for replacement players and players who resign or are dropped would be more difficult to do. I expect that one of these two kinds of players could "game" the system, but I'd have to look at the code to determine which one.

1590 days, 9 hours, 51 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

No system of Planets scoring will be perfect. One assumes that given enough games many of these chance variables (location, resources, neighbor, etc.) will balance out. Thus we should have a minimum number of completed games before a player's score stops being provisional.

As far as all these in-game metrics GP mentions... I just don't see the need and view it as making the whole thing harder to get done. Let's remember KISS.

IMO at the end of a game the finishing position should be of paramount importance - if not the ONLY thing. If you could finish higher (or first) with no combat at all - that would be SKILL! If anyone thinks race matters that much, maybe they should pick different races to play. Of learn to play the one they did pick better.

1590 days, 8 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, shoot -- if we ain't gonna do it right, why do it at all? At present, we have a system that's far from perfect. Why should we strive for something that's no better?

1590 days, 8 hours, 14 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP, We seem to disagree on what is better, or even important.

I think that collecting data on battles and invasions - etc. might be interesting metrics, but it doesn't come even close to who finishes in what position. At least that is what I play the game for.

As the hiker said when being chased by a bear, "I don't have to be faster than the bear, only faster than you." I would add, "It doesn't matter how many bushes and rocks I jump over while outrunning you."

A ladder-based scoring system that number crunched based on finishing position might not be perfect in all eventualities, but IMO it would be a vast improvement.

1590 days, 7 hours, 41 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I can't see any important difference between an additive system that measures purely based on finishing position and an additive/detractive system that does the same thing. Remove destruction points from the present achievement system and we've essentially got your ladder.

Ville's ladder is different in that it ranks without concern for finishing position. I'm not sure how that helps either, but it shares the same statistical weakness.

The ELO system (Dotman?) is likewise similar.

In short, if I filter out the word "ladder", all I hear is that you're uninterested in destruction points. If I add the value that "ladder" brings to the conversation, I hear that you want to reduce the points earned by players who frequently come in tenth.

These are decent points, and dealt with individually, I believe we could glean their essential merit and use that to create improvements or changes to our present rankings. But I feel it's a mistake to present a system, say "It's better than what we have so let's trash what we have", and then fail to account for its failings. We'd swap one flawed system for another without addressing any root problems.

Instead, I propose we examine all of the root problems from various perspectives. It seems apparent to me that no system based purely on Ville's perspective would appeal to the entire population of the site; likewise, that desire that you and I seem to share, @Mule, of honoring the winner(s) far more than any number of losers is unlikely to appeal to him. I don't propose to resolve this difference in perspective but instead to examine individual, commonly agreed-upon problems and from them design the easiest solutions for each -- or, if we happen to be gifted with some strokes of genius, perhaps even an elegant solution for multiple.

Therefore, I challenge you: Tell me, in simple words, what the problems are with the present system. One idea per post, please, and keep the posts short. We'll discuss them all at length.

Sound useful?

1590 days, 6 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> If anyone thinks race matters that much, maybe they should pick different races to play.

Are you claiming with this statement that you play Evil Empire as well as you play Rebels? If not, then how reasonable would it be to have a single number represent both?

1590 days, 6 hours, 48 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Whisperer - re: "Are you claiming with this statement that you play Evil Empire as well as you play Rebels?"

No. But I'll bet there are some players that DO play EE better than I play Rebels. Everyone is free to select any race in any given game. Just as they are free to make any of all the other choices we make in a game that results in our eventual finish.

However, I suspect maybe you are asking how we figure out who plays what race in a Championship game? I think that is a different subject than a ladder ranking system and not one I'm overly concerned with. BUT... I could have some suggestions on that as well.
--------

GP: Fair enough...

I think before everyone digs out their slide rules (yes, I'm that old), we should first try to get a majority agreement on assumptions and goals. THEN work out the details.

My assumptions:
1) The objective is to finish first, or as close to that as possible, regardless of any other factors.
2) Variables like location and resources should average out over a number of games.
3) Since Race is player selected, just as all in-game decisions leading to eventual finish, it can fairly be ignored.

My Goal:
1) A single numerical rating for each player based on his/her average finish that more or less gives an idea of each player's skill level.

1590 days, 6 hours, 19 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Most people are ignoring the fact that ladder system makes other positions than win or allied win really count too.

In my humble opinion and experience of handful finished games on this site it would be GREAT.

Most games just are boring as hell as there is no incentive trying to finish high.

1590 days, 4 hours, 30 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

OK; fair enough. I think we can safely say that your (3) is going to be controversial, but enough people agree with you that we could maintain parallel scores for per-race and general.

To clarify (1): I presume that we'd take into account win conditions in each game? In other words, that a third-place winner beats a second-place loser.

With regard to the goal: We presently use rank as a primary measure, with Experience and Achievement as secondary stats to generate the rank proper. The only analogue we now have for the single number is the Achievement accumulation, which without the other is indeed somewhat meaningless.

Query: Do you believe we should take into account Tenancy/Tenacity?

1590 days, 3 hours, 38 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk: In ladder tenacity is meaningless. Dropped players/resigned players and/or not paying attention -players naturally finish last or close to it. And finishing below zero line (around 6-7 I'd estimate) means negative points.

So problem solves itself. While getting first remains as goal, also preventing bad position becomes a worthwhile goal not to mention finishing above zero.

So everyone wins.

1590 days, 3 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

>> Are you claiming with this statement that you play Evil Empire as well as you play Rebels?
> No.

That still leaves the question that you've avoided answering _twice_:

How reasonable would it be to have a single number represent both?

This single overall score is what you're suggesting. As I hope you can now see, it would misrepresent your skill both with the Rebels and the Empire. The only way to properly represent each players skill with the various player races is to have a separate score for each race.

1590 days, 2 hours, 21 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Don't think it too complicated. An experienced player can play any race with high propability of success.

While overall score may not represent accurately the skill in certain race, it would encourage players to play more than single race.

Which is also one aspect of game quality I'm sad to say that not many players play all races.

1590 days, 0 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, then -- it seems that we agree, Ville -- any single score would have to be weighted such that a zero score in a single race would have a disproportionately negative impact on a ladder score.

It strikes me that some people may drop to avoid getting a negative end-of-game score. We want to do something about that, I think. Perhaps an automatic 12th place, slightly worse than 11th? Or should it just be a dead loss, an 11th place tie between all droppers and all dead players?

1590 days, 0 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

See, the trouble here is, I still don't think we'll have enough data from most players. I mean, if we base championship entries on any such metric without sufficient data, we're apt to find ourselves with champions who haven't won more than a game or two, but with a dozen drops or Dead finishes.

At present, of course, we have a disproportionate number of champions who've done more than their share of killing as opposed to winning, but that's an innate flaw in an accumulation system. It's the exact flaw that we see in a ladder which has too little data from some players and a great deal from others.

1590 days, 0 hours, 7 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

How can you enter championship game with drops or deads if you are placed last in ladder scoring ? I mean, you do realize that unlike achievements, ladder positions diminish with bad positions.

Imho all droppers, deads and resigners share 11th place = biig penalty, depending of competition.

1589 days, 23 hours, 31 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I think the key is to have a few different ranking systems to allow people to pursue success in the system that means most to them. For instance, none of this conversation really matters to me because I prefer to play classic games. Achievements that net me "new abilities" are not something I am interested in. I would however really appreciate a simple ladder based on finish position in classic VGAP3 based games because that is the competition I'd like to measure myself against.

If there was a second system that ranked players for PQ based games that would be great, and one that awarded achievements for new abilities, great. Basically, there are four or five groups that people fall into with respect to what they value in a ranking system. It wouldn't be too hard to set up a handful of systems based on user input that would be more effective than trying to blend all of the various goals into one system.

K

1589 days, 23 hours, 12 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Good point, @Kelmain. So you think a simple ladder based purely on planet count and relative position is what we should be looking at?

1589 days, 22 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I would prefer not to separate by game type. Typically if someone is really good in campaign games it is because that player is good at planets in general and would do well in classic format as well. If somebody dominates super melds games - chances are they would do well in regular game as well. Only their competition really matters.

If we start segmenting by game type the system gets to complex and we would need many more data points. Let's keep it simple so it is easier to implement and we get meaningful data faster.

1589 days, 22 hours, 21 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I think that would be a simple to implement and uncontroversial addition to whatever other system(s) people find valuable. It is pretty undeniable that rank in games over time will equate to skill. While things like kt sunk and turns played may be affected by skill, they are more directly measures of activity which doesn't necessarily require great skill. Also, in a game where many seasoned players are arriving at this site from other places, there should not be too high a bar to prove up some significant level of skill without being forced to play 8 games at a time to build up rank. Beating players at a higher rank than you is one quick way to see that.

@Azza I don't dispute that players who are good in one system are likely to be good in another. But where you see a few ladders as complex, I see it as simpler.

1589 days, 20 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

There are three possible results for any game we play.

Win - Solo or alliance winner

Draw - Not part of solo or alliance winners, but did not lose.

Loss - Dead, Resigned, kicked or F or F'd.

Ladder is based on these results.

Solo winner gets +15% of the average ranking of all non winners.

Alliance winners get +10% of the average ranking of all non winners

Draw - Get +5% of of the average rating of all losing players - 2% of the average rating of the winner(s)

Loss - Get -5% of the average rating of the winner(s) +2% of the average rating of all losers out on same turn or prior to his/her loss.

Everyone starts with 1000 ranking and we go from there.

Simple. Winners increase their ranking; drawers can go up or down with a bias to increase slightly; Losers can up or down with a bias to decrease slightly.

1589 days, 20 hours, 4 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Tom - nice and simple but it does not give incentives to finish 3rd instead of 4th or 5th. So there should be different levels of draw and loss :)

1589 days, 19 hours, 54 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer: >> Are you claiming with this statement that you play Evil Empire as well as you play Rebels? > No. >That still leaves the question that you've avoided answering _twice_: HUH? I guess I either don't understand the question or why it matters. Yes, I am better at some races than others. So? But, it was still *my choice* what race to play in which game. For better or worse, I am not a one-race-wonder. I am perfectly content to be scored according to my cumulative results. If someone wants to see how well I usually play as EE, they are welcome to look at my EE results.

Kelmain and Azz both make good points. But, I would prefer a single ladder. I don't play campaigns because I prefer Classic, not because I probably couldn't do as well. Whatever our skills are, I think they transfer.

What I don't like about Tom G's scoring is that, like outrunning the bear or just the other hiker, there is no incentive to finish 4th instead of 7th. If a player consistently finishes 4th and another always finishes 7th there is a difference in their skill levels. As far as drops or deads, I think they should be scored in last place *at the time* they exit the game. So the first player to exit an 11 player game is #11. The player that dies the next turn is #10. This would rewards long term survival and address the tenacity issue. What I like about Tom G's plan is that it is solely based on game finishes and addresses alliances.

1589 days, 14 hours, 55 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I think the idea that players get to choose race is somewhat overstated. Games are already hard to get started because no one wants to play the easier and deadlier races. I didn't join the game I won as EE because I wanted to play EE; I joined it because I wanted to play and the game had sat for a long while with no one stepping up to play the EE. Took one for the team.

Some races are almost impossible to join a game as, because of the rush to snatch them. Only team games or the 30 player melee mess offer any chances to play the good races. Regular 11 player games, not a chance, unless you happen to be logged in here right after it is announced.

I also think the idea that only wins count anything only encourages drops. Our mighty emperor once said something I hold to be very true: A good player can tell by turn 20 whether a game is winnable or not. If only wins reward, why would anyone stick around and waste time once it was obvious to them victory was out of reach? The only way to fix drops is to reward finishing. Even to the death. Droppers should always be ranked below folks that fought to the bitter end.

On the flip side, what about someone who is a considerate dropper? Someone who has real life happen and finds a replacement to takeover his spot? Still hand them an awful penalty? Treat them the same as the Ninjabunny types which join 10 games and then drop all 10 a few turns in? Not fair in my mind.

1589 days, 14 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@lord+pollax
I completely agree to this

On the flip side, what about someone who is a considerate dropper? Someone who has real life happen and finds a replacement to takeover his spot? Still hand them an awful penalty? Treat them the same as the Ninjabunny types which join 10 games and then drop all 10 a few turns in? Not fair in my mind.

but alas it seems there is not much interest to see different reasons why one leaves a game

1589 days, 14 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Considerate dropper case needs a bit of work from the site:

1) Player opens the position for replacement by a specific player
2) The other player then sees the game as optional under 'my games' and can choose to assign/reject the position
3) Only premium players can take over positions this way, so in case of cheating with throwaway accounts, the cheater at least needs to pay for it
4) In case people start spamming their games out in hopes of someone misclicking and accepting, the option to receive a game could be an opt-in switch in account settings.
5) the giving player receives a 'handed over' ending for the game and shares the final rank with the replacement
6) A clever rule gets devised to prevent last minute replacements from getting undeserved victories, in case it becomes a problem

1589 days, 14 hours, 26 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@lord+pollax:

Yes. The last time I took EE was just to get a game going too. Maybe EE needs help, but that's another subject. If all else fails you can always start a custom game that meets Joshua's terms for points, pick your race first, and then wait for someone else to take one for the team. So, you CAN pick your own race a lot more often than you can select any other game conditions.

I also agree about the failure of tenacity to cover all situations properly. Thus, something that only rewards finishes and scores players at the time the exit would seem to handle most such issues. I would certainly favor being able to pass your spot to another player without penalty if he just gets all the credit for the finish.

1589 days, 13 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'd be in favor of that kind of transfer Mule.

1589 days, 12 hours, 20 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

A ladder could also implement a race modifier, similar to the opponent skill modifier, where you'd get more benefit from doing well with a statistically weaker race than a stronger one. This might help convince people to join games with weaker races.

1589 days, 12 hours, 6 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@Garth: Good idea!

In my view the biggest advantage to a system based on * all * the finishing positions is that it gives more players a reason to stay and fight.

Right now, as some have indicated, it doesn't take that long to see if you have a decent chance to win or not. If the answer to that is no and you don't care about achievement points for campaigns, it is really hard to justify spending the next 6 months to a year in such a game.

If, on the other hand, you can gain points / recognition / rank by finishing 4th (for example) rather than dropping or just putting in HO-HUM turns, then we might all see the quality of play improve.

1589 days, 12 hours, 3 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Garth Vader, "A ladder could also implement a race modifier,"

I think Dotman's ladder DID have a version with a race modifier. I like any idea that encourages and rewards members who agree to play some of the tougher races.

1589 days, 8 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> A ladder could also implement a race modifier

Actually, if the ladder scores race like it does player, it could generate accurate racial modifiers for us. Of course, Classic and Standard would have to be separate.

1589 days, 8 hours, 29 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I was wondering if Dotman could remind us or demonstrate somehow the system he has already developed? There is no reason to re-invent the entire wheel if we already have something somewhat wheel shaped.

1589 days, 4 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Hrm. Actually, @Mule, I was thinking a simpler solution is within our grasp.

What if we simply reduced destruction and capture achievement points by, say, 75%, otherwise maintaining the present system? It's a very small fix, and it would seem to address the overwhelming majority of the issues raised herein. If we sort the "Mercenary" ladder further by actual achievement and experience scores within each rank average bracket, that would give us our single ladder system.

I know; it's not as satisfying as the sledgehammer approach, but it would solve our problems. I think even Ville would appreciate the results, if not necessarily the method.

The down side, of course, is that it's difficult to institute this retroactively, but going forward it should be easy enough.

1589 days, 4 hours, 2 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

In my opinion current system is flawed beyond repair, appreciating win way too much, achivements do not degenerate due to bad play/resign/drop.

If something, I like destruction + capture achievements in current system. Granted, they may be abused but thats programmers work. They are the only motivation try to do something once game has winner(s) sorted out. And they are too small.

Too much of our achievements are gained plainly from colonizing runs where players located near droppers just farm the planets and start countdown. And whats more shameful, most solo victories granting huge amount of achievements are done in these dropfests.

I would like to see ranking system which rewards active players who try to place as high as possible even if they don't win. Now the weight of a win is ridiculous.

1589 days, 3 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Dropfests, as I understand it, are common only in melee and beginner settings, thanks to the institution of the Tenacity metric. As such, I would advise that achievement points from both of these be regulated somewhat differently. Beginner games, in my opinion, should not catapult any player to admiralty; Melee games, on the other hand, are by their nature more reliant on destruction.

I hesitate to mention this, but your own view of the game seems to differ from that of the mainstream player, Ville. I'm not saying you're wrong, but rather I'd suggest that your ideal system (if it existed) would be unlikely to have widespread appeal.

Having mentioned it, though, it's possible that I'm wrong.

Perhaps we should wait for input from the design team, at which time we could poll the active membership.

1589 days, 3 hours, 45 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

I see you have not played many games at this site. Every game I play has been suffering from droppers, either mentally or plain resign/miss 3 turns.

What I see I think YOUR opinion differs from mainstream.

I think most players would WANT active games where ladder-style pecking order would be in place which would guarantee that players fight for EVERY position instead of 1 or 1+2.

1589 days, 3 hours, 36 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk In my opinion destruction score broke when Merlin build control became commonplace. A good player kills almost as much tonnage from own Merlins as from enemy ships during a game. This will be even more emphasized with the PQ.

The more I think of it the better a ladder system sounds. How can we get this started?

@Dotman @Big Beefer do your elo and ladder scripts still work with the current api? If yes could you please run them against the current data set and post the results?

1589 days, 3 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I disagree with your PQ prediction, @SMN, but in general, I maintain that simply reducing or removing destruction scores would be sufficient as a change. As well, given our missing game data, it may be problematic to institute any drastic fix except going-forward.

Do enough people agree with Ville's contention of the value of destruction scores that we should poll the players? What of the value of fourth place non-win finishes as compared to "dead" or last place or wins?

1589 days, 3 hours, 17 minutes ago

commander koski

RE: I am starting a campaign to lobby the gods of this site to...

I'm along the lines of @Smn , I think ship destruction achievement award is a poor thing:

A fun little detail: the present system punishes successful war of attrition and fighting against Crystalline.

I'm playing as the Fascists at Uri sector. I started early cloak attacks against the Rebels to force him to use his resources on minefields and not being able to build Rush.

I was successful - but I would have got a lot more achievements if I would have allied a carrier race, allowed Rebel to build many Rushes and then just banged our heads against each other.

Also I've fought many turns against the Crystalline, sweeping countless web mines. I think both parties show a lot of skill in our struggle. But our resources go to heavy phasers and web fields, the ships are rather small. Not many achievements there.

---
In my opinion, ladder would emphasis quality, not quantity of playing.

I played in a ladder system in the 90'es, it worked pretty well, it was NAVGAP, I think... Can't remember very well and can't find much in the internets.

1589 days, 3 hours, 17 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Its not mine, and I don't feel like reinventing the wheel is necessary.
https://groups.google.com/forum/#!topic/alt.test/a6xFQTQCKDA

"First of all, the base modifier is determined, according to the player's final rank.

SP Modifier by Finishing Rank
First 80
Second 40
Third 20
Fourth 10
Fifth 5
Sixth 0
Seventh -5
Eighth -10
Ninth -20
Tenth -40
Eleventh -80
"

1589 days, 3 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

By suggesting a poll, I don't mean to imply that you are the only person who owns that perspective, Ville; rather, I posit the opposite.

In that sequence, we seem to have removed any reference to a game win, which is my own primary objection. A winner in a team game, after all, may well not be the planet-count leader, and the second-place winner is somewhat unlikely to be #2 in count, in my experience. Since we have an established win condition, it seems rather arrogant to simply ignore it in any ranking method.

1589 days, 3 hours, 4 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Somehow I get the feeling that we're looking for excuses NOT to implement ladder (for some weird reason) rather than just implementing it side-by-side with current system and see which is better.

I mean, the work is mostly done thanks to Dotman.

And I think it is NOT a big deal to sort players by winning conditions. If winning condition is to form 'Gnerpkh' with Virgos on starmap then the player who does that is 1st when counting rank.

If you win a team game as a team then the player with most planets in winning team is 1st and so on. The rest sorts after the winning team by plaent count.

These are not really showstoppers.

1589 days, 2 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I think they're pretty important. Personally, I'd laugh at any system that ignores win conditions, and I do laugh pretty caustically from time to time.

1589 days, 2 hours, 43 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Naturally winner(s) come first in any point calculating. Whatever is the win condition. How the winners are sorted is usually planet count. But, in the old days we had more winning conditions like military score, tonnage sunk and so on but those are just a matter of scripting in to the calculator.

But I doubt that Dotman should do all the work, he may provide the basic setup from which developer(s) can continue.

1588 days, 23 hours, 42 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

+1 for side by side ladder and current achievement/rank system. And I vote the ladder be simply winners, followed by other finish positions.

Droppers are a real problem, but should be handled elsewhere by cutting down the number of simultaneous games for each drop, charging a monetary cost like $10 per occurrence to restore upgraded status, or some other serious penalty.

1588 days, 23 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

VK you are right and zeroed in on exactly what was bothering me for last couple of days. It seems that we are doing our best to argue into paralysis by analysis.

Let's don't let good be the enemy of perfect.

This request could be split into 2 parts.
1. First is enhancement of office profiles by adding badges for special feats (like medals). It could be Solo wins, 1st places by each race, Champion contender, Emperor, etc. Big Beefer system did that beautifully and I want that available for all.

2. Ranking system that takes all people of the site and ranks them based on the results. I am not going to tell the developers how to do it - I will be just happy if / when they do. The simpler, the better in my book (1 number however it is calculated)

1588 days, 19 hours, 18 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I have to agree with VK's - "Somehow I get the feeling that we're looking for excuses NOT to implement ladder "... And as Azz says... Paralysis through Analysis.

Anyone that goes into a game thinking their goal is to blow stuff up is not thinking strategically. When we say "win" we are saying finishing with the most planets (or mil score in some cases, I guess.) How would the rest of the players scoring NOT follow the same metric, ie: number of planets? And that points to their finishing position. If we don't want players to start dropping out or just putting in time as soon as they see they can't win, how would they not be judged by finishing position (planet count) relative to all other players - just as #1 is? To link a progressive finishing position to rank and other recognition just seems obvious to me. To do anything else just fails the logical smell test!

1588 days, 17 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> I vote the ladder be simply winners, followed by other finish positions.

This means that you don't see any difference between the 3rd-place player and the 10th place player. Is this really what you believe the scores should reflect? It will leave most of the players at/near a score of 0, which gives no indication of their skill. It will also give most players no incentive to keep playing. Somehow I don't think Joshua will be happy with a scoring system that causes new players to leave, even a secondary scoring system.

Please remember that the maximum number of players in a game is currently 35 (recently increased from 30), so scoring by rank will have to be able to handle that many players. I'd like to suggest that an equation like "Y=1000*x^2" (or maybe even "Y=1000*x^3") be used. The player positions would be evenly distributed from 0 to 1 (inclusive) on the X-axis, and their score would be the associated Y value. This equation should give a point spread similar to that which VK described above, but smoother. It would also automatically compensate for the number of players.

I believe that, if this idea is to succeed, each player will need a per-race score, and an optional (as in if someone wants to create it, I don't see that it does any harm) overall score. In addition, the last 20-50 games (pick something that makes sense) of each standardized type (Classic, Melee and Standard) should be scored by race, so that we could accumulate, and display to the public, valid statistical information on the relative strength of the races.

While I understand the desire to Keep It Simple, that philosophy is only valid once you have adequate complexity to do the job right. If this is implemented as "too simple", it won't give meaningful information, which will allow it to be easily ignored. If you really want Planets.nu to implement a ladder, you need to present them with something that works better in virtually all cases.

Please remember that Joshua has put a lot of effort into his current scoring system, and I doubt that he'll even consider bringing another scoring system into his site unless it offers features and quality that his doesn't, and never will. This means that the initial design has to be right. I expect that several different algorithms will need to be tried before the right one becomes apparent.

1588 days, 17 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Yeah; pretty much what @Whisperer said.

2 things:
(1) So long as we're gonna do it, let's do it more or less right. The logic's easy; I'll work it out in pseudocode for anyone that wants my help. The consensus seems remarkably consistent, so that's no worries.
(2) We presently HAVE a top 100 list. It's called "Mercenaries". I don't happen to like it, but it's there.

I mention (2) not to imply that we don't need some sort of change but rather to underline that we do.

Again: Personally, I feel we could get similar results by drastically reducing Destruction achievement, but I've made that point already and only a couple of people even seem to have noticed.

1588 days, 16 hours, 57 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP:

(2) Merc top 100... And this is based on relative positions at the end of the games? Or is it based on achievement or other metrics?

If it tracks relative results from #1 through #11 (or beyond) then I will stand corrected. Except that it should apply to ALL players and not just the top 100.

But, if it is not based on finishing results (as I believe) then I would suggest you haven't been listening. :)

Back to (1)... In the opinion of many of us "doing it more or less right" is tracking the relative position of all players in a given game and using it to determine rank.

1588 days, 16 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

No; that was my point too, @Mule; I just made it poorly. If you look closely at the top 100, you'll see what I mean. The system clearly needs revision if it's to be accurate.

1588 days, 16 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

One more thing WRT scoring. If you have an 11-player game with 6 finishers, 4 eliminations/FoFs and 3 resigners/droppers, each of those 13 players would need some score. While the resigners/droppers may have done something bad (they left the game), they did contribute to the game. They need to receive a score that's appropriate for what they gave the game (possibly pro-rata based on total planets for their slot vs. planets when the replacement player comes in?).

1588 days, 16 hours, 39 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Resigners, droppers and dead share the 11th position in my opinion. Places 12th and 13th until eternity share the penalties of 11th :D

1588 days, 16 hours, 29 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Resigners, droppers and dead share the 11th position in my opinion

So a player who's eliminated in turn 20 has the same skill as a player who's eliminated in turn 120? That doesn't seem right to me. I believe places should be allocated in the order that the players are eliminated/FoFed.

A player who's in second place, is dropped, and replaced by a player who ends up in 3rd or 4th hasn't shown any skill? Again, that doesn't seem right to me.

If the idea is that this "Ladder" score reflect the skill of the player, then it needs to consider the various things that can reflect skill. If it doesn't then it will probably fail.

1588 days, 16 hours, 28 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Well, correct that - dead go in order of elimination. So if 3 are alive at the end then 4th would be the latest dead.

Resigners and droppers share 11th.

1588 days, 16 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Resigners and droppers share 11th penalties

That's introducing Tenacity into a Skill-based score, and is just as bad as introducing Skill into a Tenacity-based score. It would be most honest if the score reflected the skill that the player had shown up to the point where they resigned or were dropped. Obviously, since early-turn planets are easier to get than late-turn planets, any pro-rating would have to be non-linear.

1588 days, 15 hours, 50 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Also, if a resigner or dropper isn't replaced, then they'd fall into place with the eliminated and FoFed.

1588 days, 15 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ville I don't believe you could be more wrong about droppers and melee games. A quick random look at recent games show most have had at least 30% droppage, some significantly more. Just look at any game where the victory occurs in <100 turns as a general rule, and you will find a drop fest.

The ability to transfer your position and points to your replacement serves to limit punishment (you get zero points), and rewards replacements, which is nice. Replacements should never get negative points either, in my opinion. They are doing a public service joining to keep a game functioning.

1588 days, 15 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

When I did my test run, I gave droppers the lowest rank not yet taken (ie, if 2 players are eliminated, 2 dropped, and then this player drops, they'd be given rank 7 (11-4)). I believe I followed the old ladder in that regard. I ignored ratings for replacement players, ie, they neither gained nor lost points. Similarly, the scores given to the players at the end of the game were based on the scores of the initial players, not replacements. This is probably a flaw, but also probably tends to work itself out in the long-term.

The old ladder this was based on had what they called reliability points, which I didn't implement at all, since we have tenacity for that (ostensibly); at any rate, the rating isn't meant to convey what we mean by tenacity at all, and conflating them is a mistake, as whisperer points out above.

1588 days, 15 hours, 38 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax - melees are drop fests. BUT, with 30 players when 20 have dropped we're in same position as one standard 11 player game.

1588 days, 15 hours, 4 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Melees aren't my thing, but I would think giving all the early >#11 droppers (or deads if that happens) an equivalent score of #11 is good enough. Or... just don't count Melees at all. Do they count now?

I think the system Dotman described is already good enough to satisfy me.

1588 days, 14 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Or... just don't count Melees at all. Do they count now?

Yes.

> I think the system Dotman described is already good enough to satisfy me.

That may be, but satisfying you isn't relevant to getting Joshua to put it up on the site. For that, it would have to satisfy him. I keep seeing different versions of "it's good enough for me", and requests that Joshua support it. This is the wrong combination. For Joshua to support it, it will have to be good enough for him. I believe that means it will have to do everything the current scoring system does, but better.

Simple is good, as long as the job gets done.

1588 days, 14 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@ville But nothing determines a victor in a melle game more than the location of the droppers. A guy who has all his neighbors drop is essentially assured a victory. No effort, and nothing the remaining folks can do. Especially if you are unfortunate enough to be in an area where no one drops. No skill, no tactics... just pure random luck of the droppers decides these games. Total garbage. These melee game wins should not even be counted, as they are meaningless and in no way reflect skill (in the vast majority of the games, not 100%, but most).

1588 days, 14 hours, 26 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> A guy who has all his neighbors drop is essentially assured a victory.
> These melee game wins should not even be counted

The exact same result will happen for a Classic or Standard game. Does this mean that Classic and Standard shouldn't be counted?

1588 days, 14 hours, 21 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer: Yes, about Joshua's opinion being critical to getting NU to take action. However, I believe he listens and takes into account our opinions as well. Unless we state them, he cannot do so. So... you, me, and all the rest need to communicate. As I am doing. Not for a moment do I think my opinions are any more important than anyone else's.

Nor wound we need Joshua to agree to an off-site list using a separate system. Perhaps not as desirable, but certainly do-able.

I am in agreement with lord pollax on the melee question. With the massive drop ratio we often see, assuming you have even average skills, success depends as much on luck of position as anything.

I also think once we stop paying attention to how much you blow up or how many planets 2 players can trade back and forth and focus on results at the end of the game before anything is awarded, we will have far fewer drops to worry about. We could even do away with the whole tenacity system entirely. It won't be needed.

1588 days, 14 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Quite a debate here. One comment from Gnerphk was: "I hesitate to mention this, but your own view of the game seems to differ from that of the mainstream player, Ville" So...I did a little math.

On the PRO ladder side I see posts from about 12 different players who have started and finished 145 games. On the CON ladder side I see two players who have started and finished 21 games. I bring this up because some of the posts against trying this out make me wonder if the CON side might be arguing more from theory than experience.

The positives to a ladder over the current system are not minor. It adds incentive to finish higher and improve one's final number. The current system rewards the winners with a new car while giving the other 9 or 10 players a set of steak knives. The ladder can allow a new player a chance to see his name move up the board even if he can only play a few games at a time. The current system allows a player to 'game' the system by focusing on one race, joining 8 games, quitting (or mailing in) any sector where he doesn't like his chances, while focusing on the few where he might win. The ladder makes us invest in EACH game.

Also, can we agree no one is suggesting we replace the existing system? As someone posted earlier, why not have both? Let each member focus on whichever system they prefer. Dotman had a basic model and a second one that adjusted for race. Once in place adjustments could be made to improve it where needed. It's like we're debating leather or cloth upholstery before the car's been built.

Melees should be excluded, IMO. They are a nice sandbox for new and old players alike.

We've been talking about this subject on and off for over a year so I suggest if the site isn't sold on the idea yet, maybe a 3rd party link could keep the ball rolling.

1588 days, 13 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax

I almost take offense at melee games involve no skill and only your neighbors matter. I usually don't flaunt my record but in this case I will make an exception - quite a few of those games are melee games. I play them because I enjoy fast action rather than building up for 30 turns first. I also enjoy the diplomatic opportunities with 30 people involved - you can put together quite interesting ploys with that many people.

I must be one lucky SoB when it comes to neighbors.

1588 days, 13 hours, 22 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Azz, I mean no offense to anyone that plays Melees, but I just looked at 10+ Melees and not one has more than 10 of the 30 spots still playing. Most are more like 7 or 8, which is about 25% filled.

You must admit that is a very different dynamic as far as luck of position compared to most 11 player games. It doesn't mean the players are not just as good. Just that the measurement might be different. It's statically easier to get good neighbors when 75% of them disappear. :)

1588 days, 13 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

Part of the reason 75% of them disappear is that they are disappeared. Giant melee do have earlier, faster conflict. In the game I am in there are a lot of dead races, but only two of them dropped that were of any consequence. All the other dead ones were a lot of early, vicious fighting.

I enjoy classic games more than melee (I'm just here for the Godzilla folks) but I'm not sure it is accurate to assume that 75% of your neighbours disappear. In fact, droppers happen quite often in 11 player games as well, and the results are even more disastrous, because larger kingdoms come up for grabs. Someone with 11 planets dropping is not going to address the balance of power in a giant melee when there are 4 people potentially vying for those planets.

1588 days, 13 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

Part of the reason 75% of them disappear is that they are disappeared. Giant melee do have earlier, faster conflict. In the game I am in there are a lot of dead races, but only two of them dropped that were of any consequence. All the other dead ones were a lot of early, vicious fighting.

I enjoy classic games more than melee (I'm just here for the Godzilla folks) but I'm not sure it is accurate to assume that 75% of your neighbours disappear. In fact, droppers happen quite often in 11 player games as well, and the results are even more disastrous, because larger kingdoms come up for grabs. Someone with 11 planets dropping is not going to address the balance of power in a giant melee when there are 4 people potentially vying for those planets.

1588 days, 13 hours, 0 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Furey, If nothing else, all those early drop provide more uncontested territory.

But, I don't want to sidetrack this thread from the original intent of supporting a position based ladder. Whether Melees are counted the same or not, I'd still like to encourage a ladder.

1588 days, 12 hours, 45 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Tom N

I have no opinion one way or the other, but comparing post comments is a statistically invalid method. In these situations, those desirous of change will almost always greatly outnumber those that do not. The great majority of people who like the system the way it is have no incentive to post, so are disproportionately underrepresented. Those who may not like the system, but do not dislike it enough to care will not post. Only those that want to see change have any particular reason to post. As a result, one would expect the numbers posting they prefer the ladder system over the current system should greatly outnumber posts voting to keep the system the way it is. If the postings were equal, that would actually indicate those wanting the ladder system were greatly in the minority.

1588 days, 11 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> The great majority of people who like the system the way it is have no incentive to post

Similarly, when someone reads a post, and it reflects what they would say, they usually don't bother to post, and they usually don't even "Like" the post they agree with. Post counts, either for or against an idea, really have no meaning. What does is the actual discussion, stripped of the emotion and meaningless wrapping.

> the majority of people would agree to a third party set up that was not a Planets NU system, leaving the current system intact so that they run side by side

Not only would I agree to that, but I believe it would be the best way to work the kinks out of the system, and see what algorithms work best.

I've seen many comments about "X is useless", "Y won't work" and "Z has to be done" (and made a few myself), but the only way to actually find out is to try it.

1588 days, 11 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Slash, Rightly stated, and noted. I only mentioned the numbers in response to what I felt was a false assumption that most members would not want to see a ladder system side by side to the current one, because Ville is somehow in the minority opinion. But however anecdotal these postings are in relation to what percentage of members would like to see the two systems displayed, you have to admit now that it appears Whisperer is on board the 'thread vote' is now 13 to 1 in favor. I would urge more members to take 10 seconds and type "+1 Ladder Trial" if they'd like to see it happen.

On the Melee question I would love to see these either excluded or on a separate ladder. As an example one unlucky starting position can cause an early death and then what? A "35" for finishing last? I just think it would skew the numbers and compares apples to oranges (11 player games to 35). Even if you hand out 25 "11's" to all players who don't finish top 10 it would screw up the ladder, IMO.

1588 days, 11 hours, 6 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Whisperer is on board

I've always been onboard with seeing the results, and determining whether or not they're better than what we have. I'm not onboard with requesting Joshua to spend time implementing it until I see how well it works. My intent with most of my comments above was to try to help the effort yield the best possible results.

> On the Melee question I would love to see these either excluded or on a separate ladder

Melee games are highly biased towards races that can develop quickly, either because of cheap ships (Lizards and Privateers) or lucky placement. Classic and Standard games, somewhat less so.

If you exclude Melee games, what do you do with 25-player games? 20-player games? 15-player games? To me, it would make more sense to make the number of player slots part of the equation for calculation of the player scores for the game. Once the equation is "right", everything "just works".

1588 days, 10 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

One bad start in melee - so what? It balances out in the end - rankings are not based on 1 game.

In one of my melee games I had a horrible start - I was at war with very strong players since turn 5. One after another - after killing one I was faced a fleet of 3+ Gorbies, firecloud and a bunch of other nasty stuff that was impossibele for me to deal with directly few turns away from my hw with 3 T-Rex to my name. Instead of dying or having poor finish, I reached out and organized multiple races to oppose the leader at that time. Long story short, I won't win that game but I plan on finishing 4th and it is looking likely. Not good but not bad considering I was few turns away from death.

Moral of the story is that even when you have a bad start and luck is against you, usually there is something you could do to improve your position. All else fails there is a next game

1588 days, 9 hours, 39 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

I too would like to see how the system works, so count me as a vote in favor, although I agree with Whisperer that it is neither Joshua's job to do the work of designing it and the main site is not the appropriate place to experiment and tweak it until people agree it is working. I don't have enough experience in either of the methods to have an opinion yet, so seeing them side by side sounds interesting.

What I will say is that I have had numerous instances in my life of coming in second place in a variety of circumstances, so a system that counts second place as equal to last place does not appeal to me.

On the discussion of melee games, would not a good compromise be to only count those players who finish in the top 11? Players who rank below that get nothing. This would place the rankings in melee games on a more equal footing with the standard games. It would not address the bias toward the rapid expanse races that has been mentioned, but I think it would go far towards addressing the other issues that people have brought up.

1588 days, 8 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Side note on a personal level:

Who is it you're calling out, @Tom+N? I was ready to take offense, but it doesn't seem to be me. Now, I'm just a bit confused.

There seems to be a fair consensus, and the only things we're splitting on are Melee games. I'm sure there's an approximation we can use for them -- a nonlinear multiplier comes to mind -- so all that's left to do seems to be to have someone code this. I suspect we'll have some trouble coming up with winners, but there's a way to manage that using the reported Advantage scores.

What about nontraditional games? Military score et cetera?

1588 days, 8 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> On the discussion of melee games, would not a good compromise be to only count those players who finish in the top 11?

While that might work, I think we can do better. The issue here is the exact same one as you made WRT 2nd place. It doesn't matter which position you end up in. If you're not dead last, then you've probably demonstrated some level of skill, minimal though it may be, that should count towards their score.

1588 days, 7 hours, 40 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@GP says: >> but there's a way to manage that using the reported Advantage scores. >>

I see 2 big advantages to a position based ladder.

(1) It would address all positions and give incentive to stay in the game and improve your position, and not just the top 2 or 3.

(2) It ignores meaningless trading of planets and blowing stuff up that does not increase one's eventual finish position. Now you want to put this back in?

Slash makes a fine suggestion (and IMO certainly better than using advantage points) when he says >> a good compromise be to only count those players who finish in the top 11.

1588 days, 7 hours, 19 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Another possible Melee vs Standard solution.

Every game calculated by 100 points divided by # of players at the start of the game. Also award a bonus to the top 3.

For example:
An 11 player game: 100 / 11 = 9.1 (rounded)

#1 gets 9.1 * 10 (the players he defeated) = 91 + 100 bonus = 191.
#2 gets 9.1 * 9 = 82 + 50 = 132
#3 gets 9.1 * 8 = 73 + 25 = 98
#4 gets 9.1 *7 = 64
and down to
#10 gets 9
#11 gets 0 because he didn't defeat anyone.

Apply the same formula to a 35 player game:
100 / 35 = 2.8 with same bonus structure.
#1 gets 128
#2 gets 50
#3 ... (and so on)

1588 days, 7 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Question:

Player A finishes game with 1 ship.

Players B, C, D, and E drop out and are replaced.

For a ladder, what position does Player A occupy?

- 11th, because he has least number of planets at end of game.

- 7th, because of the starting players, he did better than 4 other players. Only 6 of the starting players have more planets than Player A. Original starting players B, C, D, and E should occupy positions 8, 9, 10, and 11.

I vote 7th.

In any rating system, any positions not occupied by the starting player, should be considered lower than any starting player. Only starting players should be rated for that game. Replacement players should not be rated for the ladder, just receive Achievement and Campaign Resources.

If you are not penalized for dropping out, players could just drop out to avoid a bad ratings loss.

1588 days, 6 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule:

Perhaps I should have said, "When scanning the site for game position with regard to the number of planets someone had at the end, we could actually look at the awarded Advantage points to see if they were multiplied by anything. That could actually inform a programmer as to in which place a person came as opposed to how many planets he owned."

I just didn't think I needed to spell it out quite that much.

1588 days, 6 hours, 38 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

>While that might work, I think we can do better. The issue here is the exact same one as you made WRT 2nd place.

While I might quibble with the use of "exact" in your sentence (yes, I am sometimes too pedantic for my own good), I agree with your point. My suggestion was just throwing out a possibility. No doubt a better solution can be found.

1588 days, 6 hours, 20 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP: >> Perhaps I should have said >> Maybe I should have seen that. :)

MJS does bring up an interesting question on replacements...
I was thinking drops should be scored by their position a the time of the drop, but that doesn't answer his replacement question.

(1) Dotman / BB - could your programs extract the data as MJS describes?

(2) How much do we want to discourage replacements? I recently saw a game with the #2 spot open for a race I would have liked to play. I didn't join because I know replacements are now judged only on their ability to maintain position. So, I passed on it. The more reward we offer replacements the more likely we are to find them. I don't know that I would want to be a replacement under MJS's plan either.

1588 days, 6 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer said > it would make more sense to make the number of player slots part of the equation for calculation of the player scores for the game. Once the equation is "right", everything "just works".
I like this solution as a way to include Melee games in ladder system.

@Gnerphk, My intent was not to offend, but to help move the discussion from a debate to an agreement. Is it fair to say your position has changed from 3 days ago? In any event I'm glad to see this thread progress from 'should there be a ladder system?' to 'what would a fair ladder system look like?'.

1588 days, 5 hours, 45 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Mule makes a reasonable suggestion for melee games I think.

@Mjs68508
I understand your suggestion regarding the original players. I also understand not wanting to penalize a replacement player for taking on a losing position and not being able to improve it, their fortunes are less in their control when picking up a race started by someone else. But it seems to me that it penalizes a replacement player that does well. Should not a replacement player get a bonus to their ranking if they improve the fortunes of their race?

1588 days, 5 hours, 41 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

MJS said > If you are not penalized for dropping out, players could just drop out to avoid a bad ratings loss.

I also like VK's idea of minus points for players placing in the bottom half of the game standings. Thus, the 1st player to drop or the #11 player in a standard game gets the max minus points - however much that is.

That should discourage those that might think of a drop to avoid things getting worse. If they hang in someone else might drop first. :)

1588 days, 5 hours, 37 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's what I'm after: a metric to encourage Sten Avon. If the worst of us -- no, I mean the antiheroes among us -- are given incentive to behave in a way which benefits us all, we'll all be better off. We'll have a more reliable game and they'll have their incentive.

A positive incentive should also exist, of course, but that's another part of this discussion (and we've already agreed on it anyway).

1588 days, 4 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Ratings systems are ratings neutral. That is, points gained by some players equal points lost by other players. If there is a +bonus for first, there should be an equal -bonus for 11th.

Ratings gains by starters and replacements should exactly equal ratings losses by starters and drop-outs. If this doesn't occur, then you don't have a ratings system. You have an accumulation system like we have now, where you can game the system by working for rewards with less at risk than a starter.

No replacement player should get more points than a starter. The replacement player wasn't there the first turns when diplomacy was crucial. Joining when you already have a firm alliance shows no skill in VGA Planets.

Also, one cannot compare 3rd place for the person who begins on turn 1 and for the person who begins on turn 50. They are different talents.

1588 days, 4 hours, 7 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Don't get stuck on details. Thats why I love agile development processes since after (great) idea you'll start working on iterations, perfecting the product with every release.

What comes to melees, droppers on different turns, replacements and so on - I would not think it too much. Work on the basic ladder Dotman has provided, once it ranks standard game results then start with melees.

Once all games are rolling in then work with droppers, resigners, special cases etc.

What I'd like to see is an official release 0.1 (after the year-old NUELO) and then work/comment on it.

1588 days, 3 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm for getting it close to right the first time, not to put too fine a point on it. It's not much extra work, after all; it's just logic.

1588 days, 3 hours, 53 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

With a live system showing the actual changes instantly finetuning is also much more easier.

1588 days, 3 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

...though I suppose, when dealing with posts like that last one of MJ's, there's some wisdom to your perspective, Ville.

Controversial limits are probably counterproductive at that.

1588 days, 2 hours, 44 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I would simply disregard any replacement player in the game and reward replacements another way. A system that doesn't rank a player's ability to play start to finish is not a good measure and open to gaming the ladder. I'd rank dropping as a death at the same place in the game regardless of the current planet count.

I also can't express too forcefully that I feel we dramatically under penalize and under ostracize dropping from games. It is so impactful to a game when players drop as everyone acknowledges, but then we set up a system that doesn't discourage it. I have a very busy job and have never dropped from a game in 25 years. I realize that is partially luck on my part not to have been caught by a true long-term emergency, but an even greater part is the times I have overextended myself with games, I have been able to gut them out. We allow people to play far too many games at once as a starting practice, IMO. The right to play more than three games at once should have to be earned and dropping from a game should result in a significant hit else we teach no game etiquette.

I'd also have a second ladder for melee games since their strategy and scope is entirely different from 11 player games.

1587 days, 19 hours, 52 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Drops: Kelman says >> we dramatically under penalize and under ostracize >> and I agree with him. But, he also says >>. It is so impactful to a game when players drop >> which is only sometimes true. If a player drops early and leaves more space for his immediate neighbors or even worse if he develops an empire and then drops, it can be a big impact. But, if he plays through most of the game and is beat down to only a few planets remaining, he can drop without any effect at all beyond removing his skill set from the game. I would like to see a system the encourages players to stay playing as long as possible but does not over-penalize in comparison to the actual impact.

Personally, I have no problem with races dying. But, that is not how the site works nor how many players want things to happen. OK then... if we actually want replacement players, what metrics would make YOU take the challenge? I would not be a replacement for anything other than the same as if I had always been there. Otherwise, why not just join a new game?

I also wonder about the disconnect between us wanting to keep the balance as far as dropped players but ignore or even support such things as stellar cartography - which in my mind is like putting rubber bumpers in the middle of a perfectly good billiard table. Or Ion storms hitting some players and not others. Or random resources, planet distribution, difficult neighbors and so on... I think we worry too much about droppers. Over time we will all be sitting next to one or two, so I would rather not screw up the whole scoring ladder because of them. If we start scoring by position and stop offering rewards (such as accomplishments) to players that drop, we should see a lot fewer of them, because their ladder position will also keep dropping.

1587 days, 19 hours, 42 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

To support Gnerphk's comment, the biggest mistake most people make in projects is to cut corners in the planning stages. Every hour of time spent in productive planning and consideration of all the possibilities saves exponential amounts of time in the production phase. Admittedly, there comes a time that one has to simply get started. Planning can be taken to extremes, but that is far less common than not enough planning to begin with.

>Ratings systems are ratings neutral. That is, points gained by some players equal points lost by other players. If there is a +bonus for first, there should be an equal -bonus for 11th.

Not true. There is no requirement for one player to get penalized simply because another gets a bonus. That is usually in fact a counterproductive approach, as it generally results in people quitting and dropping out, which is what so many people have argued is a big problem. It also not uncommonly creates a fairly hostile environment where backstabbing becomes commonplace. I for one have no interest in playing in such a game.

>Ratings gains by starters and replacements should exactly equal ratings losses by starters and drop-outs. If this doesn't occur, then you don't have a ratings system. You have an accumulation system like we have now, where you can game the system by working for rewards with less at risk than a starter.

Again, not true. Ratings systems and point accumulation systems are not mutually exclusive. There are in fact numerous ratings systems that are designed using point accumulations. Any ratings system can be gamed. It is simply a matter of how one goes about it and the ease with which the particular system allows it.

>No replacement player should get more points than a starter. The replacement player wasn't there the first turns when diplomacy was crucial. Joining when you already have a firm alliance shows no skill in VGA Planets.

This is an opinion that I do not happen to share. As I stated above, I find it unfair to reward a player who drops in the first few turns more than the person who steps in and plays to the end. I think that attitude is shared by many here. Several people on these boards have stated that replacement players should be encouraged. I agree that players should be discouraged from dropping in the first place, but that should not mean that replacement players are any less important. Your example is only one of many possibilities and ignores the many replacement players who step in to a losing situation in which there were no alliances set up and the replacement player is the one to develop those alliances and turn around the fortunes of his race. That shows far more skill than the original player who bailed on the game. I think your statements about replacement players show a tremendous disregard for their efforts that I think is unjustified. In other games, I have seen replacement players who earned my respect far more than any of the original players.

>Also, one cannot compare 3rd place for the person who begins on turn 1 and for the person who begins on turn 50. They are different talents.

Well, you can, but I agree that it would not be at all fair. No ratings system is perfect. It is a different talent to play all 11 races than someone who plays the same race over and over again. It is a different talent to play the Privateers than it is to play the Evil Empire. I could go on, but the main point is that any ratings system you put in place will be comparing an amalgamation of different talents and some will not be treated as equally and fairly as others. The question is not which ones are different, but how far one is willing to attempt to separate them out. You can attempt to create an incredibly excessive system with different rankings for every skill, talent, and possibility, which will still fail to perfectly reflect the true situation; you can accept that a simple ranking will have some things that are not at all fair or accurate, or you can try for something in between. The question is how complicated are you willing to go to compare more closely related abilities?

The discussions that have gone on should have made it perfectly clear to everyone that there is no right answer. The best that can be hoped for is an answer that is least offensive to the greatest number of players. As such, rather than making absolute statements about what a system has to be, it would be more productive to clearly delineate a list of priorities that the majority can agree upon and work towards a system that maximizes those priorities. Otherwise, the discussion will continue unabated interminably. The last thing a developer needs is a bunch of conflicting statements from many people about what a system should be able to do.

Please excuse the verbiage, I get paid to lecture. I have PASD: Post Academic Stress Disorder :)

1587 days, 19 hours, 12 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Slash is spot on! Huzzah! Especially about replacements. I had a game lately where the replacement was head and shoulders above the original player and saved the spirit of the game for several of us. Although if all you looked at was his score, you would not know it. (Thanks, BT.)

Let's stop trying so hard for absolute balance in an imperfect world and accept that many things will balance out over time. We need to think about what action we want to encourage and design for that, even if it leaves a few extra points somewhere.

A ladder system that does not reward droppers with accomplishments and scores us all when we leave the game (finished or not) will do that much better than our current system. Let's get back to working on that.

1587 days, 19 hours, 11 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

The problem with discussions such as these is that so many things get brought into it that specific goals and issues tend to get lost in the mix and discrete conclusions become hard to reach.

It is good to periodically step back and say what are the overall things that need to be kept in mind to go forward. Here are a few things that I think fall into that category.

1. Not everyone plays for the same goals and any system needs to be aware of that fact.

2. Accumulation systems like the ones currently implemented measure experience (and imperfectly at that), but they do not necessarily measure ability other than by proxy via experience, which is a wildly imperfect measure of ability.

3. Accumulation systems have the advantage that it does not matter how many people are playing. Someone with a 1000 points can be roughly said to be comparable to every other player with a similar score, be there only ten players or a million.

4. Ladder systems theoretically rate ability better than accumulation systems in that it weights more the results a player has over how many games they have played. One could theoretically have a player ranking highly with consistently good results much higher than a long time player who scores poorly or drops frequently.

5. The serious disadvantage of ladder systems is that they are highly mutable to the number of players. Being top rung in a ten rung ladder means much less than top rung in a 1000 rung ladder. Similarly, being tenth in those ladders is extremely different.

6. Ladder systems assume that relative rank against all the other players matters to all the players. It is no incentive to be low in the rank if someone does not care about it. Some people prefer to measure themselves against their own performance and could really not give a damn about their ranking with other people. For those that are concerned with relative rank, ladders provide a high degree of incentive.

7. Accumulation systems provide goals that may mean more to the person who is less concerned about their relative rank than to what they have personally accomplished while at the same time providing some measure of relative rank to those that are concerned about such things, although less well than a ladder system provides.

As can be seen, neither system will work for all players as they place value on different things. Both systems have their place, but neither can be regarded as inherently better than the other unless one is talking about specific goals. It would be a mistake to try to shoehorn goals meant for one system into the other.

1587 days, 19 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well said !

1587 days, 19 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Tom N

Tom you have your answer

1587 days, 19 hours, 0 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Slash - if we are going to keep championship-game as the "ultimate goal" then ladder will definitely measure the participants a lot better than achievements.

1587 days, 18 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I think your trying too hard to simplify a complex system. In alliances games one player can provide mega support, yet not really be accredited for their efforts, only the Golden Race that wins, gets that. So perhaps there should be different categories for points. Each players stats are to reflect these different areas: IE, Support of allies, dependability in alliances, etc. I think such a break down will define any player accurately in the game under how they perform in all circumstances, isn't that what your trying to ascertain ?

1587 days, 18 hours, 39 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> If there is a +bonus for first, there should be an equal -bonus for 11th.

That's not needed. All that's needed to maintain a zero-sum score is that the game scores all add up to 0. For the computations, it's probably easiest to work with positive numbers, then, once the relative scores are known, bias them all so that the sum is 0.

> one cannot compare 3rd place for the person who begins on turn 1 and for the person who begins on turn 50. They are different talents

Except that this is what we have to do to make the system work properly. It won't be perfect, but I believe we can create something that will be functional.

> With a live system showing the actual changes instantly finetuning is also much more easier

Except that it can't.

The first step is to scrape the Planets.nu game DB, and put together a "database" (I'd suggest a bunch of text files) that contains all the necessary per-game information (win conditions, game type, win turn, win alliance and number of player slots) and per-slot information (slot #, player ID, start turn, end turn, start score and end score).

The second step is to load the above raw data into the score DB. Most of the tweaks would be in the algorithms that perform this step. If the algorithms are changed, this step would have to be redone. With the number of games in the Planets.nu DB, this could take several minutes. This is NOT "instant".

The third step would be to display the scores of the players in a usable and understandable manner.

> I'd also have a second ladder for melee games since their strategy and scope is entirely different from 11 player games.

What about 12-player games (common player count for team games)? 15-player games? 20-player games? 25-player games? Where do you draw the line? It's better to just handle them. This isn't all that difficult.

> I'd rank dropping as a death at the same place in the game regardless of the current planet count.

If there's no replacement player, I'd agree.

> I also can't express too forcefully that I feel we dramatically under penalize and under ostracize dropping from games. It is so impactful to a game when players drop as everyone acknowledges, but then we set up a system that doesn't discourage it.

While I agree with your statement (quoted), I VEHEMENTLY disagree with your "solution" (not quoted).

I believe the BEST solution is to improve the system that's supposed to penalize them, combined with the ability to have player-created public games limit the Tenacity of players entering. Recent/ongoing changes to the Tenacity system have broadened the impact of resignations/drops to include Melee games, and made the penalty better reflect the game impact. For the resignation/drop problem, we should continue to request that Joshua improve the Tenacity system.

I recently posted a list of several issues with the Tenacity system in the Tenacity Changes thread that Big Beefer created (http://test.planets.nu/#/activity/1877516). Please feel free to add your comments to that thread. If enough people make constructive criticism, the system might improve. I believe that the currently advertised changes make it a bit worse, but they do fix some huge issues.

This "Ladder" score is supposed to be a better measure of SKILL than the current system. How can we say this if we intentionally pollute it with Tenacity information? The act of resigning or being dropped, in and of itself, is not an indication of skill, or lack thereof, and should not affect the skill score. Just as it's inappropriate to have the "Rescue Bonus" be part of the Tenacity score, it's inappropriate to have a Resignation/Drop penalty be part of this skill score.

While we might WANT to create penalties for resigners and droppers because of the impact it has on the game, this is NOT the correct place to put those penalties.

-----------------------

Algorithm specifications:

I believe that this is what we have so far.

1. This is to be a zero-sum score. At any time, all of the scores should add together to 0. -- This is simple; just throw in a score bias at the end of the calculations, before the score is added to the user score. The bias would be the negative of the average of the player scores.

2. In order to maintain the zero-sum score, all players are to start with a Provisional score of 0. -- This is implied.

3. This is to score all players in a game, providing increasing scores for better placement. The rate of increase is to be greater for higher placed finishers. This indicates that there's probably a greater skill difference between the 1st and 2nd place players and the last and nest-to-last place players. -- I've already provided a basic algorithm that will do this. We just need to figure out the constants.

4. The scores are to be saved in a 2D array (player vs. race) plus 2 1D arrays (player and race). -- Saving the scores is the easy part here. There's no valid reason to NOT save the scores in the most usable manner. In addition to the actual score, we'd also have to track the number of games that have gone into each score, primarily so that we know when a score moved from Provisional to Valid.

Droppers/resigners/replacements can be handled later. For the time being, we'll gather and store their data from the Planets.nu game DB, but the score goes to the replacement (yes, I know that's wrong, but this is an interim state).

-----------------------

Outstanding questions:

A. When does a score move from Provisional to Valid?

B. Every game is a mix of player levels. If there's one high-level player and a swarm of low-level players, what, if any, score multipliers do the various players get for game difficulty? Should they all be the same, or calculated per-player?

C. How do we deliver reasonable scores to resigners, droppers and replacements that reflect the level of skill they've shown? As I stated above, we should not penalize a Skill score for an event that's not directly related to skill.

D+ Probably several others that we don't know enough to ask yet.

1587 days, 18 hours, 27 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I don't disagree at all that replacements may well be more valuable than the original player. I also agree that there are a few types of situations where dropping doesn't hurt the game that much, though the investment needed to stay in is also pretty low in those situations. I'm just saying the potential negative effects on a ladder outweigh the potential benefits, IMO.

As many have said, no one measure will be all things to all people. Give all the "achievements", or gold stars, or special ship skins, or opportunities to play in special games with ships that shoot from different distances, or whatever people enjoy. In a ladder, I want to see where I rank against other people who played a similar game under similar circumstances. Simple as that. For that you should gain rank for finishing well and lose rank for finishing poorly.

1587 days, 18 hours, 19 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@ Whisperer >While I agree with your statement (quoted), I VEHEMENTLY disagree with your "solution" (not quoted).

You didn't quote my solution because I did nor proffer one. I simply stated that I think we under penalize dropping.

I am in favor of a simple ladder. Limiting it to the original players is simple. The problem of droppers and rewarding replacements are both good concerns that I propose should be dealt with outside of a ladder.

1587 days, 18 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I think you cannot have one ladder system.

There are too may games set up variables.

There are standard games, team games and melee games, just as the basics.

Also how do you grade a Admiral entering a game that can have total newbies in (it happens especially in the Melee games)?

You would have to start from scratch and have the game set up and indicate it is a game that has valid entry into one of the ladder systems. It should have limited officer level entry and maximum levels.

So you would have a ladder system for 11 player standard games / campaign, then team of 2 standard / campaign, team of 3 standard / campaign and then Melee standard / campaign. (But then the standard and campaign could be joined together.)

But then you should grade the position based on the rank of the game. Winning a beginning level game should count less than winning a senior level game due to the "potential" quality of the players. The increased rank due to your highest level race should be used to limit things.

So no entry level and no upper level limit games should count towards the ladder system.

So no back dating the ladder system and start from scratch with every game being ranked as a set ladder game. Site games will be adjusted to always hit the limits of the ladder system. Custom games will have to be set up in a specific way to qualify for the ladder game.

So there will be a mix of regulated ladder games or non regulated private games which do not count to the ladder system.

After that is standardised you can start thinking the ranking system.

But this is only my opinion.

1587 days, 17 hours, 58 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Admiral entering newbie game means he won't have much to gain if he happens to win - but a lot to lose.

Imho there is no need for spesific rank requirements because after one or two rounds people learn that it may not be the best idea to join newbie meless to gain +5 ladder points while games against your own bracet gives a lot more.

At the moment problem is obvious. I'm playing multiple melees and by pure luck near win in two - I'll be gaining achievements nicely. Both games have difficulty factor of 0.82 so its not a bad days work. With ladder I doubt I would have joined these games.

1587 days, 17 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I went back to the ladder thread from 10 months ago. Dotman originated the post and likely had more time to invest back then. Many comments there address what we've been discussing these last few days.

Regarding modifiers for difficulty of competition, Dotman responded:
"Yeah, winning a game against high ranked players will bump your score significantly, somewhere in the range of 80 points or so, more if you were lower rated than their average score at the time. Of course, you don't see that in these preliminary numbers, since this isn't automated yet; so games that completed, say, since last Thursday or so aren't represented here (for now)."

Regarding a ladder for Melee:
Dotman said: "@Azzazzello: Yeah, definitely for campaign. Melee games have a few technical problems towards automating and would take a little longer (the rolling start/who is a replacement/who is not is tricky)."

Regarding Race modifiers:
Dotman said: "My personal opinion, I'm on the fence on the race modifier. I agree that just because it wasn't done on the RCCL doesn't mean its not a good idea. Clearly I think certain races have an advantage in classical games, in that they are easier to win with. A difficulty modifier based on this I think makes sense, as well as encourages playing the harder races, something I think we should shoot for. If done, I think it should only apply to classic games, not campaign game"

And I added (pertaining to Classic Games):
"On the "Race Modifier" idea, I was looking here: http://bit.ly/1dqWI2K at the Race Stats, along with some (albeit in no way expert) personal experience playing all races, and came up with this idea. Consider just a suggestion, a jumping off point, and feel free to hammer away.

Multiply the points gained by:

Privateers - *.75
Lizards - *.80
Borg - *.80
Colonies - *.85
Robots - *.90
Rebels - *.90
Crystal - *1.10
Fed - *1.15
Birds - *1.20
Fascist - *1.25
Empire - *1.30

The gains for the bottom 6 races = 100 points and equal the percentage points taken away from the top 6 races. I tried to take into consideration not just the 1st place finishes, but also the number of top 3 finishes.

Here's a link to that thread:
http://planets.nu/#/activity/1427142

1587 days, 17 hours, 8 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Ville

I don't disagree with you about the Championship games, you're probably right. But let's be honest, how many of us will ever play in them? Do we really want a system that only truly has meaning for a small percentage or one that has some meaning for the majority of players? I for one never expect to play in a Championship game. I will do my best in the games that I play in, but I really don't have the desire to ever be Emperor, nor do I particularly care about my relative rankings with other players outside the game I am in. I play for the fun and intellectual challenge in the game. I don't have anything to prove to anyone else. Do I like to win? Yes. Do I like to have the respect of those I play with? Yes. But getting to the top spot and being acknowledged as one of the best players honestly means nothing to me. I expect I am not alone in that regard. For others, finding the best players and challenging themselves against the top adversaries is an important aspect of their enjoyment in the game. That is perfectly fine and more power to them. That's just not me.

For me personally, I prefer the accumulation system as it meets my goals better than a ladder system. But I realize that my goals are not at all the same as many players here and that for many, a ladder system meets their goals more effectively.

I for one would feel it a terrible shame if we could not find a way to satisfy both types of players.

1587 days, 16 hours, 46 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding championships... I do not respect them or their results as long as the entry condition are based on achievements and accumulation. This does NOT mean I do not respect the PLAYERS that play in them. Just that I find the system itself a failure to measure actual game skills.

Any given Emperor might be the best player around or he might be the guy with the most time to collect the most "stuff" and maybe just good enough to parlay that into winning a single game against other players just like himself. Plus, I am honest enough to say by any measure I am probably not one of the 11 best players on the site. So it's not something I need to worry about.

Back to this thread, I think the Dotman system is 99% ready for prime time. Why don't we start with that? Does anyone have any specific objection to that? If so, what and why and what is the fix? And where can we host it?

1587 days, 16 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

This going to piss a lot of you off...

Race modifier is a non starter for the ladder. Everybody thinks races are rated differently and it cannot be subjective. Also race power depends on quality of players - for example in newbie game Privateers are one of the strongest races. In expert veteran game, they are one of the weaker ones because other veterans know all their tricks and how to counter them.

So given the following assumptions:
1. Race modifiers are subjective
2. On AVERAGE people who play multiple races tend to be better (overall) than ones who focus on single race

Why not create a ladder without race modifier and weight it by mercenary rating? Looking at mercenary list, the rankings in it are not that bad

The formula itself can be derived later.

This makes race modifier irrelevant

1587 days, 15 hours, 37 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Azz, I agree on the race thing. It seem obvious to me that the more races you play, the better you are. If only for the reason that you know how to play better against a race you have experience with. Add to that the fact that we get to pick the races we play and I see no reason for such a modifier.

However, for the same reason, I don't care if we have that modifier or not. The better player will STILL do better as long as he gets to pick a race he feels comfortable with.

If a particular race(s) always goes unchosen, give that race(s) a bump to make it more desirable. We can get carried away with trying to measure and modify everything trying to keeping the field level.

1587 days, 15 hours, 36 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Azzazzello meant no disrespect to you for your Melee game success, but in a way you prove my point. A shark amongst guppies. And many people are not shy to admit they simply join for the campaign points, then bail upon determining they cannot win.

Regular games have the benefits of everyone being evenly spaced out. Very rare in 11 player games that a race drops and no one picks up if early or before being smashed by someone. It eliminates the easy taking of space and delays the victor. The melee games are the opposite. A player drops and it is "dead" race time. All his space is free for the taking with no fight at all. No one joins these as replacements either.

Having high skill players in these along with brand new players and people looking to gain experience and campaign points simply compounds the problem. They know persistence will reward them with a high finish. 15 to 20 drops are almost assured. And you get the benefit of the add-in stellar stuff which provide almost unlimited resources for the folks next to them. No balance in the game at all.

I also disagree with your opinion on race modifiers. I say remove the subjectiveness; use their average finishes in games to develop a modifier. Looking at the Die Hard stats, it would appear the Borg would get maybe a 0.8 modifier, and the EE and Fascists a 1.2 modifier (the rest spread between these). It is not rocket science.

1587 days, 14 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Azz, I agree on your point regarding Privateers, but disagree regarding other races in Classic games. I would suggest some modifier would not only be fair, but encourage players to pick up those races that are typically the last chosen, e.g. Evil Empire and Birdmen. I understand a good player will finish well regardless of race, but if I were only interested in winning each game I would choose the Borg, Colonies, or Rebels every time.

1587 days, 14 hours, 46 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Tom N makes a good point. We need to keep the goal in mind. For me it is to encourage fun competitive games. If that means we need to give extra encouragement to get some races filled, well OK by me.

But, I still believe a good player will do better than a poor player no matter what race they play.

1587 days, 12 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The thing with a modified race score is how accurate is it? Why should one race have 1.1 x its actual score when another race has 0.9? Some one will argue according to what their theory and number crunching on this type of game they think it should be 0.95 instead of 0.90 and therefore this would make this players Federation is better than that players Robots. And what difference does that make? It will encourage someone to chose a weaker race? Will it????? Who knows.

What is the idea behind this ladder score?

To find out how good a person is at playing a specific race, yes?

To rank them in the long run for senior games /champions ship games or just boasting rights or for rank?

Or is it just the fact that some races can hit the high ranks from 1 or 2 unballanced games as Privateers and Crystalines and this is the complaint?

So I think it has to be all related to the race and game type.

All separate for the games and all separate for the race.

You get a ladder rank for playing Privateers in 11 player games. You get a score for total games and a score for your average game. This ranks you.

You cannot compare this to how a player plays as a Federation player by giving it a +20% or a -20% etc.

1587 days, 10 hours, 1 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

The modified scoring is something I could go either way on. I just don't think it is that important in the big picture.

As far as the reason for the ladder score? -
In my opinion the reasons for a position ladder:

1) To measure all players by their finish position, not just the 1-3 positions.
2) Give all players a reason to keep playing for an incremental improvement, like from 5th to 4th.
3) To reduce the value of meaningless achievements that are too easy to "game."

* I would like a ladder that only measures the last X number of games and tossed older games out. The resulting score would more accurately measure skill than an accumulation of points (by any method) earned by playing a lot of games over a long period of time.

This now leads to many players joining as many games as they can and then dropping or ignoring many. The rewards built into the current system demand this of anyone that really wants to move up the rank list as quickly as possible. Especially when they can skim achievement points from games they otherwise play poorly in. Nor does this give the rest of us the best competition and game play.

1587 days, 6 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> There are too may games set up variables.
> So no entry level and no upper level limit games should count towards the ladder system.
> So no back dating the ladder system
> So there will be a mix of regulated ladder games or non regulated private games which do not count to the ladder system.

At which point you might as well not even bother.

The Private games should, of course, be ignored. They're for testing (one player against the computer or one player against one or more of their other accounts), or a group of friends getting together for a game with nobody else to bother them. Only games that contribute to the Achievement score should be included in the Ladder.

There exists a value called "Difficulty Modifier" that is supposed to lump most of this together into a single value. It has to be collected (I added this to my data collection post above), and it has to be applied. The more difficult, the more there is to gain/lose.

When placing limits on which games do, and do not, qualify to have their results merged into the ladder, it's probably wise to remember that these games take months to complete. The more games you break out, the longer it takes to have valid scores. We're talking years here, which is one reason we need to bring in the historical scores.

Here's a quote from Dotman's ladder testing 10 months back that might be of interest to you:

"Regarding the campaign games, there's no technical reason I excluded them"

WRT Melee games, the skill that is most rewarded in that type of game is a fast buildup. This particular skil is VERY useful in other types of games, and, if mastered in Melee, could offer significant assistance to a player in other types of game. While the dynamics are different, the skills that help in a Melee game will also help in a Classic or Standard game.

> The increased rank due to your highest level race should be used to limit things.

The ladder score of the player should be applied. If the player has no score for the race, their overall score should be used. If they have no overall score, then this is their first game and they're starting at 0.

> how do you grade a Admiral entering a game that can have total newbies in (it happens especially in the Melee games)?

See my question "B" above.

In general, I expect that the higher ranking player will have much to lose and little to gain, whereas the lower ranking players will have little to lose and much to gain. This agrees with a response from @Lord+Pollax above.

I found this URL in Dotman's post from 10 months ago:

https://groups.google.com/forum/#!topic/alt.test/a6xFQTQCKDA

Further down, Dotman mentions an Average Strength Modifier:

average other players strength - your strength / 10

------------------------------

Also from the Dotman thread, FKite mentions aging scores. This is most useful for players who get to the top, then quit playing. I can come up with a multiplier to give a 1 year half-life, and apply it to players with a positive score who don't have an active game. In addition, we need a small penalty for top-ranked players who refuse a Championship game and aren't already in one. As the offer would be made manually, the penalty would also have to be applied manually.

Another suggestion from FKite was that teams should share their score. I believe that this applies to alliances as well. This is because it's impossible to determine where the various contributions come from, so saying that one of the players is 1st and the other is 2nd makes no sense. Let them both get the same "(1st+2nd)/2" score. Similar numbers can be created for any sort of team.

A potential issue with both Big Beefer's and Dotman's previous effort is that they're not "Zero Sum", and are therefore prone to inflation.

1587 days, 3 hours, 45 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

W: I got this far and just stopped reading!
> The Private games should, of course, be ignored. >

If you are referring to all custom games, even those that Joshua has approved as meeting his requirements, then you obviously do not "get it!"

There are many of us that play these games exclusively because NU does not provide us with satisfactory alternatives. They include Die Hard games, PLS games, PQ games, and so forth.

It is very important THIS is understood.

If, on the other hand, you are referring to private "test" games or others that do not meet Joshua's minimums, or are never meant for public consumption, or never request recognition... then OK. THOSE don't need to be added. But, let's be very clear on the difference.

1587 days, 3 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

You should have read the rest of the paragraph.

1587 days, 2 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The reason I support modifiers is because of what we do know: People don't win playing EE, Birds, Fascists, etc at anywhere close to the same pace as the Borg, Privateers, etc. This is regardless of skill level, since the raw stats ignore that factor (and rightly so). We don't have "better" players only playing Borg, hence their high win totals. Simply put, there are strong races and weak races (and the VGAP players KNOW this). There is no level playing ground.

The Campaign add-ons have done a great job at addressing some weaknesses, and some have even made it to the main stream games. More could be done. We know this is a fact as well.

So we know we have balance issues enough to warrant recent game changes, and we can look at the raw stats and clearly see who wins and does not win. I have no idea how you could then possibly argue that winning as the Bird is equivalent as winning as the Borg. This does not compute.

This irrational fear over fractional points someone might get shorted for finishing 5th in a game playing the whatever race is comical. Yes, the modifier would have to be assigned, but it is not completely arbitrary and is based upon game finish history. Certainly less arbitrary then assigning double beams to the Fascists for game balance.

1587 days, 2 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Given enough data-crunching, we could easily adapt the suggestion that LP is talking about. Just figure the rate at which the chosen race is winning in your chosen game style at the time of the award, normalize it to 100% at the second order, and you're golden.

Side note: It's absurd to speak of any ladder system as a zero-sum game. People are entering and leaving the Chess ladder all the time -- by virtue of learning the game and then, eventually, dying. Oh, the numbers might SEEM neat and all, but they aren't. Choose a time like, say, the last thirty years, when chess in general has become less popular, and you'll see that high versus low ratings vary over time as much as with play quality.

As such, I wouldn't spend much time even trying to ensure that any system we adopt do more than roughly approximate a zero-sum -- and that's at most.

1587 days, 0 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

How often do races win on their own?

No diplomacy? No help from anyone else? No ship trades? No arranging for someone to wipe out another player for both your advantage?

Very few games are 0 ally games.

Even the Championship have alliances going through them to arrange the winner! They just don't get a share in the win.

So a ladder system is going to be very difficult to balance.

Currently I am ranked at 8th in the first page of the top 10 for Feds (excluding the next page which includes championship players etc).

Mostly due to a slow grind with campaign team games (which I prefer) with a good balance of race combinations to compliment the Feds.

I do not think I would fare as well in a non team game 11 players. Even worse in non campaign games. So I will be nearly useless in a championship format game.

If you take the 10th player and compare his achievements to number of turns he would actually rank much higher at 4th while I would drop to 9th.

That is why I think you should separate out the ladder system and allow it to be graded by multiple means and via the games played.

1586 days, 23 hours, 11 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

You make a series of compelling points, @MartinR. The difficulty, unfortunately, is that there exist very few games (speaking statistically) for any given player. As such, it would be inadvisable to filter any out unless they are patently inapplicable.

I have never won a solo game on this site. I'm fine with that, and I don't think it makes me a bad player, or somehow unqualified to play and win a Championship game.

In the future, I think we're certain to see a Campaign championship. It may not be an Emperor's war; instead, it may be called an "Admiral's War". When that time comes, I want you to be eligible to play.

1586 days, 18 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> No replacement player should get more points than a starter. The replacement player wasn't there the first turns when diplomacy was crucial. Joining when you already have a firm alliance shows no skill in VGA Planets.

I don't think this has been adequately addressed above, so I'll provide a bit more information.

1. Diplomacy doesn't end with an "informal alliance". In fact, it's just beginning.

2. From what I can see, the majority of droppers are within the 1st 10-20 moves. At that point, they will not have made the effort to try for any alliance of any sort, and the diplomatic effort involved is minimal.

3. There's also a second group of droppers who do an inadequate job of building their economy, then encounter an adversary who will wipe them up. Instead of trying to defend their planets, they just resign or walk away from the game (get dropped). Can you honestly say that this player has demonstrated any significant amount of skill?

I believe we need a way to distribute the points between the original player and the replacement player(s).

Special case: If a replacement player resigns prior to making a move, they don't even show up as having been involved with the game.

> a ladder system is going to be very difficult to balance

It is impossible to balance "perfectly", or even well enough to make everyone happy. On the other hand, the effort required to get it to balance "well enough" is worth it to get valid scores faster.

> I wouldn't spend much time even trying to ensure that any system we adopt do more than roughly approximate a zero-sum -- and that's at most

I think that if the per-game score adjustments are close to zero-sum, the system will have a minimal amount of inflation. This is something that would have to be verified experimentally.

I looked at the Dotman run again, and it appears that he throws in a bias (initial score) of 1500. I believe that this is done so that few if any players will have a negative score. I consider that to be a nice PR angle.

------------------------------

Again, from the Dotman thread, he's using fixed initial scores of 80, 40, 20, 10, 5, 0, -5, -10, -20, -40 and -80 for 1st through 11th places. I have some issues with that, the largest being that it doesn't handle anything except an 11-player game. I previously described a method for handling a variable number of players that I think will work properly. We just need an equation to plot against.

The plot of the above scores most closely resembles a 3rd-order equation, with the 6th player being placed at the origin. I believe that's a bit too aggressive for the losers. I'd like to see the results with a few different equations (both 2nd-order with all players in Q1 and 3rd-order both with players in Q1 and in Q1/Q3) to see how it works out. Once the data is gathered from the game DB, this is a relatively small amount of additional effort.

1586 days, 17 hours, 55 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Agree with G on this. We need more data, not less.

Regarding counting Melee games:

If one ladder was developed that includes the 4 most common games (Classic-Campaign-Melee-Team), I suggest a ratio could be shown next to each player showing their preferred games. In my case I primarily play Classic Planets. Of 18 games that qualify for the ladder (excluding test and replacement games) I've started and finished 13 Classic games, 1 Campaign game, 3 Melee, and 1 Team game. That could be shown as: (13-1-3-1) or with percentages(72-6-17-6).

1586 days, 17 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Tom+N,

Good idea, but I think the ladder should contain all games, then be followed by the percentages for the common games, and a final percentage for the specials. This would only be for the overall score, not the per-race score.

1586 days, 16 hours, 44 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding a balance of score awarded between original and replacement players, I believe it should be slanted toward the the replacement and away from the original player.

Reason? Because we want to discourage drops and encourage replacements, not the other way around. The amount of work or skill involved is secondary and could also be argued that early game is easier.

1586 days, 16 hours, 37 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

While I agree with your first statement, I don't agree with your reasoning. As I've stated many times, the resigners and droppers should be handled by the Tenacity score, and we should not pollute a Skill-based score with a Tenacity-based score. This is similar to a separation of Church and State. The line's there, but it's not always easy to see exactly where it is. For that reason, I believe we should take great care to avoid the issue.

The reason I agree with your first statement is that planets are more difficult to hold onto as the game progresses. There's more effort being made to take over planets. There's a greater effort required, both militarily and diplomatically, to keep them.

1586 days, 15 hours, 45 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer: I do not see this "line" you talk about between the skill required to gather planets and the willingness to stick around and defend them. Of course, I have never agree with Tenacity as a measurement in the first place.

If we score a player lower for dropping before someone else, perhaps having less score at that time, this does impose a penalty on a drop - just as a tenacity penalty does. And I think rightly so.

The "skill" necessary to gain score is only relative and influenced by many factors beyond our control. I see nothing wrong with a single score encompassing as much as possible. The more we separate metrics the less important the ancillary ones will be.

1586 days, 15 hours, 40 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm with Mule on this one. I just don't see why a second incentive to not drop is a bad thing. Maybe I'm in the minority, but I don't concern myself with the Tenacity ratings too much. A player who starts a classic game, and then quits early on deserves hit to his ladder rating, and I DO see it as a measure of his/her lack of playing skill.

1586 days, 15 hours, 14 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I think very few player really care about tenacity unless it stops them from joining a game. Let's be honest, it is much more a device to influence behavior than a measurement. It was created entirely to discourage drops.

Using score to do the same thing is no less valid. In fact, I believe it is a better way to accomplish it. Someone that repeatedly joins and bails does not show skill in my book.

1586 days, 12 hours, 11 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> I think very few player really care about tenacity

Then you haven't been watching the newsfeed. Prior to Big Beefer's announcement that Tenacity was changing, there were a rather large number of posts where players were wondering about their Tenacity.

Currently, Joshua is creating many games with low/no Tenacity limits. Once player-created public games have the ability to limit players by Tenacity, it will become much more important.

> I DO see it as a measure of his/her lack of playing skill

Please provide this line of reasoning.

1586 days, 11 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm entirely with @Whisperer here:

(1) Tenacity is one of those major things right now. People care. Call it something else and people will care less. (I never said people were particularly perceptive, on the whole.)

(2) OK; I'm rather more AGAINST @Tom+N than with Whisperer. Tenacity, as recently conceived, is a measure of someone's skill at RL. It's only tangentially relevant to the game.

And, back to a previous post:
It's really easy to capture a ton of undefended planets. It's a pain in the ass to develop and defend them. Quitting at 100 planets and 1 starbase should grant little reward.

1586 days, 11 hours, 41 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

My line of reasoning on drops and skill is that early expansion is easy. Move some dots around and make some promises you don't intend to be around to make good on and you can easily expand your empire. It is when players encounter opposition (offense or defense) that many players decide to drop.

Even harder to track, is when they just "mail in" their turns and pay more attention to all their other games with less competition. I think this all relates to skill or a lack thereof.

Having a long term strategy that a full game requires also requires skill. Dealing with the conditions as we find then instead of quitting if things are tough. There is more to skill than just moving things about.

* All that said, I take a different view when it comes to a player that has struggled hard and long and still fails. I am much more willing to accept that randomness and circumstances may have conspired against him (or us) regardless of skill levels. But, by then, few will take these replacement spots anyway.

1586 days, 11 hours, 38 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

As far as caring about tenacity...

People care when the numbers don't add up they way they believe is right.
People care when they are told they may not be able to play in games with a low T score.

Take those away and we will find many fewer players that care.

1586 days, 10 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I guess we view droppers differently. I understand many droppers just got busy in RL and had to quit, but I suspect these are more the exception than the rule. How do I back this up? Because I've seen these same players drop and immediately join other games.

My guess is the 'typical' dropper likes the early game, and why not? It's fun. Low stress, fast, send out ships and see if there are some nice native planets nearby. They are optimistic!

Then the Mid-Game arrives. They see they are NOT keeping up in number of planets, ships, and bases. They see a Crystal to one side, and a Privateer on the other. Maybe they try to team up, but because they're behind they aren't the best choice as an ally and get turned down. Faced with a choice between several months of time consuming moves fighting to just stay alive or dropping, they drop. Some of this is due to new players not really understanding just how big a time investment these games can be.

Now this may not be the most politically correct post, but it's honestly what I've seen time and again in almost every game I've played here. So, do I think the typical dropper deserves an 11, 10, or 9 finish (and it's corresponding ding to their ladder position? Absolutely yes.

1586 days, 10 hours, 37 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Can we agree that when a player resigns or is dropped, and no replacement player takes over the position, that they should receive the lowest open completion slot available?

1586 days, 10 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If it's a ladder, @W, then yes.

1586 days, 10 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I would agree, if a player drops but a replacement takes his place the following turn, he should get a zero (no gain/loss) for his ladder score. In most cases this would mean he found someone to take over and should not be penalized. If he drops without a replacement then he gets the lowest available slot.

1586 days, 10 hours, 8 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I would agree 100% with Tom on typical droppers because this is also what I have observed. Since I am not shy about being politically incorrect, I will even say that the way achievements are now awarded and the number of games a player can join at once make this even worse.

How to score them? Also agree.

1586 days, 9 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Even the present Tenancy metric punishes serial droppers enough. There's no decent way to game the system these days -- apart from Merlin-killing, of course, and it could be argued that that particular process pays only for itself.

1586 days, 9 hours, 7 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Reminding all this thread is about a position ladder scoring.

@gnerphk says: Even the present Tenancy metric punishes serial droppers enough.

Enough as opposed to what?

Proposals:
1) You get scored in the ladder as the lowest player at the time of the drop.
2) If a replacement joins, he gets all the credit for the game.
3) A replacement gets most of the credit.
4) Other?

1586 days, 8 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm going to toss something else out there. I like the suggestions recently made here and I am seeing a good level of compromising needed to make this work out in game as something all can accept.

My toss: Maybe if we had a good AI player, the heartache and anger over droppers would be alleviated since at the very least, the race doesn't just sit there and do nothing. Or vanish magically via FoF/ kill race. Both are highly detrimental to a game. Removal of ships and defenses hurts races which acquire enemy vessels, while allowing a cake walk like the pre-turn 10 times for other races which adversely affect game strategies and resource usage. The AI player would keep the ladder points unless a replacement takes over, where it would transfer them. Ladder points still with an AI player are lost at the end of game, though it holds on to the position it finishes.

I think the hubbub over tenacity in the forum was from a handful of very loud folks who talk a lot on the forum by default. This is not bad, but in no way represents the masses of Nu players. Very few care whatsoever. Otherwise there would be a mass outcry for players who do turn activities but fail to hit "end turn" to not take an illegitimate tenacity hit. If you executed orders, you did not miss your turn. This is the equivalent of emailing in your trn file. And I certainly hate the ding for doing last minute messages, etc and hitting end turn only to get credited for missing it.

You hear a fair number of complaints, but not nearly enough to indicate some massive care factor for tenacity. Serial droppers could care less. Heck, they can just start another toon for that matter.

The idea of combining Tenacity with a Ladder sounds appealing as it will eliminate some of the lameness of tenacity and probably do a better job of ranking folks.

1586 days, 7 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

at my end, i dropped the ball maybe after fifth post or something like that but i just wanted to stick my nose in here to say "THUMBS UP FOR ALL THINGS DIE HARD!"

1586 days, 7 hours, 34 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Yes, Lord P. I think we are spending far too much time talking Tenacity.

Assuming we score a drop as the lowest at the time of the drop and assuming most serial droppers drop relatively early, this will address that problem no matter what tenacity does.

A player that hangs in until close to the end and then dies or drops, will be rewarded with a higher score than the early droppers. I think this is a elegant, measured, and simple solution.

And anyone willing to replace a dropped player needs to be given a decent chance to score some position points. Otherwise we REALLY need an AI, because few of us would want such a job.

1586 days, 7 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Maybe if we had a good AI player

I doubt that's in the cards for anytime soon. A smart AI player would be rather difficult to create, as it would need to look farther ahead than it currently does. It would also consume considerably more CPU time.

------------------------------

While I don't like the idea of punishing a dropper with a skill-based score, I do have an idea that I think might work, but before I mention it I'd like to see some more buy-in on punishment vs. earned score method for droppers.

------------------------------

I've come up with an Excel spreadsheet that allows an equation of the form "Y=AX^3+BX^2+CX" to be entered (it accepts the three coefficients), along with the number of player slots, the X-spread of the initial distribution (X[0] and X[n]) and the final scale (difference between 1st and last place Modifiers), and creates a set of zero-sum Rank Modifiers based on the provided data. If anyone is interested in playing with this, I'll make it available.

------------------------------

Finally, I've documented what I can of the history of the Ladder system in VGAP (focusing on the algorithms), along with what has been decided here. If anyone has comments on the contents, please post them. There are also a few unknowns that would be handy to fill in.

http://help.planets.nu/Ladder

1586 days, 6 hours, 59 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding game positions >11...

I was wondering if that is really much of an issue. Every Melee I have researched seems to resolve to far fewer players fairly quickly. What if all players with a lower score than #10 share an equal score equivalent to #11?

Alternately, if we really need to keep a zero sum calculation going, subdivide the #11 score into a second value set which equals the #11 score. It would give everyone more incentive to hang around or fight harder to get into the top 10.

Finally, I question the definition of "punishment" for a dropper. Are we talking about something OTHER than just giving a dropping player the lowest score at the time he drops? Because I don't consider that punishment, just a fair allocation of points.

1586 days, 6 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

In this instance, your definition of "fair" is similar to my definition of "punishment".

1586 days, 6 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Clean thoughts bounced off a dirty wall:

(1) Personally, I would rate a player higher who dies via FoF with 4 planets on turn 50 than a player who exits (resigns, is dropped, etc.) on turn 30 with 20 planets.

(2) The word Melee in Melee games indicates that these are not 4X games like an 11-player game. They are slugfests like video games. They are used to try out strategies. Win? Why?

(3) A player who gets first with a solo victory has a much better victory than a player who gets first in a 2-player alliance victory.

(4) Diplomacy, the premier online multiplayer game (50+ years, thousands of players) does not use end position. It uses number of "centers," which would be number of planets in VGA Planets. Number of planets a player has at the end of game is a much better barometer than position.

1586 days, 5 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mjs68508,

> (1) Personally, I would rate a player higher who dies via FoF with 4 planets on turn 50 than a player who exits (resigns, is dropped, etc.) on turn 30 with 20 planets.

This is the leading opinion at this time.

> (3) A player who gets first with a solo victory has a much better victory than a player who gets first in a 2-player alliance victory.

I believe this has been adequately addressed. Please see the documentation under "Alliance Win".

http://help.planets.nu/Ladder

There are a few items in that document with a note of "UNDECIDED". If we could receive feedback on those items it would be appreciated.

> Number of planets a player has at the end of game is a much better barometer than position.

Only the finishing players have planets. Those that don't finish only have position. While it's possible to come up with an algorithm that handles this, it won't be pretty. If there's enough support for this idea, I'll take a shot at it.

1586 days, 4 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

>Only the finishing players have planets.

My reading is, only the finishing players get points. Non-finishing players are ranked relative to their time of departure, am I right?

--

Again -- and I know that this is going to be unpopular, but I'm not out to win an election here -- I simply must point out that what we're proposing is going to be a near-perfect analog of an Achievements system which negates all destruction or capture points.

Since I don't think that's a good idea* (the Ville phenomenon comes to mind**), I'd instead propose we _START_ by lobbying for a reduction in points granted for these things, or at least a reduction relative to those awarded for planet count et cetera. I'm not saying "no ladder"; I'm saying "fix what's broken first". It's a pretty damn easy fix.

--
* - It's a bad idea because it encourages people to quit the moment they conclude they have no hope.
** - No offense meant, Ville; you're just the most famous example at the moment.
***- Yes, I put footnotes in a Forum post. Suffer.

1586 days, 4 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> I was wondering if that is really much of an issue. Every Melee I have researched seems to resolve to far fewer players fairly quickly. What if all players with a lower score than #10 share an equal score equivalent to #11?

From the document I've been working on for Ladder scoring:

Pro - This is simple, flexible WRT the number of players and uses the same scores that the RCC Ladder system used.

Con - This will move the scoring away from a "zero-sum" score. In this case, it will have a tendency to decrease the average score below the starting point.

Con - This means that the system sees no skill difference between last place and 12th place in a 30-player Giant Melee.

1586 days, 4 hours, 41 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

IMHO....

If an "achievement" does not lead to capturing and holding more planets upon which most planets games are scored, it seems like a pretty empty achievement to me.

If we track actual "results" - again as measured by the number of planets taken and HELD until the end of the game and which result in a higher finish, that is the real achievement.

If we want to recognize things like "blow up points" or "trading the same planet back and forth" we can give badges or service stripes or some other recognition that is as meaningful (meaningless?) as the actual accomplishment.

1586 days, 4 hours, 28 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer:
It occurs to me on divisive issues that you feel are undecided it would be nice have something like the uservoice where everyone has X number of votes to "spend" on the things that matters most to them individually. Just a passing thought ...

For myself, I'm not all that concerned with how positions 12 > 35 are scored. Just so long as they are rewarded less than positions 1 > 11.

re: Fair vs Punishment...
If we do not score a dropped player at the time of the drop, what are the other options? What word is used to describe it doesn't matter that much, but we aren't calling it punishment to score the #2 player lower than the #1 player. This seems the same to me. (shrug)

1586 days, 4 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Wait -- why, in God's name, in a ladder system, would we NOT score a dropper at the time of the drop? Score him #12, tentatively, to be confirmed at the moment of his non-return.

1586 days, 2 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

To capture some of G's thoughts here, would it be possible to ladder folks via what we've discussed and have some sort of titles for players earned via destruction and captures? A quasi dual ranking process I know, but since really the two are not necessarily similar or linked it still rewards combat success even if it does not lead to game win success. Frankly I wish it was cost of construction based, and not raw tonnage. Killing a Gemini empty equals killing one full of fighters... which hurt more? I would strongly favor counting bases and any destruction they cause, minefields too.

1586 days, 0 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's precisely what I'd like to see, LP, but I want it blended into our tracking system. Otherwise, people will lose their will to fight once their doom becomes apparent; as it happens, that's precisely when a hero's will to fight is paramount. And wouldn't we all want to be a hero?

We have an opportunity here, gentlemen: We, here, now.

We can create something that doesn't just measure, that does that but also encourages, that inspires! We can generate valor and heroism in our games.

How can you look on an opportunity like that and say, "No. Not today. I am going to aspire to mediocrity." How is that even possible?

Look, we can design a system that measures achievement in play. I don't personally care if it's a modification of our present system or an independent ladder; we can do it. We'd be foolish not to do our best, which is why, despite not sharing the conviction that stats should be zero-sum, I'd encourage us to use @Whisperer's algorithm (or one based on similar principles, at least; I prefer something second-order, but that's a minor quibble).

Destruction achievement should be less than end-game achievement, and it should be based on tonnage and on military score. That's evident; we shouldn't even have to discuss it. How MUCH less than end-game... well, that's worth a good argument. But almost all of us agree on the principle, that end-game scores are more important.

For myself, I'd like to say that I'm tired of seeing us represented in championship games by players who never win. We lucked out in Capricorn; we got a solid group of players, and as a result the game is going on bloody forever. I'm loving it.

If we can devise something that fixes that, I'll be happier. But if we build something that's "good enough", you'll be facing my scorn for a long time.

And you'll deserve it.

1586 days, 0 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Minor modification: @W's algorithm IS second-order. Don't forget to halve it (quarter post-square); we're looking at a deviation from a mean. IMO, we should probably do the same thing for per-race win percentages, but halve it again, since it can be argued that a better player will simply seek out the best race. (Go Fascists!)

But that's a side note. Just wanted to say sorry to have misread you, @W.

1585 days, 20 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"> (3) A player who gets first with a solo victory has a much better victory than a player who gets first in a 2-player alliance victory.

I believe this has been adequately addressed. Please see the documentation under "Alliance Win".

http://help.planets.nu/Ladder"

"Alliance Win

UNDECIDED

If an alliance wins, all members of the alliance receive the top positions. The total points of these positions are averaged, and all members of the alliance receive that average."

-------------------------------------

So, if I understand right and if:

1st gets 50 points
2nd gets 40 points
6th gets 0 points
10th gets -40 points
11th gets -50 points

then solo win gets 50 points

and alliance win with 1st and second places averaged gets 45 points each.

Bad, very, very bad. A solo win is 2-3 times harder than an alliance win. I have done both. A solo win is a gruelling experience. An alliance win is fairly easy.

------------------------------------

Remedy: All points should be halved in a 2-player alliance victory or divided by 3 in a 3 player alliance victory.

2 Player alliance victory example continued:

1st gets 25 points
2nd gets 20 points
6th gets 0 points
10th gets -20 points
11th gets -25 points

Even losers should lose less in a 2 player alliance victory. It is harder to survive fighting a hidden 2 player alliance (Most alliances are not announced until near the end of the game) (especially if they ally before the game) than survive against a solo victor.

-----------------------------------

There definitely needs to be 3 ladders to join the one ladder we have now (blitz):

8-14 player non-team games

15-30 player non-team games

Team Games - And, team games should have each exact team rated, not just a player's team game score.

Example: Fred, John, Frank would have a separate rating from Fred, John, Tom or just Fred, Tom. They are all separate entities. It would also encourage true, lasting teams to form.

1585 days, 20 hours, 26 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Two suggestions:

Melee scoring:
80,40,20,10,5,3,2,1,1,1,1,1,0,0,0,0,0,o,0,0,0,0,0,-1,-1,-1,-1,-1,-2,-3,-5,-10,-20,-40,-80
It keeps to the zero sum rule.

@Gnerphk, I know you feel strongly about improving the existing Achievement system. Can I suggest a new thread for that? I feel we've gone down this road enough. Points for blowing stuff up are a head scratchers to a Crystal player as well as many other cloaking races. I won't even go into the "PO" public opinion score as I have no idea how that one gets computed. Let's focus on a clean ladder...simple and easy to understand, and those that want to visit the Achievement scoreboard can always do so.

1585 days, 20 hours, 20 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"> Number of planets a player has at the end of game is a much better barometer than position.

Only the finishing players have planets. Those that don't finish only have position. While it's possible to come up with an algorithm that handles this, it won't be pretty. If there's enough support for this idea, I'll take a shot at it."

---------------------------------------------------------

Use the same algorythem as Diplomacy. Each player starts with 1/11 of the planets (in theory). 1/11 of 500 is 45.45.

If you end up with zero planets, your rating result for that game is -45.45.

If you come in first with 232 planets, your ratings gain is (232 - 45.45).

It is a zero sum rating system.

Of course, for alliance victories, all results are divided by the number of alliance members.

In team games, say 4 teams of 3, all teams start with 3/12 of the planets.

-------------------------------------------------------------

Why are people trying to reinvent the wheel? The Diplomacy ELO system has worked for 50 years. Why do people think they can come up with a better idea?

1585 days, 20 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's an interesting perspective, MJ. I don't grant your basic premises, but your logic is impressive.

In my opinion, it's far more difficult to win with a team than individually. When done in a competitive environment, it requires coordination, unity of effort, extensive communication, and above all an individual committing to a cause that is greater than himself. That takes a lot of effort.

Granted, as a matter of convenience, two powers might possibly agree to ally in order to end a game; this is the exception to the above statement. But that's hardly worthy either of punishment for them, and in any case it should have no impact whatsoever on any player who resigned long ago.

@Tom+N - I'm concerned that this be done properly if it is done at all. If it's not done, then at the very least we ought to make some alterations to the Achievement system, and in point of fact that would be a wise move independently. Nevertheless, it would be absurd to build an inaccurate or imprecise system just because it CAN be built; if it's going to be done, it should be done well.

1585 days, 20 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Clarification: I'm not referring to the Diplomacy ELO in my preceding post.

1585 days, 19 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Tom N
The drop off in melee scoring needs to be much more gradual.

I don't think there is much difference between 25th or 30th place finish. Bottom of the barrel is bottom of the barrel. There is no need for it to mirror top half where 1st is a huge difference from 2nd.

Something like this would make more sense imo (adjust the values as you see fit)

50,30,15,10,5,3,2,1,0,0,0,0,-1,-2,-3,-4,-4,-5,-5,-6,-6,-6,-8,-8,-8,-10,-10,-10,-10,-10

1585 days, 19 hours, 48 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I like Azz's Melee scoring better. Is it zero sum? Who cares? Will lighting strike if it is not?

+1 in support of MJS's solo scoring vs ally scoring.

Also Teams. Give teams a name (NY Yanks?) and score them as compared to other teams (Barcelona Conquistadors). Give them a separate ladder.

Is there anyone other than GP that wants to insert "blow up points" into our proposed position ladder. I know I certainly don't. Until or unless the official current NU scoring is abolished that kind of achievement is still represented. Or we could give badges without changing position results. But to lower the standing of player X because he manages to win without blowing up so much stuff? Say it ain't so!

1585 days, 18 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule mine is zero sum as well. It is just skewed toward the top and stretched at the bottom but all numbers do add up to 0

1585 days, 18 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Azzazzello, I like it. My balanced version was just a jump off point. Yours makes more sense.

@Mule, Agree that a ladder without "Blow-up points" is not a rushed or weaker version, but actually a more pure version. Folks can still go to the other scoreboard for all that other stuff, which I still contend means close to nothing to many players.

1585 days, 17 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I see a lot of good thought this morning. Let's see what we can do with it.

@Mule,

> If we do not score a dropped player at the time of the drop, what are the other options?

There are three general options available for scoring resigners and droppers.

1. Put them all below the eliminated/FoFed. This is the method used by the RCC Ladder (in fact, they had several different versions of resigner). Due to the difficulty in getting replacements for an email game, this was appropriate.

2. Put them into the lowest unoccupied slot. This currently appears to have the most support.

3. Merge the work of all players in the slot and create an algorithm that "evenly" distributes the end score among them.

I was originally thinking about #3, but getting agreement on the algorithm would probably be impossible. At this time, I'll support #2, but I won't like it.

@Tom+N,

The scoring you presented is even worse than the scoring presented by Mule (treat all under 11th as 11th), as it shows virtually no difference in the middle, which is where we actually start to see a difference in demonstrated skill, and a huge difference at the bottom, which is where we see almost no difference in demonstrated skill.

@Mjs68508,

> There definitely needs to be 3 ladders to join the one ladder we have now (blitz):

Now we're back to requiring years to get valid scores for most players.

> team games should have each exact team rated, not just a player's team game score

Virtually no teams would ever receive valid scores. We might as well not score them at all.

> Use the same algorythem as Diplomacy.

I don't have the time right now to research that, but the algorithm you presented implies that all resigners, droppers, and eliminated/FoFed players have the same level of skill. Somehow I don't believe that's correct.

What might work is to use this only for the finishers, and to dynamically create a formula, based on the points that the finishers would receive from the distribution on the curve. I can do this, but it will take some time.

> Why are people trying to reinvent the wheel?

Because the "wheels" we've seen all have major problems with our "car". The RCC wheel looks to be from a Honda Accord, and the Diplomacy wheel looks to be from a cement truck. Unfortunately, we have a big SUV.

@Azzazzello,

> The drop off in melee scoring needs to be much more gradual

I agree. I'm getting close to an equation that I believe will work. I made a tweak this morning that pulls the number of player slots into the equation, and it looks like it may work better. In the end, we'll need to see how these numbers work with real game data before any decision can be made.

Speaking of real data, the documentation I've been working on (http://help.planets.nu/Ladder) lists the information that I believe will be needed for each game. Does anyone think that additional information will be needed? Does anyone have the programming skill and API experience necessary to extract this information so that we can run some models? I think we're getting close to the point where it would be useful to experiment with real data.

@Mjs68508,
> A solo win is a gruelling experience. An alliance win is fairly easy.

@Gnerphk,
> it's far more difficult to win with a team than individually.

I believe that both are correct. A solo win is more difficult militarily, but an alliance win is more difficult diplomatically. It all depends on how the player thinks. As we have no way to judge that, I believe everyone will need to compromise somewhat. I also believe that Mjs68508 should rerun his analysis using the RCC ladder numbers, as they're closer to what I expect to see than what he used.

------------------------------

WRT "zero-sum", the closer we can stay to it, the more stable the end results will be. We should not design in intentional significant deviations from a zero-sum scoring system, as that would tend to cause inflation or deflation of the overall scores. Either one would imply a potential method to "game" the scoring system.

While the Average Strength Modifier is unlikely to be zero-sum in any one game, I believe that, over the number of games we're dealing with, it should average out.

------------------------------

Has anyone thought about how (or if) we should implement the Planets.nu Difficulty Modifier?

------------------------------

For players who exit the game in a single turn, what order do they take their places? In terms of demonstrated skill (lowest to highest), I would put it at Eliminated, FoFed, dropped and resigned. This is due to the overall size of their empires at the time of departure. For the dropped vs. eliminated, the dropped obviously doesn't have the skill to click on "Resign", so is placed lower than those who do have that skill.

Are there any other ideas?

1585 days, 17 hours, 38 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding existing data:

Just looking at my own history I see games that should be included and others that should not. In most cases a sort to toss out any games that do not have any achievements attached correctly identifies which is which.

However, I do see a game (Old School) that meet all Joshua's requirements and which he said he would count, but it appears he never got around to it.

So, after doing an initial game sort and coming up with a list of games to include, I think (at least to get started) we should have a chance to lobby for additional game inclusion or exclusions. This would help assure a more correct data set.

Perhaps BB would be willing to make appeals rulings?

1585 days, 17 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Azzazzello (and anyone else who might be interested),

For an 11-player game, right now my algorithm for equal distribution along an equation yields the following for 11th through 1st:

-56.47058824, -54.58823529, -49.56862745, -41.41176471, -30.11764706, -15.68627451, 1.882352941, 22.58823529, 46.43137255, 73.41176471, 103.5294118

This still needs some tweaking, but it is zero-sum. I think I need to make the curve a bit steeper near 1st place.

If we combine Mjs68508's suggestion for finishers (score based on planets), this might actually be good enough. Opinions?

1585 days, 17 hours, 12 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

>Melee scoring:
>80,40,20,10,5,3,2,1,1,1,1,1,0,0,0,0,0,o,0,0,0,0,0,-1,-1,-1,-1,-1,-2,-3,>-5,-10,-20,-40,-80
>It keeps to the zero sum rule.

As Melee games tend to pick up a few beginners and ruthless experienced players know to pick on the beginners you will see a lot of beginners stung with a quick exit.

So I would like to see any ladder system being modified for entry level.

Beginner level with max rank limitations should have a lot less "punishment" for being killed early on. Droppers should be punished but those who try but get mobbed in a beginners game should not be punished so much.

Yes you want to reward the winners but to totally demoralise a rookie in his first few games as he is learning the game is it a bit wrong.

Dropping yes should get punished (but is that not what the tenacity system is for?). Being attacked by 3 neighbours and fighting to the bitter end should not get punished.

So beginner levels with limited rank should not have a loss for less than 11.

But this would make the ladder system unbalanced for low ranking.

But possibly best to only start the ladder system after a person gets to a certain rank or finished a certain number of games?

More to add to the possibilities

1585 days, 16 hours, 51 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Martinr,

> those who try but get mobbed in a beginners game should not be punished so much

How do you identify a player who "gets mobbed"? I'm not sure that it can be done short of a complex AI system, and I don't think that's worth the effort. If you look at the Rank Modifiers supplied by Azzazzello and myself, you'll see that the curve has a much lower slope for the early exits, which means that no player gets a really bad score.

Also, the Average Strength Modifier does a lot to compensate for a mix of high ranking players and beginners.

------------------------------

The above numbers I came up with used a 2nd-order equation. By adding in a bit of 3rd-order, I'm able to get a difference of about 40 points between 1st and 2nd in an 11-player game. Using that same algorithm, the difference between 1st and 2nd in a 30-player game is only about 16 points. This is keeping with the 160 point spread between 1st and last that the RCC system uses. Increasing the difference any more in a 30-player game will result in extreme compression of the last several places. Right now, we're down to a ~0.2 point difference between 29th and 30th.

1585 days, 16 hours, 44 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I am in agreement with W on this one.

As far as getting "mobbed" I can see how this would be difficult or near to impossible for a program to determine. A better solution would be Melees restricted to beginners. Maybe a stern warning in a new player's welcoming packet about being careful not to step up too soon. Then they can make their own choices.

1585 days, 16 hours, 36 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

So in low rank melee games you don't count for the ladder or only reward a high win and not coming last unless you resign or miss 3 turns.

Or have a low rank melee end game moderator who gives a special "you tried your best but were beaten to a pulp horrifically by your neighbours award"

1585 days, 16 hours, 33 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I disagree, you should be evaluating a player on several levels: Support to allies, dependability in politics to agreements, knowledge of race they play, success playing their chosen race, experience of the game (how long playing), number of wins, number of losses, and number of allie wins. This total in points will definitely define the player.

1585 days, 16 hours, 29 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I still think my "count everyone <11 the same in Melees" answers these problems.

Beginners can play without getting a huge hit for not doing well and we won't have experienced players exploiting hugely slanted Melees because we are trying to make them zero sum balanced.

1585 days, 16 hours, 29 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

While that's technically true, @NinjaBunny, I have two objections:
(1) It's rather difficult to measure some of those things in numbers.
(2) You recently joined a team game, played a couple of turns, and dropped, stranding your allies. Your opinion is uninteresting to me.

1585 days, 16 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

For what it's worth, I'm with @Mule on this one. Zero-sum is ideal only as a concept; don't be a slave to it.

1585 days, 16 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> exploiting hugely slanted Melees because we are trying to make them zero sum balanced.

Please describe the "exploit". If it's real there's probably an easy fix.

@Gnerphk,

> Zero-sum is ideal only as a concept; don't be a slave to it.

If it's easy to have the scores reflect what needs to be reflected, then we should stick with it. Only when it becomes difficult should we consider other methods. So far, it's been simple.

The implied question in my last post was whether or not a 16-point difference between 1st and 2nd in a Melee game was enough. Based on the responses, I think not.

@Mule,

> I still think my "count everyone <11 the same in Melees" answers these problems.

With the increased compression at the low-end of the scale, we're getting closer to that, but staying just far enough from it to say that better placement gives a better score.

1585 days, 16 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@W, My Melee suggestion was a rushed example to move the debate along only. I agree Azz has a better 'zero-sum' example. Ny the way, thanks for your efforts on that "Ladder Page"! Nicely done.

I hope the site keeps a Melee game always available to beginners and experts alike that is exempt from ladders and tenacity. Nice to have a place to go to try out new races and tactics for a while. Call it a "Practice Melee".

As for our leaderboard, once the algorithm is agreed upon I suggest two ladders. One for Teams and one for all other qualified games: Classic, Campaign, Melee, Custom games that meet site minimum requirements.

1585 days, 15 hours, 49 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I've seen 3 plans and just showing top and bottom scores:

Tom: 80,40,20,10, >> -10,-20,-40,-80
Azz: 50,30,15,10, >> -10,-10,-10,-10

In Tom's 1st out takes a pretty big hit (-80).
Azz's is better for both extremes without the high end being too far out.

W's list (note: This is reversed order from above): -61.6091954, -59.89694808, -57.91291546, -55.65709756, >> 71.23945416, 80.01811902, 89.06856916, 98.3908046

W's would work OK but -61 for first out is still a pretty bit hit. I guess that's OK. Just don't be first out. :)

Looking at the 11 player...
-56.47058824, -54.58823529, -49.56862745, -41.41176471, -30.11764706, -15.68627451, 1.882352941, 22.58823529, 46.43137255, 73.41176471, 103.5294118

It appears close enough to be fair,...

OK, no huge exploits jump out at me. :)

1585 days, 15 hours, 38 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

BTW...

Before we start dishing out minus scores it would be really nice to make it clear to everyone. There will be new players that are not privy to this discussion and join a Melee - or even worse, just be assigned to one - and suddenly find they are getting a minus score on a ladder they were not even aware of. They will not be happy campers.

We need at least enough cooperation from NU to make sure nobody get blindsided.

1585 days, 15 hours, 33 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule said> "There will be new players that are not privy to this discussion and join a Melee - or even worse, just be assigned to one - and suddenly find they are getting a minus score on a ladder they were not even aware of. "

Another reason to have two types of Melee games available, one for fun and one that counts toward the ladder.

1585 days, 15 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Who cares that they get a minus? It should not impact them in terms of resources or anything else - it just ranks them relative to other players. If they are new, then likely they are not as good as others and should be moved lower.

Getting -x is not necessarily a bad thing, it is just an adjustment to your standings compared to the rest of the crowd.

Resource allocation can still be done via old system

1585 days, 15 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Perhaps there should be an "opt in" feature or games which clearly count toward ladder achievements and games clearly marked for practice/ no ladder movements.

I'm not a big fan of the melee game, since it is not VGAP to me, but some perversion of it. Won't make me popular, but hey... as Gnerphk said I ain't running for office. Same goes for Blitz games. I am thinking the ladder feature needs to segregate the game styles. Same ladder can be used as a mechanic, but just separately applied to each classification of game. Thus Emork may be top dog for the classic, and Mule top dog for the Team, and Ninjabunny top dog for the droppers, etc.

This would allow folks who specialize in Blitz to get properly ranked with other Blitz players. Same with Melee and the others. I even like the idea of formed teams that have some permanence. A league could be made. Ranks assigned to team outcomes.

I like W's scoring as it seems fair and reasonably awards the victors.

I think I would like destructions/ captures included but I see it as extra frosting on the cake. I could live without it. I would love to see badges developed though at a minimum.

1585 days, 15 hours, 5 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Azz, I understand your point, but not everyone here even cares about resources - assuming you are referencing their use in campaigns. I, for one, could care less about that. But, I do care about ranking based on game positions relative to other players. I can only assume I am not the only one that just wants to play Classic games and be ranked by the results.

All I am saying is players should be made aware of the ladder. And, as Tom N suggests, maybe have some games that are not included in the ladder. Especially those Melees that new players are just assigned to.

1585 days, 15 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I believe you've got a good idea there, LP. Perhaps we could even restrict it to Senior games or something similar.

I dislike the thought of discouraging beginners in ANY way at all. I'd just been nerving myself to strongly recommend that only the top 100 ever be displayed at any point for just exactly that reason.

1585 days, 14 hours, 53 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP: Regarding just listing the top 100... I could not disagree more!

Provide a penalty free "sandbox" for beginners to play in or experienced players to experiment in is enough. But as soon as we start just having an elitist top 100 ladder is when I really will bail out of here. :(

Please do not take this personal, but this sounds like just another way to submarine this whole position ladder idea that I think many (most?) of us want.

1585 days, 14 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> The implied question in my last post was whether or not a 16-point difference between 1st and 2nd in a Melee game was enough. Based on the responses, I think not.

That's about as good as I could do quickly with a 3rd-order equation. I moved to 4th-order equation and got a difference of 21, but with the first 20 places having a negative Rank Modifier. I think the returns are diminishing, and the overall picture is becoming bleak for this approach when used by itself. Is it worth trying to get more than this?

What we don't want to do is have scores of Melee games eclipse scores of Classic, Standard, Campaign or Championship games, which means the spread between 1st and last needs to be similar. Additionally, we need to have some point-spread between the lower players, so they can see that continuing to play will help their score, or at least hurt it less. A side effect of both of these is to decrease the gap between 1st and 2nd in large games.

I believe that it's possible to have some combination of multiple approaches that will work here, but I don't have the time and energy right now to think about it.

Does anyone have any ideas that might solve this mesh of problems?

1585 days, 14 hours, 44 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

W: We can paralyze by analysis seeking perfection, but let's not do that.

I give your plan a +1. :)

1585 days, 14 hours, 33 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Multiple ladders

Ladders for Classic, Non classic / campaign, Team, Melee and Blitz etc.

Each ladder has a score which then feeds into the overall score for who is the best of the best.

People who prefer a set type of game can compare themselves to their game peers (those who only play one set of games). This can be used to arrange special leader board games etc.

People who do a mix of games can compare the whole list.

Also make it so that people can change the ladder so they can measure their rank on tons destroyed or ships captured etc (I am quite proud of my Fed tonnage destroyed :-) ).

1585 days, 14 hours, 30 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I have no issues with multiple ladders or sorts as long as we have one ladder to rule them all. :)

1585 days, 14 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I just added these ideas for Replacement players and Returning players to the document.

http://planets.nu/#/howtoplay/Ladder

Replacement Players

When a replacement player comes into a game, the score table will expand by one. If, at any time, the replacement player is dropped for missing turns, they will receive the drop score, but only if it is negative after the modifiers have been applied. Allowing a positive score change for a player dropping would allow replacement players to abuse the system.

If the replacement player resigns, or if the game completes before 20 turns have elapsed, they will receive no score, and the created slot will be removed.

If the player spends more than 20 turns in the game, they will be eligible to receive a positive score when they exit, but only if they play to the end of the game.

As the replacement player has done a service to the game, they will not receive a negative score after adjustments, unless they are dropped from the game.

Returning Players

A returning player is one who exited the game (usually by being dropped) and wishes to return. The Planets.nu system will only allow a player to return to the slot they had previously.

The current suggestion is that if they missed any turns, they will count as a resignation or drop, followed by a return. If no turns were missed (rejoin on same or next turn), they will be considered as having a single contiguous play.

1585 days, 13 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

(1) I am curious. More and more games are disappearing from people's profiles and the completed games page. How will those games be factored in? When one of your losses disappears, will your rating go up? When one of your wins disappears, will your rating go down?

(2) I like the idea of starting from now and games being announced in advance that they are rated, just like in chess and Diplomacy.

(3) If you improve a position from 11th to 9th, I don't see why that person should be punished with negative points.

1585 days, 12 hours, 9 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Good points all, MJ.

1585 days, 12 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@MJS68508 said: "(2) I like the idea of starting from now"

I'm not on board of throwing out several years of data. These games take a long time as it is. The more data we start with the better, IMO.

1585 days, 11 hours, 47 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

If the data on old games is no longer available, and not just hidden or stored somewhere, how about if we just use all the data that is available?

Or, as a compromise just go back 3 or 4 years? Like Tom N, I'd hate to think all those years are just so much trash to toss out the back door.

1585 days, 10 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> If the data on old games is no longer available, and not just hidden or stored somewhere, how about if we just use all the data that is available?

I consider that to be a sensible plan.

To find the games, just pull the information from the "Finished Games" list, and throw away the games that have fewer than 8 human players.

1585 days, 8 hours, 39 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

Each time I check back this gets closer to what I hoped it would be from the beginning. My hope that everyone on the site isn't looking for "speed planets" is being restored. Huzzah!

1585 days, 5 hours, 52 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@MJ: (3) If you improve a position from 11th to 9th, I don't see why that person should be punished with negative points.

-> he is not punished, he is awarded with LESS negative points :D

1585 days, 4 hours, 29 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Didn't I see a post that replacements would not get any minus points? Yes, I just found it...

W said: >> As the replacement player has done a service to the game, they will not receive a negative score after adjustments, unless they are dropped from the game.>>

Doesn't this deal with the problem? Or do we want to say under no conditions?"

1585 days, 3 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> W said: >> As the replacement player ...

The idea was originally from @Lord+Pollax, about three days ago. There was no disagreement, and it was very good, so I put it into the document, then quoted the document.

I added the part about getting whatever is appropriate if they ignore the game for three turns and get dropped.

1584 days, 20 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

On replacement rankings, let's say their ladder score won't go down based on their finish. This is different than stating they won't get a negative score. Point being a top tier player who receives a +10 would likely see their ladder position drop.

That brings to mind another thought, and it's early here so shoot it down if it's wacky:

Why not have the ladder score LOOK like a finishing score? In other words, our best players might look like "PLAYER John Doe = 1.50 (15-3-4-3)" A quick glance would show others that John's average finish is between 1st and 2nd place, and he's played 15 classic games, 3 campaign, 4 Melee, and 3 team games. A perfect score being "1.00" and the other extreme an "11". Melees and other non-11 games would have to be mathematically converted so a 35th place finish would be given an "11". Difficulty modifier could allow a player to even IMPROVE on a perfect 1.00, similar to students getting a 4.5 grade point average in school.

1584 days, 18 hours, 24 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> On replacement rankings, let's say their ladder score won't go down based on their finish.

That's effectively the same as what I wrote:

> As the replacement player has done a service to the game, they will not receive a negative score after adjustments, unless they are dropped from the game.

Non-negative after adjustments.

> Why not have the ladder score LOOK like a finishing score?

The Ladder system scores skill, not placement in a game. This would require a parallel set of calculations.

While I see nothing wrong with this request, I believe it's outside the scope of the Ladder system, and this project is complex enough without scope creep. If you really want this, I think it would be best if it were put into another thread.

1584 days, 18 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> I can come up with a multiplier to give a 1 year half-life, and apply it to players with a positive score who don't have an active game.

The magic number is 525/526. This will give a half-life of 1 year.

1584 days, 18 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm not sure it IS the same thing. As an example, a player has 10 games started and finished ranked: 1,1,2,2,3,1,2,1,2,1. He would likely have a very high rank on the ladder, let's say 5th place amongst 2000 members. Now he joins as a replacement in 7th position. He does well and finishes in 5th place. While he's given a +5 for his finish, wouldn't this lower his average and hence his position on the ladder leaderboard? If true, then I'd suggest any finish that lowers his ladder average not be counted. If false, then maybe I misunderstand the whole system and I'd suggest we're looking and a ladder that rewards quantity over quality.

1584 days, 18 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> While he's given a +5 for his finish, wouldn't this lower his average and hence his position on the ladder leaderboard?

No.

1584 days, 16 hours, 54 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@ Whisperer >The magic number is 525/526. This will give a half-life of 1 year, if applied on a daily basis. The question is whether or not we want to do this. The "justification" for this is that the player is getting stale while not playing, thus losing some of their skill.<

So this decay occurs even if a players is actively playing games? As someone who prefers to play fewer, deeper games, I am not a fan. If the decay begins if a player goes inactive, then I'm all for it.

1584 days, 16 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> So this decay occurs even if a players is actively playing games?

Absolutely not. The intent is to keep active players at the top, and to move inactive players farther down, but only to the middle. Any player enrolled in a game, even one that's on hold, wouldn't see their score get aged.

1584 days, 14 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Any player enrolled in a public game with 8 or more slots for human players (i.e. eligible for Achievement score), even one that's on hold, wouldn't see their score get aged.

Hmmm. I think we need to widen this a bit, and at least include Training games.

1583 days, 20 hours, 14 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

No new posts for a few days. All questions answered? Maybe all interested parties have (http://help.planets.nu/Ladder) link bookmarked? Anyway, I would hate to see this idea make like an old soldier and just fade away. :)

1583 days, 19 hours, 51 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

O just thought I would re-post this from the Leo Championship thread, as it related to this thread.

>> ville kauppinen: ... ladder ...
Its safe to say our game was badly infected with different player skills. Achievement listings are extremely faulty in measuring player skill-level and only in special cases provide players with nearly equal skills to same game. Perhaps Capricorn had that kind of luck. >>

So, let's not have this die.

1583 days, 19 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> let's not have this die.

I don't intend to, but I need some ideas and agreement before I move on. Everything in the Ladder document that says "UNDECIDED" needs to be looked at from several angles. We need the best answers that we can get, limited to what we can implement (e.g. no requests for tasks that require AI will be considered at this time).

What this is waiting for now is user feedback.

1583 days, 19 hours, 11 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer: Consider the possibility that a lack of feedback may also mean everyone that cares about those specific issues has already spoken or that nobody cares enough to question whatever you decide. :)

1583 days, 18 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

There are still a few items that have no idea whatsoever, or a general idea that needs some refinement. Perhaps people would like to see the results of some of the answers before they make a decision.

Also, I need to make another pass through the design to be sure there are no significant holes, and that everything is clearly stated. I can't do that today.

Finally, I don't have the skill with the API to extract the data (I'm not a JavaScript programmer). If someone could extract the data as listed in the document (dates may or may not be available), I can start putting together a simulator so that we can see what we have (I'm quite good with C). Ideas are nice, but until we can see the results, we won't know whether or not they're any good.

1583 days, 18 hours, 28 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

W: Boy, can I relate! I used C and dabbled in C+ before giving it up. And coding was not my primary occupation anyway. I'm OK with tweaking Java, but that's about it.

Anyway, now that you have described who and what you need, perhaps we can get those people skills on the job. (Anyone listening out there?) Ditto for getting the final undecided issues addressed.

1583 days, 17 hours, 52 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Just trying to keep this moving along...

The following is a summary of Undecided from ladder page. (Details at http://help.planets.nu/Ladder) The linked page does list some Pro/Cons and suggestions.

1: Large Games (>11 players)
2: Replacement points and impact on total points.
3: Dropped & Returning Game Player.
4: Alliance point distribution.
5: Team point distribution.
6: Aging of inactive players.
7: Championship refusals.

1583 days, 15 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

One of my concerns is the idea of "active players". I am someone who is always playing, but usually 1-2 games. With the average game life completely out of my control (# turns per week keeps changing per RL issues of other players), I see a rolling time period for the ladder as a punishment. All my games will drop off and I'll have maybe 3-4 games as score-able...based on my read of suggestions. Even if they are all wins, this isn't going to get me high on the ladder.

I am thinking "active player" should mean a player without long breaks in game play. Maybe 6 months or greater with no games inactivates someone. Drops them from the ladder. You are restored upon your return with the points you had, not the ladder position you had. This ought to ensure you restart lower on the ladder as others have gained more points while you were away. Not rewarded and not punished, simply put into stasis.

1583 days, 15 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

As this is not an additive-only system, there is no guarantee that a 6-monthly gap would necessarily give a lower slot in the ladder.

I'd like to spend some time redefining what"active" is, so that users don't wonder why their score dropped.

I can see that a 30-90 day gap could happen from time to time, and I think I can add that gap as a prerequisite for score aging.

1583 days, 13 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm hoping for the system to allow "breaks" to occur, since frankly we all get tired of the game from time to time and simply want to step away from games for a bit. I chose 6 months because that is a very long hiatus for just taking a break. 30 days seems short, since some places encourage vacations that long. Probably could live with 90 days.

1583 days, 12 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm with LP that 90 days seems about right.

1583 days, 11 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Regarding the 7 undecided issues:

1. Large games: Can't they be prorated/averaged in so a 35th finish = 11th finish, etc.

2. Replacement points: Guaranty the replacement player gets >0 points to ladder, otherwise whatever their finish rewards them.

3. Dropped and returning player: If system can allow, let them continue without penalty.

4. Alliance Points: Give winning alliance (+70 and +50), which reserves +80 for a solo victory.

5. Team Point Distribution: I defer to others. Not my chosen format.

6. Aging of inactive players: I vote at 90 days they're name is removed from the ladder until they return, with some minor 'aging' to their score if they return beyond 90 days.

7. Championship refusals: I defer. Not sure this needs to be addressed right now.

1583 days, 11 hours, 43 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

I think counting players inactive at 90 days a bit quick. I wouldn't go less than 6 months. Heck, if banks give you a year to cash a check, that seems a fair time to wait before one is considered inactive.

1583 days, 6 hours, 37 minutes ago

chaka

RE: I am starting a campaign to lobby the gods of this site to...

I believe that banks give you only 6 months to cash a check. Depends on where you are living maybe...

1583 days, 6 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Just got back from the tax guy, and I can now try to do something useful.

@Tom+N,

> 1 ...

That's what I'm looking at now. There are several posts above on my progress.

> 2 ...

With the exception of the replacement player being dropped due to missed turns, that's the plan. Do you believe we should remove that exception? What is your reasoning for this?

> 3 ...

As it hasn't been coded yet, it can probably be made to happen. Could you please provide the reasoning for this? Why should a player who repeatedly ignores the game be allowed to continue to do so without penalty, instead of giving someone else a shot?

> 4 ...

I think I can figure out a way to do this.

There are two ways to reward alliance winners. One way assumes that an alliance is "run" by the player with the most planets, and rewards them as such. The other way assumes that an alliance is a team where the effort is shared, and rewards them equally.

Due to the benefits of a "solo win", I'm assuming that the players are somewhat closer to sharing than master/slave. Is this in error?

> 7 ,..

Agreed, but it will need a decision sooner or later. I think input from the people who have played the Championship games would be useful.

> 6 ...

> I'm hoping for the system to allow "breaks" to occur, since frankly we all get tired of the game from time to time and simply want to step away from games for a bit.

Let's look at this in detail.

I'm sure it CAN be done, but it's an increase in complexity. For that increase, there should be an overall improvement in the quality of the scores.

The question is whether or not it's appropriate for a player to come back after a 6-month or 1-year break and be EXACTLY where they were. This is a dynamic game. The players are learning and the game itself is changing. Here's an example that should clarify this.

Given: We have two players with very close scores. One player continues playing, but doesn't improve their score. The other player takes a 1 year break.

Question: At the end of the year, which player will be better?

My expectation is that the player who's continued to be active will be better, even though their score hasn't increased. This expectation is based on the fact that the player who continued playing has kept up with the changes, and has continued using the interface, while the player who took the break hasn't kept up with the changes and has likely become rusty.

While it's probably POSSIBLE to hibernate the score of a player taking a break, and it will make the player happier, is it best for the other players that this happen? Does it make the scoring system more correct or less correct?

1583 days, 3 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer big thanks and respect for the effort to form the ladder rules!

Issue 7: Championship refusal.

I have a feeling that there will be very few really good players at the top of the ladder who would refuse a championship game. It could be a non-issue really. Also I do not trust visibility for 3rd parties on this, because the signup process is lengthy and not transparent, and ratings change during it. So admin effort would be needed.

On expiration:

I agree that there are and will be names from the past who gained a row of first places when the median skill of the player community was a lot lower than what it is now, when MBC and other advanced tactics weren't common knowledge yet. Players who haven't been active in several years should slowly expire from the ladder.

However we need to be extremely careful on how we do anything like that. The fact is that there are a lot of very good low-volume players. This is emphasized when people put real effort in the games and really try to win against tough competition. Those who put a lot of effort in tend to due well but as a consequence might only finish one game a year or so. Whatever system we create, it must cater for this player, otherwise the whole thing is pointless.

Expanding this to the idea of taking a break from planets. Many players do this, some of them very good. Let's not create too much pressure for people to keep constantly playing. Planets Burnout is a real thing. We have currently two emperors taking a longer break from Planets. Not only would it be unfair to them to lose anything significant on the ladder from taking a break, it would also be unfair to the other players who subsequently get steamrolled when they come back.

Lastly, there are the private/experimental games. A player could be playing several private games but no public games. There is no way to rank the private games sensibly for the ladder. My point here is to emphasize again that ladder loss for long breaks should happen very slowly and very carefully.

1583 days, 1 hours, 24 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I can probably fairly comment on the Penalties and Championship refusal as I refused the last Championship game.

The reason for that was that I was very busy and I didn't feel I could commit the time. I don't play more than 2 games at once, and even 1 can be hard work when your doing well.
The other reason which is becoming very apparent to me is that they simply don't look like much fun. From the threads on each there seems to be a lot of people unhappy and not enjoying the games, so you've got to question why bother?

The no allies format just seems to strip out a huge part of the game, and I also like the Campaign options having played the Fascists recently.

Should I be penalized for being busy or not wanted to do a game that I probably wont enjoy? I don't think so.

BUT

I've learnt in this game that life's too short to do things you don't enjoy. Its also too short to worry about a position in a ladder that you don't really care about so I cant get too excited about it, and if people really want to measure their enjoyment based on their place in a ladder then they must have a different mind set to me on WHY they play the game.

I wouldn't object to the penalties, but I don't think they are right.

1583 days, 0 hours, 51 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Too much discussion :D Now we need someone to implement it one way or another (3rd party or developer) and just say this is the way it is going to be, like it or not.

OR, just say we don't need a ladder, achievements are great and end of story.

1582 days, 22 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

We don't NEED a ladder. (I'm not convinced it's useless, though.)

Achievements COULD BE great, if only.

1582 days, 21 hours, 29 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

Thats the spirit. But like in every project there comes a time to do actually something or bury it :D

I've already registered as a lazy bastard but I'm hoping there is at least one productive member in the feed.

1582 days, 20 hours, 29 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Smn & @Pobs,

Thank you for the information on the Championship games. Based on that, it's on hold for now until we have the results from the simulations.

@Smn,

> Not only would it be unfair to them to lose anything significant on the ladder from taking a break, it would also be unfair to the other players who subsequently get steamrolled when they come back.

I mentioned above that limitation to a public game might be too restrictive. If we expand this to include private games, I don't believe it will be able to be simulated using the extracted data, as I don't believe any information on completed private games is in the data that's publicly available. This is on hold for now as well, again, until I can see the data.

In addition, I'm not sure that there's any interface, except checking each user individually, that will provide information on active private games. If this is the case, then the entire score aging idea might die. If we can't get the data without placing undue load on Planets.nu, we can't do the aging.

It's obvious that when they come back, some of their technical abilities will be lost, but they'll be fresh and ready to absorb the changes. While I believe that a short break will probably not cause loss of anything, and that a small amount of aging is appropriate, selecting the numbers for that will be a bit tricky. With luck, the simulations will provide helpful information, possibly even that we don't have the raw data to do this.

@Ville+Kauppinen,

> Now we need someone to implement it one way or another

We're working on some final details of the design, but the request for data has gone out. As the RCC scoring won't work for us, we need to find a scoring system that will. I'd like the results of testing on 11-player games to have a similar order to the results using the RCC scores. After that, we can expand it to all public games with 8 or more players.

Fail to design === Design to fail.

1582 days, 20 hours, 9 minutes ago

ville kauppinen

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer - a bit off-topic but in my experience from various organizations, small and global, projects with (too much) design have almost always failed someway. Budget, time or result-wise.

Best results are achieved with agile development processes where prototyping starts almost immediately after concept is ready and then finetuning it version and iteration by iteration.

1582 days, 19 hours, 1 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I have to agree with Pobs. I see no reason for penalizing anyone for refusing an invite to a Champ game. It's not everyone's cup of tea, nor is it any kind of measure of skill level.

Achievement points = blow'em up points. If blowing up stuff leads to better finishing positions, they will be rewarded by a position ladder. If not, it is silly to acknowledge them except by badges or service strips or whatever.

1582 days, 18 hours, 29 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ville+Kauppinen,

Not off-topic at all. You want to have some reasonable amount of assurance that something will be delivered.

The design we're working on is of WHAT it will do, not HOW. The HOW will be done by Agile methodologies. Much of the WHAT needs to be known beforehand, so that Agile has some enough initial guidance to get something out the door.

1582 days, 18 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Why should someone be punished for not committing to a game they don't want to play in due to various reasons?

Why should a person on a ladder be punished just because they cannot commit to a Championship game? A game format they may not like or have RL issues at the start of the game and they cannot commit fully to a game that takes a lot of effort to play well.

In fact it should be the opposite. I think the people who turn down a Championship game should get recognition. A badge to say they were deemed worthy and were offered the chance to play but they thought that someone else should have the chance to play in their place.

People who are not playing in any games should be flagged as non active. Their overall scores should be frozen (apart from any element about recent games) and it be reinstated if they start playing again. Say if someone has not been in a game for a month or two they are flagged as inactive and any ladder system can be judged by active and inactive players. If your not actually on the site anymore should you still be eligible for Championship games?

If set up properly the ladder should be able to read from all angles by the people who want to.

Of course entry to certain games will be regulated by the site in its set method.

1582 days, 17 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

In the computer age, why can't there be multiple lists? that should be the easiest part of the programming.

List 1: Active
List 2: Everyone

1582 days, 17 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer,

Regarding point 2, Replacement player gets dropped for missing 3 turns I agree they should receive lowest available score, same as an original player would.

Regarding point 3, Dropped player who returns. I'm on the fence here and doubt this comes up very often. When it does I think it's a case of a player who had a tough, busy week in RL, got dropped, then returned to find his race still available. My opinion is let them return to the game without this blip affecting his ladder score. He will likely pay a price 'in game', but if he can continue to the finish then he's doing a service to that game and deserves whatever finish he gets.

Regarding point 4, My experience with alliances is the member with the most planets has done the heaviest lifting and deserves the higher point total. If it didn't make things too complicated I would even suggest an equation that rewards the alpha member an even higher portion of points, e.g. if there is a lopsided planet total, say 200 for alpha and 50 for the beta member. In any case a solo victory should alway receive the highest available points IMO.

@Pobs, Agreed on Championships. I even suggested championships ought to simulate what most of us play (and have played for 20 years), which are games that allow alliances.

@Ville, I'm with you on getting the ball rolling, but unless I missed something we don't have a volunteer yet. We might have a better chance of attracting one if we can put forth exactly what that entails. Are they creating a ladder as demonstration only? Who maintains it?

1582 days, 16 hours, 55 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I've marked the penalty for refusal to play in a Championship game as "Refused". I'd like to keep it in the doc, as it shows that this was considered but was refused, and the reasons that it was refused.

I'm thinking about the words needed for some of these other ideas.

> unless I missed something we don't have a volunteer yet

While I can work on the back-end code, I'm not all that good at writing UI code, and I haven't ever worked with JSON (the Planets.nu API uses JSON). Also, I can't host this.

Fortunately, all we need at this stage is to run the simulations to create viable algorithms. Later, we'll need to occasionally create some static pages. I can write the "C" code to do both of these, and I expect I can learn to use JSON. As there has been no response to the request for data, I expect I'll have to figure out the API, which will add time. The good news is that the code should be portable.

We need a volunteer for long-term (1-2 years?) hosting. Once we have that, I can start working on the back-end (I need to know the OS). As Planets.nu runs Windows, a Windows/IIS server is preferred.

1582 days, 16 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I found the API documentation (such as it is) at http://planets.nu/api-documentation.

Parsing JSON doesn't look all that difficult, and I found "LINQ Pocket Reference" on my Kindle, so that should help. I'll start moving forward with gathering the information. As LINQ-to-JSON requires C#, it pretty much limits us to a Windows/IIS server.

I expect that this may take a bit of time.

1582 days, 15 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer My thoughts on your analogy with 2 players at the same level: If a guy plays for a year and doesn't move relative to the guy that was away a year, he did not play well and did not deserve a rise in position. I am thinking with both not moving, yet the entire ladder is, that both would drop on the ladder in your scenario. One guy for not playing, and the other for not playing well.

I agree that declining a Championship off should be no penalty in ladder position points, but he SHOULD be marked somehow to denote he was offered and refused so he is automatically passed on in the future. Opportunities for Emperor come but once, in my mind. Seize it, or be forever unworthy. Until you work your way back up a ladder that is for another crack at it. Otherwise the top will be log jammed with folks not wanting to risk anything and people with aspirations of greatness denied their proper spot on top. No one is being forced to play, but you are forced to make a choice. Opportunities at Emperor should not be bankable (to whenever I feel like doing it). My opinion.

1582 days, 15 hours, 4 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Championships:

1) Until NU decides to toss the achievement based ranking in favor of our ladder, we don't have to worry about Championship Games. I suspect we will be running 2 ranking systems in tandem for some time. So, let's not sidetrack getting a ladder working on an issue that may never be an pertinent.

2) While some might consider playing in a Championship game a rare opportunity, I don't think that should be a factor in a ladder that proposes to rank players based on their game results. It is an single apple in a crate of oranges.

1582 days, 14 hours, 52 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax

So if it happens to be a really bad time the day you get invited and can't do it right then, you are forever unworthy? Your statement make it sound as if you believe that people either need to make the game their life or they aren't worthy of playing in the Championship games. Please tell me if I'm misreading you.

Your comments seem contradictory. You first say they should be automatically passed on in the future. Once turned down, they can never try again. Then you say they can have another crack at it if they work their way back up. Which are you advocating, because they are mutually exclusive opinions.

If a person turns it down and they take no penalty for it, then they should simply have to maintain their position on top until the next game rolls around. Is this not correct? I agree with you that a person should not be granted an offer simply because they did not take the chance the first time, they still need to be on top of the pile when the next game starts. But they should still be allowed that chance if they have earned it.

1582 days, 14 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@slash Nothing in here is real; its a game. I get that. I completely understand when RL gets in the way of gaming. It is a good choice to deal with RL. But your RL isn't my RL. Why should any RL issues I have impact you or the rest of the site? If the issues are that important, frankly the rankings I would lose on Nu would not even be on my radar.

I said I don't believe you should be penalized any points, not ladder positions. Keep your points to show all how good your were, but lose your rank/ ladder position by being bypassed to allow others to progress to the Championship. A simple asterisk would work to mark your unworthiness to be Emperor.

It is not the games fault, or other players fault, that RL bites someone in the butt. Its tough luck...like getting ion storms over your core. But we don't end games because someone is having tough luck in RL. We accept their resignation with our best wishes.

You don't want to play for Emperor? Fine. We accept your resignation from the ladder and wish you well. No different.

I agree about not getting sidetracked. But I also think the group as a whole are thinking about this because I can see this becoming an issue later on. Too many folks don't have the courage to step up, and frankly I feel many use "RL" as an excuse. Only play what you can manage, and finish what you play. DIE HARD!

1582 days, 13 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

And looking at the current Championship games is it really worth the effort to become Emperor?

Sounds with all the bitching from both games its not worth it.

Sounds a lot like our current election campaign in the UK.

You stress out making deals with everyone trying to negotiate your way to power. Your promise this and that to everyone. Then change it a few days later when the other side gives a better deal.

And like the UK elections all the promises that you get disappear as you get stabbed in the back at the point of victory.

1582 days, 13 hours, 21 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax I have no interest in becoming Emperor myself, so I'm not speaking out of personal interest, other than a general sense of fairness. It has nothing to do with courage and frankly, assuming that of anyone is rather insulting. For me, it has everything to do with the amount of time I choose to spend on the game and my lack of interest in that type of comparison.

I would never expect someone to end a game due to real life, but pausing a game is reasonable. In the MvM game I am in right now, we have told Gnerphk that if he needs to pause the game right now, we are fine with that. Sometimes RL does get in the way and it is only courteous to give them a chance. So yes, I don't have a problem with other people's real lives affecting me. We let other people's lives affect us every day. It is generally considered common courtesy to be considerate of other people's situations. Why should the game be any different?

Of course, there are limits. You can't stop the game every time someone is late for a turn. Being late with one's turns is inconsiderate to the other players. If someone can't or won't get their turns in, they should be dropped.

I completely agree with you that people should only play what they can manage and finish what they play. But sometimes, we need to allow for consideration of real life. If someone has to go to the hospital for themselves or a family member, say, we should cut them some slack and not question their courage because they can't play right now.

1582 days, 12 hours, 52 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Like Slash, I just don't care about Championships.

Right now, because I don't think the achievement accumulation method of ranking and inviting players to join is valid. Even if that were to be changed to a ladder, I would still not care that much. I will probably never be invited to play, and that's fine. I know I am not serious enough about the game to put in the time and attention, game after game, that rising to the top (under either system) would require. But, why should my results for the time and attention I do earn be tarnished with any notes one way or the other? Especially based on the games I choose NOT to play?

A ladder is about recording and ranking players based on their game finishes. Nothing more. When it becomes time for the powers that be to invite players to play in some special game, what they use as a yardstick is entirely up to them. If we don't agree with it, that's a different argument for a different time.

1582 days, 12 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Making player profiles and achievements more interesting are features that a lot of games are doing. But this isn't a fast run and gun game type. This type of game doesn't need flashy graphics and innovative ladders and statistics.

Certainly, it isn't easy to make it to the championship matches. And it is certainly fun to play in one. But it isn't about that. It is about playing the game against people that are going to be committed to giving you what you want out of the game.

If you are a fan of challenge and depth, you have to put in the time to get to the people who will give you that same depth. That is what the championship games are about. People who have proven their dedication to the game and have made it their responsibility to put out their best at the cost of their time. I had to put a whole lot of other things on standby when I took on the championship match. I can honestly say I do not have that kind of time right now to give that competition on that scale.

The only way I think that people will get the game they want to play consistently is to implement some sort of auto joiner that pairs similar strength players together. But that is a way difficult thing to consider with the amount of data I'm sure would have to be sifted through. This is just the reality of it, there are way too many variables here to determine good players with an exact scale. Even a ladder system isn't exactly relevant. I might be able to beat a guy that someone better than me can't. Just because I look at things differently, or have a different weakness/strength. How in the world would that be ranked together? Especially when there won't be thousands of matches made by a specific player to get the batting average.

There isn't a best way, I think it needs to stay the way it is. Experimenting with a bunch of ideas can ruin the amount of work and time some have put in. You know, the players who are really worth their salt have put in 4 hours of studying this game for every hour they have played it.

1582 days, 12 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Missed a direct question, since I've been away:

I do believe some changes need to be made. While I have no practical objection to a ladder system, as presented, it has no application to the championship game setting. This seems a natural consequence of the stated indifference of many of the contributors here to the War. As such, I don't think it can justly be said that I would have us "go back" to an achievement system since we don't seem to have left it.

But I do believe we ought to see some modification to the official system as well.

With regard to another point: I guess several of us have stopped contributing to this discussion not because we're either in agreement or uninterested, but rather because continued repetition seems pointless. There have been many interesting points raised here and then ignored, only a few of which were mine.

1582 days, 12 hours, 0 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@mycroft: If you are happy with the way scoring is done now, you should be happy there is no indication NU management plans to change that.

Those of us that prefer a ladder ranking with the opportunity to do better or worse relative to other players can continue working to create a ladder.

What you want and what we want do not have to be mutually exclusive. Nor should the fact that no system is perfect stop some of us from seeking, if not perfection, something that we believe is somewhat better.

If we are willing to give it a try, time will tell if more players prefer one ranking system over the other.

1582 days, 11 hours, 12 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Continued:

GP: I don't believe anyone's input has been ignored. But, the thread has moved along with the majority of the posters driving the conversation.

I have yet to hear any argument on why the accumulation of achievement points based on blowing stuff up is better than a ladder based on relative finishing position. What I hear is some saying since no system can be perfect we should not change anything. And I hear some saying that while some modification might be necessary, we should still maintain the current achievement system. Please correct me if I have missed anything.

My opinion is that unless it changes your position relative to that of others, blowing stuff up is meaningless. And if it does, then you will be rewarded by a ladder tracking your game finishing position.

I am more than willing to discuss why I believe the proposed ladder, as we have worked through different inputs, would be an improvement. I suspect others here are as well. I just hope it can be something like A is better than B because...(and list the reasons) rather than simple statements of happiness or unhappiness with something.

1582 days, 10 hours, 57 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

I don't think anyone is saying don't do the ladder as a separate ranking system. Those of us with no use for it don't care if it is available for those who do want it. That is perfectly fine. We just don't want a ladder to be the only system, and I don't think anyone is really arguing for that anyway. I thought we had gotten past the "which system is better" discussion. We can discuss possible improvements that can be made to the current system and good ways to set up a ladder system, but they are separate discussions.

1582 days, 10 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Concur, Slash. Mule, I think you got it in a nutshell.

Separate discussion about fixing Achieves - when I return, possibly Saturday, I may launch such a thread if no one else has, and I'd appreciate it if you folks would contribute.

The Wars are important to a lot of us, and I think we can improve things a bit.

As for the ladder: take the time to do it right.

1581 days, 21 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Martinr

I don't think there is really any bitching re: Capricorn. It's a cutthroat affair that I'm enjoying massively.

The only one regularly "complaining" is Ted, and he does it purely for propaganda reasons, he doesn't really believe it. He is decidedly hypocritical as he has actually had Ciocco as a fully compliant vassal since turn 36, flying the cubes around to fight his battles, giving _all_ of his own fleet to him. Even made Ciocco to give up his entire home territory without a fight in exchange of servitude.

1581 days, 18 hours, 24 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Interesting take on Cap Champ game. I think it looks agonizing long and likely to get much longer. Probably more so with the level of players and all the non-alliance alliances. While that is interesting, I really am posting this note to get us back on the Scoring Ladder topic.

RE: Slash's, "I thought we had gotten past the "which system is better" discussion."

I suspect it will keep coming up from time to time. For myself, I am OK with running the two systems side by side for a while to give us a chance to judge them on results instead of theories or preconceptions. Long term, I DO hope the ladder is adopted by NU as the official system, and the "counting coup" or achievements are demoted to just ancillary data or badges. Just as I am sure there are some that would hate that.

And, before that happens we probably WILL need such a discussion on "which system is better."

1581 days, 18 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Actually, I'm kinda with @Mule here, except these things specifically:
(1) I rather hope Achieves wins out long-term, but only after it's fixed. Just a preference, though. And I very much approve of playing for medals.
(2) Capricorn is VERY fun. It's why I'm down to four games, and I'd be down to three if more real players would take on MvM games. I really don't have the time at the moment for two teaching games, one experimental setting, and one War.

Not sure where to set up an experimental ladder. I wonder if @Thin+Lizzy and @Spacesquad would see their way clear to hosting it on the Magazine, at least temporarily. I'd be willing to maintain a post if someone else would manage the actual execution.

Having said that -- still, do it right. Someone wanna set up a SurveyMonkey poll, maybe? (I know someone mentioned it and someone else said no, but that's not really a discussion. And I for one liked the idea.)

1581 days, 17 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> we probably WILL need such a discussion on "which system is better."

I don't expect that to happen for quite some time.

There are a lot of holes that would need to be filled before we're ready for that. One of the large ones has to do with player score vs. player rank. We don't even have a starting point for this, but I have a few thoughts.

I believe all players should start with a rank of Midshipmen, as they do now. After playing a non-MvM game to the end (I don't believe the ladder should track MvM games), the winner(s) should be promoted two ranks (to Sub-Lieutenant after first win) and all other finishers should be promoted one rank (to Ensign after first completion). Non-finishers should be left at Midshipmen until they have a valid score.

All MvM finishers should receive a promotion to Ensign.

This will slow down the current rampant advancement of new players. Admiral after a single game is insane, and highly indicative of the problems with the current scoring system.

1581 days, 16 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

No argument there, @W, but that's not something that a ladder can do unless it's made official. The vast echoing penetrating and incredibly loud official silence that I've heard in this topic implies that such a move isn't planned for the immediate future.

Though, to be sure, they do like surprising us.

I'd actually suggest that the low ranks continue to be fairly easy, but that they be altered somewhat as the progression continues, and that additional ranks be created within the structure in order to stretch out the promotional levels. If an MvM game were to end sooner than 80 days, perhaps that single promotion would be sufficient, but given the present structure I rather like that the early promotions occur around Turn 20 or Turn 40.

The difficulty that we face is that a metric suitable for recognition of good newbie play should not be applied equally to high-end players.

1581 days, 16 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> additional ranks be created within the structure in order to stretch out the promotional levels

The Midshipmen at Annapolis actually have four different common ranks (4th-class through 1st-class) and six specialty ranks (Midshipman Ensign through Midshipman Captain). As the specialty ranks could be confused here with real ranks, we probably shouldn't use them. Perhaps we could use the common midshipman ranks to stretch out the lower levels. The Rear Admiral (lower half) is a Commodore, but I think we're missing a Vice Admiral. This adds a total of four ranks to the list, which should help.

> I'd actually suggest that the low ranks continue to be fairly easy

While I understand, and somewhat agree with you, the one thing I do not want to do with a beginner is give them a promotion based on a provisional score, then have to take it back when their score becomes valid. I believe that this could easily lead to them leaving.

Perhaps we could use something like a "temporary field promotion" that "must be verified by HQ" for players with provisional scores. If they lose that when they get a verified score, it may be less likely to cause them to leave.

1581 days, 16 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

No offense meant to you Slash. I am simply noting from the many folks who have posted here at Nu saying they would never risk their rank for a championship game, especially for the races that have little chance of winning. Aside from having the title Admiral, it sounds like many could care less for a ladder.

So my point it that you would keep the title everyone wants, but drops on the ladder that apparently doesn't matter. I cannot see why anyone would care about the ladder if the goal isn't to play in the Championship.

1581 days, 15 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I can only speak for myself @Lord Pollax, but I would like the ladder as a means of ranking myself relative to my competition. I have no intention of ever playing in a championship game (although kudos to those that do), but I am still competitive about this game. The current ranking structure only reflects number of at bats. A ladder could potentially reflect batting average.

1581 days, 12 hours, 34 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

300+ comments and still climbing. You can't say there is no interest in a ladder. :)

Lord P says, "it sounds like many could care less for a ladder" and I think I hear a lot more saying they could care less about a Champ game and would prefer a scoring system based on a ladder. Maybe we both hear only what we want to.

I don't care if someone else wants to play in a Championship Game. Good on them! But I do take offense if anyone wants to downgrade me or anyone else because we don't care about such games and might refuse an invite if it were ever to be offered.

There is room enough here for us to have different interests without seeking to impose penalties for that. This has nothing to do with skill level or results from games played.

1581 days, 12 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, take offense. I downgrade you, Mule! :op

No, you're right; there's plenty of room for you losers to play too.

...

Sorry. I really couldn't resist. I kept trying, but the put-downs were just too easy. It was like a warp well.

1581 days, 11 hours, 49 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP: I don't mind. :)

I consider blow'em up achievement points that don't lead to a better finish pretty silly and have my chuckles at that. Rank based on that? Seriously?

I think I'll be happy with my brand of losers.

1581 days, 9 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Lower rank, yeah. For the low ranks, blow-em-up works beautifully. Upper rank? Not so much.

But I've been informed (with some huffiness) that's better discussed in a different thread. So I'll take my discussion and go home.

Probably launching it tomorrow, actually, even though I won't be home yet.

1581 days, 7 hours, 48 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Lord Pollax, you are right, many people couldn't care less about a ladder, but don't let that be confused with a desire to squelch a ladder, as long as it doesn't remove a system that works for us too. The problem I had was your curious stance that the desire to play in a Championship game had anything to do with one's courage. While that may indeed be true for a small number of people, it is most decidedly not true for most. As Mule stated, some of us simply enjoy a different type of game. I for one think that the players in those games deserve facing players who are willing to give it their all, truly enjoy that level of competition, and have the time to commit to that level of intensity for the length of the game. My enjoyment of the game decreases at those levels of competition, so it is not fair for me to take the spot from someone who would enjoy it more. There is no reason to downgrade people because they play a game for different reasons. There is plenty of room in this community for multiple types of players with respect for all. Attempts to push things in ways that limit that diversity only hurts the community as a whole. I daresay that within the confines of a specific game, we would find much commonality.

As to your question about why someone not interested in a Championship game would want a ladder, I think Furey answers that quite well. Here is yet another type of player. I don't share his interest in a ladder, even though we both share a lack of interest in a Championship game.

As I tell my wife not uncommonly, I don't have to understand why someone feels a certain way, I only have to accept that they do and work with it.

The point of all this longwindedness is that there is a lot of fruitless discussion about why would someone want to play in a certain way and which way is better, when it would be much more productive if we simply said these are the goals people want and what can we do to see as many of them done as possible. With more than 300 posts in this thread, the disparate discussions have become rather hopelessly entangled and very hard to follow.

There is much that people agree upon, but the agreements are easy to lose in hundreds of posts that meander worse like the Mississippi river.:)

1581 days, 7 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk,

> Probably launching it tomorrow

I'd like to suggest that you delay a bit. I believe that the software that can test out the algorithms for Ladder can also test them out for Achievement. I think you'll have a better chance if you can present something that's well thought out and has been demonstrated to deliver a better result.

@Slash,

> There is much that people agree upon, but the agreements are easy to lose in hundreds of posts

That's why I'm tracking most of the ideas in a document. If I've missed something, I hope someone will inform me.

1581 days, 7 hours, 31 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer,

Excellent. I remember seeing a brief mention of that somewhere, but was one of those comments I lost track of. Could you please remind me where we might find that document if it is available for viewing?

1581 days, 1 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Azzazzello: Sounds like you need to lobby harder.

Can you afford to lobby like American lobbyists?

1580 days, 17 hours, 54 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, the 300+ comments haven't come from 300+ people Mule. More like 290 comments from 5 people and 10+ from drive by folks. Lets be honest hear, we hardly have a plurality.

I like the ladder, but don't need one to know where I rank amongst players of VGAP. Most folks know where they sit, relatively speaking. I just think you guys are expecting too much from this ladder, and feel like it will REALLY rank you. It cannot.

Aside from Blitz, there are no head-to-head games. You face 10 random players normally, and every game you get into has a different combination. Comparing 10 games I have played to 10 games any of you have played is almost meaningless if trying to assign anything other than a generic general classification like "good" or "bad". I can look at your games and know if you are good. I can see if you are bad. Can I tell if you are really #1577, or should be #1382? No.

So lets not kid ourselves into thinking this ladder is all that and a bad of chips. It will be a great conversation piece and will draw some interest from players, but as far as being a useful tool for ranking individual player skill, not so much. Because my 10 games are different than yours... apples and oranges. But they are indeed fruit. : )

1580 days, 17 hours, 24 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Lord P: Every point you make is valid. Nobody (to my knowledge) has said that a ladder will be absolutely accurate as far as ranking players.
But, I do believe a ladder would be more accurate than ranking players based on how much stuff they blow up.

As on any subject, those most interested will have the most to say. No argument there. Until someone sets up some kind of voting mechanism, nobody knows what most players want. Not you and not I. But, I would welcome such a poll. Even then, we must recognize, those that vote would be those that care the most and many others will still not contribute.

All that aside, consider the following...

When most games are posted, the victory conditions are based on how many planets you take and hold at the end of the game. How can that NOT be the prime measurement?

Now, if the game's victory condition were the winner is whoever blows up the most ships and trades the most planets back and forth, then I would certainly agree that ranks should be based on those items instead. And, I have no problems with keeping track of such numbers. But, basing rank on them? Nonsense.

This seems like a very simple exercise in logic to me.

And for anyone that thinks this part of the argument is finished and we shouldn't be talking about it... obviously not.

1580 days, 17 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Because me taking away a bunch of planets from a gaggle of new players does not prove I am a better player than you taking fewer from Admiral level players. That is why it is a lousy measure in its pure form. No measure of skill, simply a number.

1580 days, 17 hours, 10 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Mule, let me throw a bit of uncertainty into your logic:) Per chance have you considered the situation thusly?

Let's look at a couple of different cases. In the first case, a player goes through 200 turns in one game, which is hard fought, with a multitude of ships being destroyed and every scrap of territory epically fought, requiring expert stratagems, until finally the player wins.

In the second case, a player goes through four games in which he easily wins all of them in 50 turns a piece because several players dropped and the remaining players were far below the player's skill level. The player has little challenge and relatively few battles of note.

Which one deserves the most recognition? Which one deserves the most points? If one goes by the number of games won and the number of planets held, clearly the second player is far superior.

Which one would you rather play?

There is a role for counting "blow up" points, I think.

1580 days, 17 hours, 5 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Good point Slash, but I tend to agree with Mule that it isn't exactly the same thing. Crystals and Privs would be disadvantaged, and it could be exploited easily.

I still think they should be counted, just separately perhaps as badges.

1580 days, 16 hours, 48 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Lord Pollax, I quite agree that it isn't exactly the same thing and that there are difficulties with counting them. I am not saying I know the absolute best way to do so, because I don't. My only point was that they should be included as a factor and should not be delegated to the dust bin. Precisely how they should be counted...well, that is a more difficult decision, one that I am more than happy to leave to people with more more experience.

1580 days, 16 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax, You said > " Comparing 10 games I have played to 10 games any of you have played is almost meaningless if trying to assign anything other than a generic general classification like "good" or "bad"." I think these things will average out. We've all had easy wins and tough ones.

I think Furey hit the nail on the head, comparing Planets to baseball. I see a player can theoretically climb up the Achievement leaderboard by just playing more games than the average player. They're getting more "At Bats" and will get more "hits" because of it. It tells me nothing about his batting average, and in fact rewards players to "Win or Not Care". A player solely interested in climbing the current board would do best by joining 8 games, focus on the 1 or 2 he might win, and just 'mail in' (or resign) the other 6 where he has little chance.

I also think these most recent posts are ignoring the fact there will likely be a difficulty factor added to the ladder calculations. A win against a bunch of new players or droppers will not help you climb compared to a win versus other top players.

1580 days, 16 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's why ladders should be viewed from various angles.

Ship captures for ranking of Crystals and Privs (obviously minus the ones that were just swapped back and forth between "friendly or hostage" players).

Tonnage and defences destroyed by other races (obviously minus the Merlin and other assortment ship that some people blow up for PBP / PP recycling).

Anyone discussed how the new ladder is going to go through all the lists and modify it for the previously suggested Admiral level players secretly teaming to harvest open to all Melee games. Or the player who destroyed hundreds of ships but only scraped in third. Or the tonnage blown up to remove the PBP / PP recyclers. Or the swapping of planets between two friendly races to farm planet captures for resources. Or the multiple account users who give everything to one account???

Or will the ladder need a large pinch of salt added. Similar to the current achievement point ladder?

1580 days, 14 hours, 13 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

If we want to continue to chart and rank based on achievements as they are now, honesty would demand we should also base who wins the games on the same measurement and get rid of the whole planets count thingee for everyone. Is that what we want to do?

Anyone can come up with situations to support whatever they like. What-ifs are easy. However, assume all situations will average out over time and / or a difficulty modifier is applied. That should handle most of those what'ifs.

Further, let us agree that under the game descriptions it SAYS the player that takes AND HOLDS the most planets until the end is the winner. And all the rest of the players are also listed by the game according their planet count at the end of the game. From most to least.

Every game ALREADY has a position ladder based on planet count. All we propose here is to capture that data and show how it averages out for all games based on a mathematical formula that will not just favor the players that play the achievement accumulation game.

1580 days, 13 hours, 56 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Mule, I want to correct something here.... I've followed the conversation passively, and you've often wrote things like, "blow'em up achievement points that don't lead to a better finish". Those statements make me think you misunderstand what is part of the "Achievement Points" a player has accumulated. (More below)

You wrote, "If we want to continue to chart and rank based on achievements as they are now, honesty would demand we should also base who wins the games on the same measurement and get rid of the whole planets count thingee for everyone. Is that what we want to do?"

That makes no sense - because the destruction element is only one SMALL PART of our achievement score currently. The majority of the achievement points earned on this site are based on the planets a player holds when the game is over, with bonuses going to those who win based on planet count. Look at some actual numbers of top players, and you'll realize that their "Destruction based Achievement Points" are a small portion of the total calculated in Achievements.

Here is mine, for the Rebels, for example.

7947 Total Achievement Points

113 from Capture (1.42%)
541 from Planet Defenes (6.81%)
1771 from Destruction (22.29%)
and
5522 from Planets (68.49%)

Thus, planetary rank at the end of game (multiplied by winning bonuses) accounts for about 70% of my point total.

---

You can complain about "blow'em up points"; but keep in perspective how many points that might actually be for the normal player, don't get biased based on a few scores who cheated the system.

1580 days, 13 hours, 56 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

p.s. I'm, as always, in favor of a ladder. I worry that you guys are really over-complicating this thing though......

1580 days, 13 hours, 23 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@J-Zan: I "get" that the current rank system isn't entirely based on achievements. It's not even the biggest part - for the few winners. That is why I keep adding "points the do not lead to a better game finish." If you destroy an enemy ship that leads you to a better finish, then I am suggesting you will be rewarded by a ladder - by a better finish.

Those points that already are counted because of a 1st > 3rd based on planet count are certainly valid. However, they are also never averaged against the number of games played. Therefore, they favor players that play the most games, rather than the players that finish more of the game they do play - better.

Nor does this address many other factors... #4 > #11 (assuming 11 players) planet count. Calling trading the same damned planet back and forth or building a Merlin - give it away - blow it up, etc. and calling those "achievements."

Now, I may be wrong. It's happened before. And I will not be upset if you want to point it out. Fire away.

1580 days, 13 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Not to put too fine a point on it, but trading a planet back and forth is a way to mine campaign resources, not achievement points.

That in no wise makes it a bad example, mind. It's an excellent example of ways in which the system can be gamed. It's just not one of the ways in which this particular aspect of the system is gamed.

Merlin mining, on the other hand, IS a method by which achievements can be gamed. It's not a great method, but it works. Plus side there is, if anyone's got the setup to make it go off, they're building other ships for combat purposes, in all likelihood, and if their enemies are Merlin-mining too, it'll balance out... more or less.

It's not a good thing, but it's not one of the world's great evils, is kinda the thing I'm aiming at here.

We could balance it in the Achieve system by tracking the mass of gifted ships and taking a fraction of that from destruction achievements; likewise, we could reduce destruction gains by 50% or more after a certain threshold. I kinda like that last.

After all, the wise general wins a war before even going out to fight. Only a very poor one actually wants to take his army and go kill the enemy army. Just ask Sun Tzu; he'll tell you.

1580 days, 13 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

PS: What I'm launching is not a ladder; apologies for the miscommunication there. It's a new thread on how to fix the current achievement system. Expect it tonight, unless someone else wants to jump in.

1580 days, 13 hours, 7 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

My apologies for my part in carrying on an argument in a thread that had been working toward details of what a ladder should look like. I'll put this on ignore for a few days and let others discuss it.

1580 days, 13 hours, 6 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Mule,

Start from the beginning, this is false:
"I "get" that the current rank system isn't entirely based on achievements. It's not even the biggest part - for the few winners."

The current rank system is (in function) based ENTIRELY on Achievement Score. Nothing else matters but Achievement Score.*

Your mistake is that you're over-simplifying Achievement Score. It DOES have stuff you don't like included in it ... but those things are the MINORITY of what the score includes. Not the majority. You're (as always) welcome to dislike it in favor of a Ladder. I prefer a ladder as well.

My post was to correct your assumptions throughout this thread that Achievement Score is about "blow'em up points." Achievement score is not ONLY or SOLELY or even MOSTLY "Blow'em up points." Doing the math, 70% of my Rebel Achievement Score and 57% of your Rebel achievement score is calculated based on the number of Rebel Planets you hold when the game is over (or in games not yet finished). Notice that even a player who doesn't "win" a lot of games with a race (You, as Rebels), does still get a majority of their points via planets. Not the other factors.

As for your other concerns.... Trading "the same damned planet back and forth" is Achievement score neutral - it's worth ZERO ACHIEVEMENT points except the SINGLE point the planet is worth when the game is over. If you and I did that, only one of us would get 1 point throughout the ENTIRE process. People do that annoying thing to farm a different thing (campaign resources.)

Blowing up a "Merlin" in staged combat WOULD create achievement points... as would repeat-capturing it.... Call those "blow'em up points." But in normal play (without someone who is cheating the system), "blow'em up" points would only be a fraction of a total achievement score.

Thus, as I stated above, ACHIEVEMENT SCORE is primarily based on the number of planets a player holds. Yes, it certainly favors players who play more games. That's another reason I like ladders.

* - There is the experience score, but that's meaningless.

1580 days, 13 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> I'll put this on ignore for a few days and let others discuss it.

But you'll join us in the new thread, I hope? It's hard to debate with someone who isn't there or won't talk.

1580 days, 7 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The only problem with the "at bats" analogy is comparing a guy's at bats with a high school team to a guy's at bats with a AAA team to another guy's at bats with the NY Yankees. The quality of the pitching matters.

Competitors quality matter. Races played matter. I am simply asking that we make this ladder a better judge of skill, considering the factors which most influence game outcomes. The 2 items I listed are perhaps the biggest outside influences, and combined with a player's own skill determines the outcomes of the vast majority of games.

1579 days, 23 hours, 39 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Lord Pollax,
I agree totally with you that the competition you beat matters. Finishing first in a newbie drop fest is hardly comparable to finishing 3rd in a game where you lose to Emork and Jobo for example :)

All I was trying to point out was that the current system gives absolutely *zero* indication of skill, merely of playing lots of games.

A ladder, even an imperfect one, should be able to reflect relative skills *better*.

Is there a means to reflect caliber of opponents in a ladder the way in which chess ladders work? Probably, but I don't know how hard that would be to code/implement. Would a ladder system be imperfect and able to be cheesed/manipulated? Probably. But would a ladder system give us a little more information than the current system? I totally believe that it would.

I 100% agree with your concerns regarding race choice and opponent skill level.

1579 days, 22 hours, 21 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I agree that there should be a greater reward for beating quality players versus beating weaker opponents. This seems possible if the average rank of the players in the game is used to adjust the points. If players at the top were worth 3-5 times the relative advancement and players at the bottom 1/3-1/5th, this would reward play in high level games and minimize rewarding clicks per minute.

1579 days, 20 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> there should be a greater reward for beating quality players versus beating weaker opponents

That's already been figured in. It's called the Average Strength Modifier. This is a direct port from the Dotman implementation (same name) and from the RCC implementation (called Difficulty Modifier there).

------------------------------

I've come up with a way to handle the Planets.nu Difficulty Modifier. This is a number that is supposed to summarize the difficulty of the game environment, including such things as mineral availability, native frequency and Stellar Cartography.

"The Planets.nu Difficulty Modifier is a single number that is supposed to represent the relative difficulty of play for different sets of game generation parameters. This score will be added to the player's score, with the last-place player receiving no bonus, the winner(s) receiving the full bonus, and the rest of the players receiving a pro-rated portion of the bonus, depending on their finishing position."

1579 days, 19 hours, 31 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> This score will be added to the player's score, with the last-place player receiving no bonus, the winner(s) receiving the full bonus, and the rest of the players receiving a pro-rated portion of the bonus, depending on their finishing position.

I just realized that this is not zero-sum. A simple change that makes it zero-sum is as follows:

"This score will be added to the player's score, with the last-place player losing the full bonus, the winner(s) receiving the full bonus, and the rest of the players receiving or losing a pro-rated portion of the bonus, depending on their finishing position."

While zero sum isn't absolutely required, it will make the system more stable in the long run. In this instance, it was simple to do, and it made sense.

The overall impact of this should be minor. I've seen difficulty modifiers in the range of 0.50 to almost 4.

1579 days, 19 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I am curious, how is this info going to gotten?

1. Many games are missing from the Finished Game page.

2. Records of drops and even starting players was not recorded on the individual game page (in the Events section).

3. Only about 70-80% of the adds and drops are recorded in the events section of any game page. For some reason, 20-30% are not recorded. It has always been that way.

So, if you are planning on using the Finished Game Page and the Individual game page events sections, you don't need to be that precise in your calculations, since you will be only be working with 60-70% of the relevant data, since the rest is missing.

Anyway, where is the info you plan to use going to be gotten from?

1579 days, 17 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

We'll do the best we can with what we have. I expect to have to get the full game details to find out when and how slots were freed and players replaced.

1579 days, 17 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

We'll do the best we can with what we have. I expect to have to get the full game details to find out when and how slots were freed and players replaced.

1579 days, 16 hours, 55 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding those games that have dropped off the radar...

1) Since the proposed ladder has both an additive value for good finishes and a subtractive value for poor finishes, it seems like having some older games unavailable should average out. Maybe even be a good thing, assuming our early "learning games" will be poorer finishes.

2) The cynic in me ponders if some of those that might find this a problem could be those players that have accumulated points, not through good average finishes, but by simply playing more games than other players. Redressing this is something I see as a good thing.

3) The alternatives to just going with the data that is available seems worse. Everyone starting over at zero or not to do anything because all the date is no longer available. Are there any other acceptable options?

1579 days, 16 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

What if the ladder was a fixed duration thing, like a rolling 2year timeframe? Or it reset each Jan 1st, with the previous year's results used for rank stuff the entire next year? That would keep it from having to constantly be updated.

Just tossing out some ideas her, not advocating any of them.

1579 days, 16 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Or everything starts from scratch.

We all start afresh and every time a new game is started the site extracts the information to a separate area (another site) for recording from the start point forward.

It does not use old data that may no longer exist.

So old games are disregarded and everything is started anew.

This will mean you can set it all up to extract all the data you want and it can disregard the data you don't want.

This will mean everyone starts all over again.

But this will be beneficial for players who are starting now. The old timers will have to start afresh but it will now be measuring their current ability. Which I think will be better for a new ladder.

1579 days, 16 hours, 5 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I would vote for the 2 year roll-over as a compromise between the two extremes and so we don't feel like ALL our previous work is of no value.

1579 days, 15 hours, 58 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

The problem of trying to track all the player changes during a game could be addressed by simply scoring only for the original players. I still think that not doing so compares apples to oranges at the end and makes things way more complicated. Drop, get killed, or resign and get the lowest finish position that hasn't already been assigned for that game. Incentivize replacement play some other way outside of the ladder.

Historically, we'd need to use finishing players since the record of starting and replacement players isn't there, but going forward I think starting players would be the way to go.

This may not even dis-incentivize replacements. I'd be far more likely to gamble on saving a position if I knew it wasn't going to hurt my ladder rank.

K

1579 days, 15 hours, 53 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I am thinking that if you check the activity you can see when players drop or join and by comparing that to the planet count on that turn, the data is available. How difficult it is to extract automatically is something I do not know, having never seen the raw data.

In the Die Hard games there is a badge for replacements, but I'm not sure how effective that is compared to the desire for a full game roster. How many turns would any of us play for a badge?

1579 days, 15 hours, 40 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Not a lot of replacement games for me, but aren't they usually a case where you are joining in middle or lower positions? Isn't that more likely to NOT help your ladder position? I DO know that you join 'blind', and many times find your abandoned race is already under attack, and in need economic management. I would be more likely to join as a replacement if I knew my final ladder score would NOT count.

In one case I joined as a Birdman replacement to get acquainted with the race before starting my first full game as a Bird. My point is why not offer these replacement positions up for players to try stuff out without pressure?

1579 days, 15 hours, 32 minutes ago

redphoenixchaos

RE: I am starting a campaign to lobby the gods of this site to...

If you want to encourage people to join replacement games though, you might want to see a rating boost for them. Nothing to unbalance anything but enough to encourage people in the behavior you wish to see in the game. I think before there was an effort through the tenacity rating to encourage people to replace dropped players. If a new rating system is being implemented, I would think it should be friendly if not encouraging for replacement players.

-Chaos

1579 days, 12 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> Regarding those games that have dropped off the radar ...

At this time, we're not sure that there are any for the interface I'll be using (going through games to determine the players, not going through the players to determine the games). I don't believe there's any need to borrow trouble. Let's see what it looks like when I load it all up.

I think I know how to locate resignations, drops and replacement players. Let's see what I can and can't get before we worry about this.

> I would vote for the 2 year roll-over

For some players, that's only 2 games.

@Kelmain,

> This may not even dis-incentivize replacements. I'd be far more likely to gamble on saving a position if I knew it wasn't going to hurt my ladder rank

That's what's currently planned. If the replacement ends up with a positive score, they get it. If they end up with a negative score, it's ignored.

@Redphoenixchaos,

> If you want to encourage people to join replacement games though, you might want to see a rating boost for them

That sounds good. Got an idea?

1579 days, 10 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I was thinking the boost for replacements being there is no negative outcomes for joining as a replacement. Even if you play badly, you still won't go less than zero points. Artificial, yes. But it would be a reward for having to join blindly and sticking it out. Drops would be discouraged because of the lack of risk.

I just witnessed a replacement guy lose 200+ advantage points joining a game, because he took over a guy in second place set up for a rapid collapse. Not his fault. He should have never had to take that beating in points. But he was a Die Hard type and suffered through it doing the best he could with the hand he was dealt. I would not want to see a repeat of that in our system.

1579 days, 8 hours, 47 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

So if I am reading this right, a player can join as many games as he/she is eligible to play as replacement players; mail in turns and never risk losing any ladder position, but can gain position for any top half finish even if they did nothing particular to earn that placement. Nice. That's certainly an incentive to sign on to lots of games.

I didn't see how the actual placements are to be calculated, but I think the actual ladder position should be based on a player's average score per game rather than some accumulation of total points. If you finish lower than your average score, you go down some. If other players finish higher than your average, you go down some. If you finish higher or other players finish lower, you go up.

1579 days, 8 hours, 39 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

My thought isn't to have them mail in turns. They need to improve their position to get more points, but won't lose any if they don't (was my way of thinking).

The odds of someone joining and doing nothing, yet retaining a top 5 spot are nil in my mind anyhow. Especially if folks are really concerned about ladder position. Everyone should be battling hard to claw up a spot or two as it means points.

1579 days, 7 hours, 51 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> I was thinking the boost for replacements being there is no negative outcomes for joining as a replacement

That's almost what's currently planned. If a replacement misses three turns and is dropped, they will receive a non-positive score for the game.

> I didn't see how the actual placements are to be calculated, but I think the actual ladder position should be based on a player's average score per game rather than some accumulation of total points.

Accumulation of points is completely valid, as it's a zero-sum (or nearly so) scoring system. For every positive point, there will also be a negative point. We're trying to keep it that way by balancing as much as possible.

In addition, there's the Average Strength Modifier, which compares each player's score to the average of the other player's scores. As an example, if an Admiral enters a game with a bunch of Lieutenants, the Lieutenants will have a lot to gain and not much to lose, while the Admiral will have a lot to lose and not much to gain.

These two items allow continuous accumulation of scores to deliver valid placement after a while (I hope it's only three games). You should read the document I referenced previously. I believe that will make everything clearer.

1579 days, 7 hours, 28 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@W I did read the document and it is very helpful, but it did not make absolutely clear (to me at least) that the score would be cumulative versus averaged. I'm far from a mathematician, but an average score would still be based on a zero sum total I believe. It's a small point though.

Overall, I am quite happy with the plans as they have come out, I do still dislike any system that rewards (or rather, doesn't penalize) lots of mediocre play and lots of player changes.

1579 days, 5 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk,
@Whisperer: Well now I know why docs aren't done yet!

1579 days, 5 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

RABBLE RABBLE RABBLE!

(Need to unsubscribe before a hundred emails while I sleep)

1578 days, 20 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Glyn,

> Well now I know why docs aren't done yet!

From http://en.wikipedia.org/wiki/Internet_troll

> troll: a person who sows discord on the Internet by starting arguments or upsetting people, by posting inflammatory, extraneous, or off-topic messages in an online community with the deliberate intent of provoking readers into an emotional response or of otherwise disrupting normal on-topic discussion.

Go away Troll. Your disruptive comments are neither required nor desired.

1578 days, 20 hours, 6 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"I just witnessed a replacement guy lose 200+ advantage points joining a game, because he took over a guy in second place set up for a rapid collapse."

I don't see how this is possible.

1. If he had 200 planets he would probably already be a solo winner or part of an alliance winner.

2. But even if not, it takes 50 turns to become fully vested. So, if he lost 5 planets the first turn, he didn't lose 5 Achievement Points, he lost 0.02 x 5 = 0.1 Achievement Point. If he loses an average of 4 planets a turn over those 50 turns until all 200 planets are gone, he would only lose 100 Achievement Points. With an empire of 200 planets, he must have had at least 100, if not 200 ships. If he couldn't kill 10,000 kt of ships (100 Achievement Points) in those 50 turns, he was playing very poorly. That is only 10-15 big carriers and alchemy ships. His star bases alone would kill that many ships. Even if he just pressed end turn each turn he would gain Achievement Points.

1578 days, 20 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> it did not make absolutely clear (to me at least) that the score would be cumulative

Thank you for that feedback. I have updated the document to improve this. I also moved some of the information around in an effort to make the document flow better.

1578 days, 20 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, technically, you don't earn any points for having your starbases kill things. At least, that's how I understand it.

1578 days, 20 hours, 1 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> you don't earn any points for having your starbases kill things

You don't, but that's not the same as losing Achievement points.

There's no way to lose Advantage points once a Campaign game has started.

1578 days, 20 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Hmmm, that would be an interesting experiment. Start a new, unregistered ID. Play only 1 Race and always be in 4 games. Only join games with established positions. The only thing you do each turn is max your planets/starbases, so each turn takes 5-15 minutes. See how high in rank you can get. You can call yourself Captain Starbase.

1578 days, 19 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"Well, technically, you don't earn any points for having your starbases kill things. At least, that's how I understand it."

You don't earn pbp's. You earn full Achievement Points.

1578 days, 19 hours, 54 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Associated links that I think are useful to post occasionally:

Dotman's proof of concept list:
http://planets.nu/_library/2014/7/nuelowtrans.html

Whisperer's Ladder Scoring System:
http://help.planets.nu/Ladder

1578 days, 18 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer, In Dotman's ladder I had a 3rd and 4th place finish and each lowered my ladder score. You mentioned earlier in this thread that such a finish would NOT lower a player's score (we were discussing replacement players and how their finish could only improve their score). Could you clarify?

1578 days, 18 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Tom+N,

Dotman used the MCC scoring, which was (1st to l11th): 80, 40, 20, 10, 5, 0, -5, -10, -20, -40 and -80

That scoring system has a VERY large impact for 11th place vs. 10th. In order to deliver results that are more indicative of the demonstrated skill difference between the early players exiting (resign, drop, eliminated or FoF), we're looking at a curve with a flatter low-end. In an 11-player game, this should leave the top 4 players with a significant positive Placement Score. In addition, there will be a positive Difficulty Modifier (score for environmental difficulty).

This leaves the Average Strength Modifier. In a competitive game, the ASMs should all be near zero. If, on the other hand, a high-rank player enters a game with a bunch of newbies, their ASM could be significantly negative. Only in circumstances such as this would a 3rd or 4th place finish net a negative score. However, if a high rank player can lose to a bunch of newbies, it could be argued that they didn't actually deserve that high rank.

I believe that the overall effect of this for players who care about their Ladder score will be that there will be less mixing of ranks in the games, which will make the games more competitive. Expect a pile of "mid officer" games to appear where neither newbies nor high ranks are allowed in.

1578 days, 18 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

FYI, right now, we're looking at scoring curves for 11-player games similar to (1st to 11th) this:

112.571, 72.914, 40.343, 14.171, -6.286, -21.714, -32.8, -40.229, -44.686, -46.857, -47.429

This will need to be verified experimentally.

1578 days, 17 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@mjs68508 Die Hard game, zero ally, our PLS70 trial. Take a look at what happened to @Battle Toast and tell me if I'm wrong. I believe I was being conservative...I think the message I got from him was closer to 300 points than 200.

The former player screwed up royally on the diplomatic front, which ended in a coalition forming for his destruction. He quit as soon as he saw the fight start to retain the points he achieved. That left whoever took over in a terrible point trap.

This is why I advocate for the major point penalty for quitting, and any points achieved should be transferred to the new player when a resignation is needed for RL events (and you go get a replacement) so in worst case the player shouldn't go negative.

1578 days, 17 hours, 2 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> This is why I advocate for the major point penalty for quitting, and any points achieved should be transferred to the new player when a resignation is needed for RL events (and you go get a replacement) so in worst case the player shouldn't go negative.

That's the way this ladder system is supposed to work, but the resignation penalty isn't all that large (see the point list above).

Your statement reminded me that players usually improve with time. I'm wondering what everyone thinks about dropping the lowest finishing score for every 10 completed games. As a bonus, drop another low finish for every 5 finishing Replacements (20 turn play minimum).

Does this improve anything?

1578 days, 16 hours, 54 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

The problem I would have with scores dropping off by number of games is that would favor those playing more games. Someone that only plays 2 or 3 games a year could take 10 years before a poor game score goes away.

I would prefer the drop off based on years (2-3?) or some combination of number of games and years.

1578 days, 16 hours, 45 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer: Your response was a troll itself. I'm sorry I find it humorous you are so distracted from responsibilities you chose to take on. I also find it humorous that you and Gnerphk are always "WOAH WOAH SLOW DOWN THE CHANGES!" while I'm waiting to be able to use the Docs to answer questions.

And it is funny... but your thin-skin is much less so. I know EXACTLY what is going on, it happens to everyone that works on a project.

1578 days, 16 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> The problem I would have with scores dropping off by number of games is that would favor those playing more games

Not necessarily. The only scores that would be eligible to be removed are for finishes, and only other finishes would allow that to happen. Most players who are in a lot of games have a large proportion of drops, resigns, eliminations and FoFs. None of these would be eligible to be removed.

Of course, this would only be the lowest negative score. Even a low positive score improves overall ladder placement.

1578 days, 16 hours, 34 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

W: A drop, dead, or ForF would never go away?

That's kind of harsh. A problem I have always had with the Tenacity scoring is it's lack of discrimination. Someone that drops on turn 5 is far different than someone that lasts until turn 50 or 90 and then drops. That they receive a ladder score at the time and in relation to other players handles this sufficiently. Let's not get back into punishment mode.

Even more different are those that get eliminated or killed. They tried and failed and should NOT be stuck with this forever if someone that merely survives is not. That just leads to hiding out and mailing in turns, or having to turn slave to survive. that's not right!

IMO, all game scores should be treated the same.

1578 days, 16 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> IMO, all game scores should be treated the same

OK. I can see this. Lowest score is lowest score. If 10 finishes earns a recovery, then it should earn a recovery of any score. How about the oldest negative score? That removes the cause of the score, and stays in line with the intent.

> A problem I have always had with the Tenacity scoring is it's lack of discrimination. Someone that drops on turn 5 is far different than someone that lasts until turn 50 or 90 and then drops.

Not any more. That's one of the few improvements in the new Tenacity system.

1578 days, 15 hours, 59 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Lord Pollax,

Thank you for specifying which game. His result (-255) is an error. He should use the Contact button at the bottom of the page to get it corrected. In fact, I will PM him now. Thank you for pointing this out. His actual Achievement for the game should be 5 (4 x 1.1 x 1.09).

1578 days, 15 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

For what it's worth, @Troll, it's usually ME that's going "Slow down; slow down". Whisperer is far more open to change, I think. My experience with complex systems has made me extremely conservative.

1578 days, 15 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk: Actually that was hilarious! I'd say to keep using it, but then I'd be tempted to use Mechnomaniacs 'Nerf' term as well. It took a year, but I have you and MJS68508 aliases finally memorized.

1578 days, 15 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

It really WAS an honest mistake. But I did think everyone -- including you -- would get a kick out of it. :o)

1578 days, 15 hours, 11 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

In an attempt to get this thread back on topic...

Whisperer: Thanks for reminding me about T-score changes. Sometimes the old gray matter fails me.

So... You will treat all game scores the same. That's god. :)

1578 days, 14 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule: "get this thread back on topic" https://www.youtube.com/watch?v=UoUEQYjYgf4

The thread topic is "vote for this", not debate. This thread went off the rails a long time ago. Seriously, Tom N pointed out the debate has been done to death in the 5th post.

I do not think Azzazzello wanted the thread to turn out like this, but keeping it near the top of Activity feed is, assuming it results in more votes and thus more attention and the desired adoption of a ladder system.

Either vote or don't, but keeping this "on topic", well then talk about voting for it.

1578 days, 14 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> You will treat all game scores the same

I put these ideas into the document as Undecided.

Replacement Player Bonus
Score Aging - Active players

1578 days, 14 hours, 44 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

W: What's the best way for us to respond / provide opinions on "Undecided?"

If it is here my input is:
Replacement scoring: If cannot lose score it's OK.
Score Aging: Track last 3 years history.

1578 days, 14 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> What's the best way for us to respond / provide opinions on "Undecided?"

Post of course. If it's a public statement, there's much less of a chance that some Troll will come along later and write that this isn't what was asked for or agreed to. This is why I'm trying to put most of the viable ideas and suggestions into the document.

> Replacement scoring: If cannot lose score it's OK.

It only removes negative scores. That's why it's called a bonus.

> Score Aging: Track last 3 years history.

According to my understanding of what you wrote, I have a the following issues with that:

1. It would remove both positive and negative scores.

2. It would make changes when players do nothing. This could have an even larger negative impact than the "Score Aging - Idle Players" entry that many seem to not like.

3. It adds complexity to the system, as the finish date for each game would have to be tracked.

4. The finish date would have to be checked against the current date, and old games removed from scores. This would require an independent task/subtask that would need to be run on a daily basis.

5. For players who play several different races, most of their scores would not have enough history to be valid vs. provisional.

6. Time does not make a player better; playing does.

1578 days, 13 hours, 47 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

re: Aging.
OK, mark me as don't care. :)

That said..

5) IMO ~80% of player skills transcend race and only about 20% are race specific. We could argue about what the percentage is and my numbers are a pure guess-estimate. Still, most actions taken during the course of a game are similar enough that this does not seem like a big issue to me.

6) This is your best point. One HOPES more play makes one a better player. More thought and attention paid to any specific game will also accomplish this, but it's not something we can easily track.

1578 days, 10 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Good grief! I am so confused! Why is it that the only ones discussing this topic so hotly seem to be the better players?

Start passing out negative points and players would jump ship as soon as they believe they were in trouble. Soon we could see games being waged between the top 6/7 players. I prefer to play against 11 players. And, when in trouble, stick around, do my best to stay alive and annoy the front runners. Don't give me negative points or move me down the ladder for being an annoyance you can't get rid of.

If a ladder means managing to stay alive until the game ends, and receiving negative points or downward movement on the ladder, I vote no.

... Can see not receiving positive points or advancement on the ladder.

1578 days, 10 hours, 2 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I never have seen the necessity of a zero sum (as many minus points awarded as plus points) formula. Nor has anyone explained adequately why the sky would fall if it were not.

I would like to see everyone has some "skin" in the game though. Enough of a minus value so they will think twice about being the first to quit but not so much anyone is afraid to play. Maybe a modifier to remove minus points based on "time served."

1578 days, 9 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I think there are a lot of players like Jusme. In fact, I believe he represents the majority. That is why I don't think Joshua will ever institute such a ladder. The one possibility is if people must opt-in, like Campaign Games.

After all, look at the original post. Azzaz wants a ladder like the Die Hard ladder. I think that would be a great idea. But, notice how many players are demanding that there be more Die Hard games so they can be on the ladder. The huge majority of players simply say, "Fine for them, not for me."

We live in the age of Warhammer and Reset buttons on video games. The mind set of most players today is accumulation (Achievement Points), not Competition (Ladder).

Nobody is going to leave the site if no official ladder is instituted. Scores, if not hundreds, will leave the site and new memberships will drop to a trickle if a ladder is instituted. Guess which way Joshua will go?

If you want the ladder to succeed:

1. Make it private with a fancy name, like "The Grognards," etc.

2. Make it opt-in.

3. Make it start from now.

4. Make the games private with an individual host. Use the Interest Section and simply label each game, Rated Game 0001, Rated Game 0002, etc.

1578 days, 9 hours, 48 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I would have no problems with an "opt-in" option or even some kind of vote or opinion poll. That is the only way we would know what "most" players would prefer or what would make them leave.

For myself, I think most players would prefer a ladder over an accumulation of points that favors maximum number of games over a ladder that tracks the best average results.

But I have as little data to support that as does MJS to support his opinion.

1578 days, 4 hours, 39 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> I never have seen the necessity of a zero sum formula. Nor has anyone explained adequately why the sky would fall if it were not.

I have added a section for that in the document. I welcome the assistance of anyone who can state the reasons for a zero-sum scoring system more clearly.

While the sky wouldn't fall if it weren't a zero-sum system, it is be obvious to me that as you move away from a zero-sum score towards the current purely additive score, the ability for Quantity to counteract Quality will increase. This is why I want to try very hard to keep the individual elements of the scoring as close to zero-sum as possible.

I'm sorry that I can't explain it any better. While I understand the math well enough to know that the above is true, I don't understand it well enough to explain it to someone who doesn't. You might want to try adding a pile of random numbers. If the range is -1 to 1, the sum for a large number of values will be near 0 (non-inflationary). If, on the other hand, the numbers are in the range of 0 to 2 (or even 0 to 1), the result will be much larger (inflationary). Inflation of scores is bad, and contributes directly to the problem we're trying to solve.

@Jusme,

> Start passing out negative points and players would jump ship as soon as they believe they were in trouble.

That will only increase the number of negative points they receive. While the results of a game may be negative, and the internal score may be negative as well (they're easier to work with if they're balanced around 0), the score that the players see will be biased, probably by 1000 (to be determined experimentally).

I expect that 3 games will be required before the score moves from Provisional to Valid.

1577 days, 23 hours, 32 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

In answer to the requirement for a zero sum score.

To begin with, it is a very bad name to use and confuses the actual purpose. As someone who has had considerable experience in the psychological effects of numbers, it is one of the worst ways of describing what is going on and will cause players to quit because the idea of giving negative scores is anathema to most people, so I propose that people interested in this stop using that term.

What is going on is better called normalizing the data. It is statistically flawed to try to compare the averages of two or more groups if the sizes of the groups are different. What normalization does is to convert absolute numbers to a relative rank so that two groups can be compared without the differences in the spread of numbers affecting the rank. The more scores in a group, the wider the range of scores will be. For instance, if one is trying to find the average height of a million people, if you take 1000 samples, the range of values you will get will be bigger than if you only take 100 on average, because you have more chances to draw samples from the more extreme ends. This is true even though the true average height has not changed.

Basically, if you assume the scores will assume a standard bell curve, with the average scores clumping in the middle with fewer and fewer people as you get away from the middle, it is fairly simple to normalize the scores to get rid of the difference in sample size.

How does this translate into the ladder system? If you assume there is a "value" describing the quality of the player, each game can be considered a sample of their playing ability. If you take the scores of each game for each player and you convert those scores to a scale from -1 to 1, then they are all on the same scale and can then be reasonably compared. Thus, you can take a player with a thousand games and compare his real playing ability to someone who has only played ten (although the player with a 1000 games will have a much more accurate score).

Why the -1 to 1 scale? Going back to that bell curve, what the scale does is put the average at zero, with each score being given as a relative distance away from that average, either above or below.

May I suggest that the scores for each game be given with a score range that is all positive, with no negative numbers. Then you can report the effect of that score on the RELATIVE position on the ladder as either a positive or negative score,. It is imperative that the position is reported as RELATIVE to the other players, not a negative score to that player. While it may not be a difference in reality, the psychological effect is enormous.

Hopefully, this helps explains the confusion of the zero sum scoring for some people. If not, I can provide a more thorough lecture:)

1577 days, 22 hours, 58 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Having said all that, I would hope that everyone would now realize that in a game such as VGAP, in which the number of games a person can realistically play is quite small, making this sort of scoring system about as accurate and reflective of true ability as attempting to shoot a quarter from a 1000 yards with a BB gun when you can't actually see the target. Not to say that the accumulation system is better, just that any system to compare players is not really going to show a truly accurate ranking of ability with any real degree of precision. The best you can really hope for is to say that the top ten players will be superior to the bottom ten. All the scores with any degree of closeness to a particular player will be essentially meaningless. With a 1000 players for instance, any score within at minimum 50 to 100 will be essentially identical in terms of actual playing ability.

1577 days, 22 hours, 39 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@MSJ 1) Whatever 2) Fine. In fact I'm all for making it "opt in" 3 & 4) No . Not enough data to make the ladder meaningful that way.

@Mule I'm surprised to see you advocating for an accumulation system given your posts about achievements. The biggest reason for a ladder is the ability to reward high quality play rather than high quantity play, IMO. The ability for players to endlessly add to their relative score by playing lots and lots of games with mediocre finishes versus playing a few very well completely undermines the whole concept to me. The actual point scores are unimportant. Make them all positive. Make them all negative. Hide them altogether. It is the relative position that a ladder is mean to reflect.

@Slash It is true that the ranks will not be precise because we aren't measuring precise things as well as the fact that our sample size is small. That hardly makes the ranks meaningless. In a given game player A may beat player B and in a subsequent game the opposite may occur and who is to say which is better? But if Player A wins over player B (or players similar to B) several times, that will shake out in the averaging.

1577 days, 22 hours, 30 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Kelman. No, really, it won't. Given the ambiguous nature of player ability and ways to measure it, combined with the low numbers of games people can actually play, it is statistically impossible to separate players according to ability with any real precision. Anyone with anything approaching a similar rank will not be able to be distinguished. If you honestly think that rank 56 is really any different than rank 89, then you don't understand the math.

1577 days, 22 hours, 21 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@Slash In a general sense, I'll have to simply disagree and leave it there. Your numbers 56 and 89 are meaningless out of context. Maybe they would be meaningful maybe not.

1577 days, 22 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I've mentioned this before, but it deserves a second look:

NO ladder system can be zero-sum. New entries have a base number, and if people die or quit the game, they take their points with them. If it's essential to create a zero-sum ladder, you must account for both.

(New entries would be given a score equal to the arithmetic mean of all non-inactive players. The scores of inactive players are distributed to the active ladder.)

I object to someone taking away my points if I die. I want those to outlast me. I want my record in this game to stand as a monument to my ability, to be honored by future generations of Planets players. ("Grampa - did you really play against Gnerphk?" "Yes, my boy. He was a real b------.")

To counter the tendency that will arise of clogging the upper reaches of the game, rather than decay I would advocate a moderate accumulative effect on a per-game basis. This will provide a relative disadvantage to infrequent players; however, since those individuals are more likely to end up with win or loss streaks due to chance, that can only improve accuracy in the upper levels.

From a practical standpoint, it would provide encouragement to all but the poorest players. The average person would see his own position(s) rise or fall based on effort rather than ability. Let's face it, folks: Not everyone is a born logistical genius, and not many people want to be; you have to be half accountant, and that's kinda boring for a lot of us.

A zero-sum ladder would deprive many of our players with the only bankable currency they have: hope.

1577 days, 22 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Slash and @Kelmain - The proposed ladder could be made meaningful for only the very top few players. The majority would toil away in the vicinity of 1000-4000.

1577 days, 21 hours, 46 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Have to disagree with you there, Gnerphk.

Any ladder system that purports to show relative ability between players has to be "zero-sum" (please see my explanatory post about why that is such a bad way to describe it).

The problem you are talking about is a different purpose than what a ladder system provides and we should be careful to not conflate purposes.

I completely agree with you about not wanting people to take away my points if I die as a record of my accomplishments, but that is better served under an accumulative system. I happen to like accumulative systems better than ladders because they provide an accounting of what I've done and is more important to me, but it is a poor way to decide who is best.

If you want to have a system that is both at all meaningful and compares player ability, you have to eliminate the variance caused by the number of games a person has played. To do that, you have to normalize the data, and that is the "zero-sum" information people are talking about. They are just talking about it in a piss-poor way (I rather expect that most people discussing this are either engineers or computer programmers, or both, groups that are not traditionally associated with explaining things well to people outside their field). Admittedly, as I have stated, the nature of this game precludes any normalization attempt to be anything other than fantastically rough, but it does at least address the problem.

If one is to have Championship games (a perfectly reasonable expectation in a site such as this), then one has to have some way of deciding who gets to play in them. If people want the very best in the game, then simply assuming that longevity equals ability is not going to get that. While there is a correlation, it is far, far from being 100%. If one wants to reward longevity and accomplishments, then the accumulation system is better. If one wants to do both, it will require combined scores between the two systems, but the two systems have to be kept separate for tallying.

That is why, along with several others, I have advocated having both systems. They each serve different purposes that are, in many ways, mutually incompatible. Because winning these games has a strong luck factor, no system is ever going to be perfect, but one does the best one can. But to do that, we have to be clear on what each system can and can't do well. A hybrid accumulative ladder system that tries to do everything will not work as well as both systems running separately.

1577 days, 21 hours, 9 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@ Gnerphk I didn't grow up in the reward everyone with a trophy generation. I realize there are different schools of thought and that rewarding effort has its place. I'm not suggesting that we do away with all rewards for participation. The whole idea of the ladder to me is to have a second system that measures not participation, but success. If you tell me people will quit if they are exposed to any form of negative reinforcement, I say bah! Wargamers, and particularly wargamers who this game appeals to, are not that soft. Some of them will strive to become better players instead of looking for the easy game in response and that is a net gain to me. The very continued existence of this game, largely unchanged is a testament, not to the casual gamer who happens to hyperlink in and give it a try, but to the player who likes the logistics and diplomacy and actually wants to be challenged. Undermine that challenge, that complexity, that fight for survival and domination aspect of the game and it will be gone with the next round of phone aps. THAT would be a real disservice to the game.

Most players may well ignore the ladder. Some diehards, however, will have another reason to keep playing and honing their skills.

1577 days, 21 hours, 5 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk I agree, the farther you get to the extreme ends, the more meaningful it becomes, if you have enough data to work with.

@Kelmain I picked those numbers at random. I have no idea how many people play on this site. If there are 4000 or more as Gnerphk implies, then there is no opinion about it. It is mathematically impossible for ranks within 50 points (and probably much more) to be at all meaningful, the margin for error is simply too big. The rank would only mean anything at the farthest ends of the scale. But those are the ones that are most important for the purposes of the ladder, yes?

1577 days, 20 hours, 52 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

Yes. I'd even go so far as to say they are only important at the top.

1577 days, 20 hours, 48 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Well, 409 comments and counting. The ladder shouldn't scare management, IMO. It could be a small link on the homepage. Easy to ignore, for those not interested. The positives have been discussed ad nauseum.

What's next? Are we still looking for a programmer willing to create a prototype?

1577 days, 19 hours, 40 minutes ago

slash

RE: I am starting a campaign to lobby the gods of this site to...

Allow me to suggest another way of ranking players for a Ladder system (this is separate from the accumulation system). First, figure out some system for scoring players, however people choose to do so. Make that scoring system normalized, or "zero-sum" as people are calling it.

Instead of reporting people's actual scores, which gets into a whole host of psychological BS, report their standard deviation from the mean, in this case, 0 because it is normalized.

Under this system, because of the many uncertainties within the scoring, everyone within one standard deviation could be reasonably assured they were roughly equal. Most players will fall between -1 and 1 SD and they can all count themselves as basically being equal on the ladder.

By the time you get into the 4 and 5 SD (aka Sigma) range, you will have precious few players. A 4 Sigma player can be assured that he is likely on par with a 3 or 5 Sigma player, at least enough that he stands a decent chance of winning against them. A 5 Sigma player can be assured he is better than a 3 Sigma player.
Each level could be associated with its own rank, possibly. You could institute games that are, say, 5 Sigma games, where you have to be at least that high to enter, or a 0 mean, 2 Sigma game, where you have to be within -2 to 2 Sigma to enter. Many possibilities here.

I would recommend that, because people's values will change over time, entrance to games is based on CURRENT scores, but that people be awarded badges or something marking the highest level they got. For instance, if someone achieved a 5 Sigma rank, they will always keep the 5 Sigma badge, even if their current rank drops below that.

Using this system will help alleviate a lot of the uncertainties in scoring that bedevils a purely linear system and makes a good mathematical reason for the different ranks.

1577 days, 19 hours, 31 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Kelmain: I am NOT advocating for a cumulative system. But I am saying there is an undesirable impact to using, or at least displaying, negative numbers in a ranking system. After a poor finish we still might slide "in relation" to other players, but psychologically that is a different thing than seeing a minus in front of our names. We are all willing to take a certain risk as long as we perceive it to not be too big a risk. How much is too much is debatable but we are dealing with humans that are not always analytical. As Slash has pointed out, our proposed numerical system only seeks an approximation of RELATIVE ranking. Sometimes we are better off not watching the sausage being made.

We can crunch the numbers any way we like, negative numbers and all, but when the machine spits out the results it will be our individual rank relative to all the other players. Someone will be #1 and someone will be at the bottom of the list. On Dotman's list I am 766. I would rather be #1, but I am also happy I am not #2500.

As for all the rest...
* When I become inactive, by whatever measure, by all means remove me and my score from the active roster.

* I define cumulative systems as anything that favors the players that may not be as skillful but have just been around collecting longer. We should risk loss of points, or at least a risk of losing relative position, when we join a game. We only disagree on HOW MUCH risk of loss. I do not see that it has to be evenly balanced with the possible gain. We want prospective players to see a greater chance of gain than loss. This will keep them playing.

1577 days, 17 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk,

> NO ladder system can be zero-sum. New entries have a base number

The base number internally will be 0, so this is not an issue. In order to make the reports more pleasing, a value (probably between 1000 and 2000) will be added to the internal scores prior to displaying them to the user.

> if people die or quit the game, they take their points with them

How so? The points still stay in the system, maintaining the "zero-sum" philosophy.

It will be possible for the Active player scores to become unbalanced. This will have to be monitored.

@Tom+N,

> Are we still looking for a programmer willing to create a prototype?

I'm currently starting this process. It is not fast, and I will not allow it to become fast at the expense of a good result. Also, I have many RL tasks that need my attention, so I can't work on this full time, or even anywhere close to it. If I had access to Dotman's code, that would probably make this process faster.

If someone else wants to step up to the task, please feel free.

------------------------------

The document was updated to reflect the types of games to be scored within the system. Also, I added some links to various web pages, including this thread. Finally, I added the Die Hard ladder system to the history section.

1577 days, 15 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I did actually find my code from last year, happy to give you a look if you like @Whisperer, but its a bit of sparsely commented cluster. It's actually several python programs I ran in order; to pull the data, then normalize the data, then finally to run through and figure the rankings. The hard part was never the computations, the hard part was determining who joined and when, etc etc.

I just looked at my email address associated with Nu here, to see if I had yours already, I don't think I do, PM me and I'll send you the code. Also, looking through that inbox, I apologize for the PMs folks have sent that I haven't responded too!

1577 days, 15 hours, 35 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Dotman,

While I haven't worked with Python yet, I probably know enough other languages to fill the gap until it gets to be complex. I'll give it a try.

Message sent.

1576 days, 22 hours, 37 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer
I am also starting to create a ranking program. I'm doing it mainly to experiment with the PlanetsNu API and some other technologies - just doing whatever feels right rather than trying to incorporate all the community feedback. It may make sense for both of us to push forward and compare results later.

@Dotman
How did you determine when players joined and resigned? The game/loadinfo API seems to only include the join turn of the final player, without even mentioning previous players who resigned or dropped. I would also be interested in seeing your previous work on this. Sending PM with my email.

1576 days, 21 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

The only way I can think of to get the internal details of a game is to use the Load All API (http://planets.nu/#/post/attention-developers-new-load-all-api). In the year+ since it was released, it has not been added to the API documentation (http://planets.nu/api-documentation).

1576 days, 7 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@whisperer Thanks. It looks like the "load all" API lists usernames on a per-turn basis, which should be sufficient to infer join/resign events. Plenty of other information in there, too.

My first test query took about 10 minutes for one game. Is that typical? There are thousands of public games, so that translates into weeks of constant querying to get all the data for ranking.

1576 days, 5 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If you're just looking for event info, try the loadevents API call, ie:
http://api.planets.nu/game/loadevents?gameid=54512

It's not 100% accurate on old games, but for the most part it works.

Load All packages up every turn file from every player for every turn. The files can be huge (partly why the download takes so long). Might be overkill here.

1576 days, 5 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> If you're just looking for event info, try the loadevents API call

Yet another undocumented API call. No wonder I didn't know about it. Are there any other "known" but undocumented API calls that we should be aware of?

1575 days, 22 hours, 37 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer
The loadevents API is used when viewing the "events" tab of a game. For example, post to http://api.planets.nu/loadevents?version=1 with the gameid post parameter set to 100282.

@BigBeefer
Is there a certain date after which the loadevents API is accurate and reliable? Some others in this thread have talked about the events tab only showing the adds/drops about 80% of the time. If that's the expected level of accuracy in general, that would be a very frustrating API to work with.

1575 days, 13 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

> The loadevents API is used when viewing the "events" tab of a game. For example, post to http://api.planets.nu/loadevents?version=1 with the gameid post parameter set to 100282.

Where did you find this documented, specifically the "version=" argument? I'm starting to do the documentation for the API, and am looking for examples of functional code that uses it, along with any existing documentation.

1575 days, 13 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If you're looking for API commands and can read some javascript, look in the APICOM section. It contains pretty much all the calls used by the client, which is most of them.

1575 days, 12 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Big+Beefer,

> If you're looking for API commands and can read some javascript, look in the APICOM section.

I found them. Now I need to figure out how to actually use them and what they return, then document it. This is almost as large a task as the game documentation, and a lot more specialized (fewer readers).

I would expect that there's internal documentation on most of these. Is there any way I could get access to that, or at least the API parser on the server (to tell me what parameters are needed/valid)? That would save a LOT of time on this task, and it would keep the "private" codebase private.

1574 days, 17 hours, 51 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I was looking at the April, 2012 ladder for most Achievement Points per Destruction Point and guess who was #1?

Joesnoffy, the current consensus leader in Leo. He had .17587 Achievement Points per destruction point. He could shoot 1 torpedo at you and your whole empire would collapse.

On the other end was humanlightningrod with 443,376 Destruction Points. He would kill you, your family, your pets, and your neighbors before he would accept victory.

1574 days, 17 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Nice. Gonna share?

1574 days, 17 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Joesnoffy, the current consensus leader in Leo. He had .17587 Achievement Points per destruction point. He could shoot 1 torpedo at you and your whole empire would collapse.

Somehow I doubt this is the case. I think that, as a Crystal player, he preferred taking ships to destroying them.

1574 days, 16 hours, 41 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Exactly. Even that tiny bit of info generated disagreement. Now just imagine the hootin' and hollerin' when you lump regular games with melee with team games with custom games made public and you have games missing. Lots and lots of bellyaching.

1574 days, 16 hours, 37 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

MJS: And your point is?

Not do anything different? I suspect there will always be disagreements, even if we changed nothing. I certainly question the system currently in place.

1574 days, 16 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I've been cordially invited to not express this opinion here, but:
I think we could do a lot with the present achievement system, simply by tuning the method by which awards are generated and then rescaling the rank system to fit. A ladder could be interesting, but we're missing the bet if we don't do both.

1574 days, 15 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Trip down memory lane. Here some ladders, lists, rankings from April, 2012. I am just showing these to show what is possible. (Note that the headings are off. I don't know how to program so I didn't get it quite right. Most headings are for 2 columns - a name and a number).

(1) Cumulative Racial Points

http://www.rainbownerd.com/nuladders/0412g.html

This list shows the cumulative points for each race at this time. It is interesting because it is very similar to the ratios used to establish the Campaign system base Advantages - Pirates start with the most and fascists start with the least (before 2X beams became standard).

Another interesting point is that the Privateers also topped the destruction point list, unlike the Crystals, who were in the middle of the Destruction Point list.

This list cannot be duplicated today with the current program I wrote. Back then we didn't have Melee, Team, or Custom games. So, there was 1 of each race in each game. Nowadays, such a list would have less meaning since you can have 6 of 1 race in a Melee game and 2 of another race. In this case you would expect the first race to score 3X as many achievement points, all else being equal. So, you would need to find out the number of Achievement Points per position player per race.

(2) Cumulative Player Achievement Points

http://www.rainbownerd.com/nuladders/0412ach.html

This list shows the cumulative number of achievement points for each player and ranks the top 100. This is similar to our current Mercenary Leaderboard, which also combines all the races a person has played. The second list shows a similar ranking, but everyone's achievement point totals are normalized for race.

For example, vepr is lower in the second list because he played mostly Privateers. The achievement points he got for playing privateers were multiplied by 68020/113949, gotten from the first chart.

Likewise, halion rises to the top of the second list because he had been playing a lot of Empire, and Empire Achievement Points were multiplied by 68020/46503.

This list cannot be generated with the current program for 2 reasons. First, it uses the first list. Second, I would have to manually go in and add back points lost in championship games.

(3) Cumulative Player Experience/Destruction Points

http://www.rainbownerd.com/nuladders/0412ED.html

This list ranks the top players in experience points and the top players in destruction points. Experience points aren't that interesting. But, by comparing a player's Achievement Points and Destruction Points, we get the last pair of columns.

The last pair of columns (joesnoffy/175.87) shows the top 100 players who need the fewest destruction points to get Achievement Points. In other words, they get more of their Achievement Points by winning games than destroying stuff (like I do - Kablooey!).

The second pair of columns (dennis/80.75) ranks players who need the least experience to get achievement. Since experience is basically based on turns played, it ranks players by who needs the fewest turns to get the most achievement - a mark of efficiency.

So, these are just a small sample of things we can do if we massage the data. So,

1. Download all the relevant API files and HTML pages.
2. Grab the data from these files/pages and put the data in a huge database offline.
3. Use the huge offline database to make any lists you want in HTML page format.
4. Put the HTML pages up somewhere and and post a link here.

1574 days, 15 hours, 31 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mjs68508,

In most of the headers in the linked pages/tables, the TH tag should contain a 'COLSPAN="2"'. They would be easier to read if the headers lined up.

Interesting information.

The process you're describing at the end of your post is similar to what I was looking at doing. The DB probably wouldn't be all that large, but it would still take some time to go through the whole thing, which would be required if the points algorithms changed.

1565 days, 14 hours, 44 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Based on this thread (http://planets.nu/#/activity/1929168), it seems that the Difficulty Modifier that the system has isn't what I thought it was. As we're already figuring in the relative skill of the opponents, we need to create a number that defines the physical complexity of the cluster that the game's in.

I think that we should be able to come up with something by multiplying several parameters together.

Mineral Availability. There are four resource values in the game status. We should multiply these together.

Native Availability. We should multiply the percentage of worlds with natives by four, then divide by the average native government level (IIRC, this is a 1, 2 or 3 for low, normal and high).

Planets near Homeworlds. There are three parameters that determine how close planets are to the homeworlds. What to do with them? More close planets -> easier game.

Players. More players makes the game both more difficult and less difficult.

Number of planets. More planets -> easier.

Map size. Bigger -> more difficult.

Perhaps it would make sense to combine the three previous parameters into a player & planet density value. Ideas?

Sphere. Off = 1.0 multiplier, on = 1.2 multiplier?

Stealth. Off = 1.0 multiplier, on = 1.1 multiplier?

Classic/Standard/Campaign. I think we can ignore these. They emphasize different skills that are used in the Championship games, but I think that this aspect can be ignored in the overall game complexity.

Stellar Cartography. We need to figure in the various settings. I don't have time right now to come up with a viable algorithm. Ideas?

Is there anything that I missed?

1565 days, 14 hours, 36 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Please file this under the "For Whatever It's Worth" file:

When I say or hear "Difficulty Modifier" I think in terms of the players in the game and NOT the resources or whatever that we are all equally striving for.

If we are considering different games with different mixes of resources or conditions, I still question this kind of comparison. It seems to me to go far beyond the original suggestion of having a ladder based on player finishing positions and not one we need.

1565 days, 14 hours, 31 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

While I understand what you mean, I believe that we should at least consider it. This is because it's much easier to implement now than later. To implement this later, we'd have to rescan ALL the games and recalculate the ladder.

If the answer is that it's not worth ever considering this, I'll accept that. If the answer is that this can be added later, I'll want to either implement it initially, or at least gather the information needed to implement it later, so that we don't need to rescan the games to gather this data.

1565 days, 14 hours, 12 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I believe there's a logical problem when considering starting positions. If everyone faces the same conditions, the relative modifier is zero.

Example: If everyone in the game was required to build a SDSF on round one and immediately recycle it, that would be a uniform handicap. It would have an equal effect on all players and, therefore, would have no impact on the game as a whole.

The only validity that a relative rank-based difficulty modifier has is that it takes into account player ability, which is NOT equally applied or balanced.

1565 days, 14 hours, 6 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

RE: GPs comment.

Huh? I think I missed the boat somehow.

I understand Whisper's position and I am content to wait for other player inputs about if they want to have a modifier based on game conditions rather than just player experience / assumed ability / ladder ranking - or whatever we might choose as a player modifier.

As far as using any form of modifier to compare players, it seems to me that is exactly what we are proposing with a ranking ladder and that discussion was left behind some time ago.

1565 days, 13 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The present difficulty modifier is intended to give beginning players a bonus if they defeat veterans, and (from the text) to give veterans a handicap in games full of beginners.

I believe it's legitimate because player ability is, by definition, unbalanced.

"Sphere", on the other hand, is uniformly applied. A Low Minerals game is no harder to win than a High Minerals game, or a large versus small, or one that requires you to wear a top hat and monocle during each turn. There is no reason to alter awards based on that sort of balanced condition unless it's race-specific.

1565 days, 13 hours, 55 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

GP: Thanks for the clarification. That's something I can agree with. :)

1565 days, 13 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"To implement this later, we'd have to rescan ALL the games and recalculate the ladder."

Don't worry about the cost of rescanning games. There are only about 2200 public, non-training games. Any time the ranking algorithm is improved, there will be no problem recalculating from scratch.

1565 days, 13 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

> There are only about 2200 public, non-training games.

I hadn't counted them yet. Thanks for the input. As long as we're not changing what we need per player, this isn't enough to matter.

We can figure out the details of this one later.

Could you please check the per-player information in the document and tell me if you think anything else is needed?

@Gnerphk,

Interesting argument. I'll have to think about it.

1565 days, 12 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"Could you please check the per-player information in the document and tell me if you think anything else is needed?"

Reposting the page we are talking about: http://help.planets.nu/Ladder

That document's per-player information roughly matches the information I'm collecting for a ladder implementation. A few small differences:

That document lists beginning and ending scores. I am tracking scores for each turn because this API provides that quickly with one call per game. As long as the cost for collecting data is low, I tend to err on the side of collecting more.

http://api.planets.nu/account/loadscores

That document lists the difficulty modifier for a game. We are calculating more accurate ratings for each player anyway. Might as well calculate our own difficulty modifier by averaging the ratings for players in the game.

Win conditions sound useful. I am not collecting those so far, and probably won't worry about them for version 1. This information was not included in any of the API calls needed to collect the rest of the information I needed.

I am collecting the list of players who are in the game at the end, and what position each of them is playing. This should be redundant, but looks like it will be needed to partially compensate for "missing events". Unfortunately, the events API is not very reliable.

1565 days, 12 hours, 33 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

RE: "That document lists the difficulty modifier for a game."

Is that current "difficulty modifier" data based on the accumulation points system or the proposed ladder? If the former, I would think a ladder would prefer using it's own metrics.

1565 days, 12 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"Alliance Win

UNDECIDED

If an alliance wins, all members of the alliance receive the top positions. The total points of these positions are averaged, and all members of the alliance receive that average."

------------------------------------------------

Wow! I just agreed to come in 4th place so our alliance could win more quickly and my partner could start another game. We let the losers come in 2nd and 3rd so they could do their happy dance. And, now I read that it is undecided whether winning should be considered a positive goal on the ladder.

Kind of invalidates the whole idea of a ladder, doesn't it?

1565 days, 12 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Win conditions sound useful. I am not collecting those so far, and probably won't worry about them for version 1. This information was not included in any of the API calls needed to collect the rest of the information I needed.

I believe it's in the results of loadgameinfo, but it's coded. It's important to note that if the win condition is Military Score, then the score we track has to be Military Score instead of planets.

1565 days, 12 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mjs68508,

That particular item is listed as "undecided" because I haven't seen enough feedback on it. It could be argued that placement should be only by score (planets/military as appropriate), or it could be argued that the Diplomacy to come in 4th and be part of the winning team is an achievement in itself, and worth a 2nd place in the ladder score.

1565 days, 12 hours, 11 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"I read that it is undecided whether winning should be considered a positive goal on the ladder."

Everyone agrees that winning should be rewarded. The details are less clear (undecided).

Some say that all players are divided between winner and loser categories. Very simple. Totally ignores the difference between the person who barely missed second place and the person who dropped on turn 10.

Some say that when two players win together and one of them has more planets, the winner with more planets has accomplished more.

Some say that a player who has five planets at the end of the game, a player who dropped, and a player who was wiped out on turn 20 have different levels of accomplishment. If so, it is not clear how to assign numbers to those levels of accomplishment.

1565 days, 11 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If winning is not considered the highest possible success in a game, then the whole ladder is invalidated.

Likewise with a Team. In a team game you have division of labor. One team member can end in 8th place and end up losing ladder points just because the team members thought that Biocides were better ships than Crystal Thunders so the team planets belonged to the borg. I think our Crystal ended up with 8-14 planets. Yet, we had so many webs, it was one of two games cited by Joshua when he instituted the minefield limit. Between that and the fact that team won the game, I have trouble understanding that any ladder with any validity at would consider that player a bottom dwelling loser.

That is like saying a pitcher for the Dodgers with a record of 30-3 and an ERA of 1.00 is a terrible baseball player because his batting average is a measly .110.

Division of labor is the mark of superior players, not losers.

1565 days, 10 hours, 40 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

A reminder... this is about a ladder that keeps track of the player's game finishes. Not what the site may want to reward them with.

Assuming we are talking about an alliance and not a team, then I think all players of an alliance should only get whatever their position would net them without the alliance. The benefit of the alliance has already been achieved in the game by having someone else to work with and everything that entails. If a player doesn't gain by such an alliance, why did he join it in the first place? If he is not playing for his own best finish under whatever the circumstances are, he is nothing but a slave/vassal and should not get to ride to glory on someone else's coattails. And we should not be encouraging someone to work for someone else's benefit anyway.

We keep thinking in the old dynamic that only #1 (or 2 and 3) matter that much. In a ladder, every position matters. We might have alliances of all sorts of position possibilities. Just how complicated do we want to make this? Get too complicated and those that oppose any ladder will win, because we will never achieve anything!

A team should average their positions, with a tiebreaker consisting of the single highest position.

To use another baseball simile, your era does not change with how many games the team wins.

1565 days, 10 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"Assuming we are talking about an alliance and not a team, then I think all players of an alliance should only get whatever their position would net them without the alliance. The benefit of the alliance has already been achieved in the game by having someone else to work with and everything that entails. If a player doesn't gain by such an alliance, why did he join it in the first place? If he is not playing for his own best finish under whatever the circumstances are, he is nothing but a slave/vassal and should not get to ride to glory on someone else's coattails. And we should not be encouraging someone to work for someone else's benefit anyway."

Winning in a diplomatic victory doesn't count as gaining? That seems very strange to me.

If the victory conditions for a game indicate that two people are able to win as an alliance, then you will likely see a lot more cooperation in a game because people do place value on that second winning slot.

ERA measures just one aspect of a player's skill. It would be like coming up with separate measures for logistic skill, combat skill, and diplomatic skill. Interesting future work, but I think that can wait for version 5. :-)

1565 days, 9 hours, 56 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

1) You assume we actually want to see more cooperation. From the other player's perspective this is not always desirable.

2) It is true that ERA is only one aspect, just as MJS's baseball example was. To use the baseball simile better we should actually count the Dodgers as a single unit (player) and maybe an alliance as a league. And which league wins more games just doesn't matter.

3) But, I think what we are discussing here are how many reward points to give each individual player for his/her finish if they are also allied. I maintain that giving them extra points for working with someone else to achieve the finish makes no sense. If anything, we should deduct points for not doing it solo. But you might argue this discourages diplomacy and i would happily agree but still support it. :)

3) Rewarding diplomacy... If a player doesn't get rewards in game play (trades, skills, help, etc.) then he shouldn't make the alliance. We shouldn't have to ALSO change his position score because of it.

1565 days, 9 hours, 22 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

(1) More or less diplomacy is deliberately tuned by choosing zero-ally or single-ally games. We are specifically talking about games in which allies and diplomatic victory have been selected by the players/host.

(3) It does seem reasonable to give more reward points for a solo victory, because it is much harder. Doing so would tend to reduce the number of "vassals on coattails" because the main winner wouldn't want to share those points.

(3) Suppose there are extra points for solo victory, but the leading player still values the 4th place player's support enough to share the victory, as in MJS's example? This victory sharing is definitely what the players/host intended by choosing a diplomatic victory game, and such sharing should be allowed. (With less credit than for a solo victory? Probably. But it is still a victory - for both players.)

@mule I'm getting the impression that you are actually opposed to the "diplomatic victory" rule set. Should we just not include such games in the rating calculations? That would make more sense to me than ignoring the rules that the players/host agreed to, but also means a lot of players won't be rated at all. Diplomatic victory is very common.

1565 days, 9 hours, 15 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I don't think you understand the concept of division of labor. In the example I gave where I came fourth as the winner, I gave a whole bunch of planets to my ally because he could build Gorbies that were much better and cheaper (10/10/10/1 star base) than I could build Vickies (10/10/10/8) star base). I am not the only one that thinks 10 fascist Gorbies + 10 fascist Vickies is a better mix than 20 fascist Vickies. I also gave him about 3 merlins to make the Gorbies, 20 Meteors to tow ships around faster, a couple of Super Freighters, and about 15,000 clans to collect taxes with. While he collected 200-300 pbp's in combat, I collected 400-500 pbp's in combat destroying our enemies. I have trouble seeing how I was riding his coattails. Any player who plays VGAP should understand that it is best that the big carrier race owns the majority of planets. First we defeated a Crystal/Lizard alliance and then we forced a Lizard/Robot alliance to surrender. So, there was no 2 vs 1 involved.

Actually, since this is private, you can do whatever you want. However, to be truthful and not trick people you should explain that the ladder doesn't involve how good people are at winning, just how good people are at building fighters and torpedoes (military games) or holding onto planets (diplomatic games).

1565 days, 8 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I don't at all object to a ladder, but I would mock and deride any ladder which purported to show player ability but discounted alliance members in a diplomatic victory.

I don't care how good someone is at fighting if they can't seem to win. I'm here to win. If I have to ally with someone who has more planets than myself in order to win, that's what I'll do.

Because I play to win.

(Or did I mention that?)

1565 days, 8 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Off-the-cuff examples:
"Ladder of Losers"
"Oh? Just how far down does your ladder go?"
"Mediocrity for the... no, not 'win'..."

1565 days, 8 hours, 42 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I think I understand very well the concept of division of labor. I also think there is a difference between a team and an alliance. In a team, I would agree wholeheartedly that the two players should exchange their skills and resources in such a way to maximize their use AND to be scored as a team.

But, an alliance is another creature entirely. It is (or should be) individuals cooperating for mutual benefit. But, still individuals.

I am all FOR diplomacy! The sum is stronger than the parts. I only take issue with scoring them differently just because they have decided to avail themselves of this benefit during game play. They already will have benefited each other and should need no "special" points awarded after the game ends.

1565 days, 8 hours, 41 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

In short, @Mule, you're fine with diplomacy as long as it doesn't gain a player anything?

1565 days, 8 hours, 41 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I agree with Gnerphk that difficulty rating should not be impacted by the game settings, i.e. mineral density, natives, etc. Those factors neither make a game harder or easier, just different. Higher level player however certainly should be factored in.

On the allied victory question, I do think allies who achieve the designated victory conditions should float to the top of the position designation for the game as they do in the achievement system. Not because they do in the achievement system, but because winning is the game objective.

K

1565 days, 8 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mjs68508,

> to be truthful and not trick people you should explain that the ladder doesn't involve how good people are at winning, just how good people are at building fighters and torpedoes (military games) or holding onto planets (diplomatic games).

I believe that this is what we're trying to decide.

To me, Mule's belief that all players are to be rewarded according to their actual place by the score decreases the meaning and intent of Alliances and Diplomacy. While formal alliances are not used in the Championship games, I have heard that there is a very high level of Diplomacy and agreements in those games.

I think that the way the winners are documented will work best, but I only see three other opinions (2 for and 1 against, if I understand correctly). As this is one of the crucial pieces of the puzzle, I'd like to see more opinions.

1565 days, 8 hours, 32 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

@G >In short, @Mule, you're fine with diplomacy as long as it doesn't gain a player anything?<

That isn't what he said. Mule implies that diplomacy should only be rewarded as far as it gets you a higher planet count. This does fail to acknowledge or at least agree with the fact that the typical win conditions are pure planet count OR alliance planet count.

1565 days, 8 hours, 32 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I have given my opinion on scoring players according to their position regardless of alliance status.

So let me ask, what is anyone's suggested alternative? And how does it stay fair to the players that are not allied? What about those that only exchange free passage or info sharing? What about undeclared alliances? What about 1-way alliance status?

Do we count a 9th place as 2nd just because he allied with 1st place? What does the real 2nd place player get?

1565 days, 8 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule: I don't believe anyone can finish better than "win".

1565 days, 8 hours, 25 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

Players who don't ally should get a solo victory bonus.

Do we count 9th place as 2nd because he allied with 1st place? IMO, yes. Deciding when and whether to ally is possibly the biggest decision anyone will make in a game and it should have ladder consequences.

1565 days, 8 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm fairly confident that, in the overwhelming majority of cases where points-stacking is not used, no 9th-place player can ally for the win. That would require 7 other players to have 50+ planets each and one player to have 200+. Tough to do in a 500-planet cluster.

You're unlikely to see anyone below 4th place ally for a win.

1565 days, 8 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"And how does it stay fair to the players that are not allied?"

I agree with you there, Mule.

Checkers - Max Allies = 1+
Chess - Max Allies = 0
Tiddlywinks - Melee with 100 turn limit

You cannot compare these three types of games, so there is absolutely no way to be fair.

-----------------------------------------

In WW2 North Africa the Italians had the most soldiers and garrisoned the most cities and towns compared to the tiny Rommel and the Afrika Korps. So, although the Italians spent most of their time running away and the Tommies were afraid of the Afrika Korps, this ladder would rank the Italians as the superior force if the Axis would have managed to take Alexandria and close the Suez Canal.

1565 days, 8 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Actually, MJ, I believe the proposed ladder would rank the Italians as the superior force regardless.

In truth, they DID ally for the win, as I recall...

1565 days, 8 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Actually, most winners come in 1st, second, or 3rd in a max ally game 1 game. Part of the reason I agreed to come in fourth was so that I could be part of a unique, elite club of commanders that was able to do that.

In fact, not only did I come in fourth, there were only 4 of us left in the game, so I came in last. Have you come in last and still won your game? Thought not. Sorry, you can't join our club. You're not good enough. No "Last Over Losers" club for you.

1565 days, 7 hours, 33 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> And how does it stay fair to the players that are not allied?

As the winners were smart and skillful enough to form a winning alliance, I don't see how it's unfair to anyone. The goal of this game is to win, not just to have the most planets. Planet count is a means to the end, but that end is a win.

> What about those that only exchange free passage or info sharing? What about undeclared alliances? What about 1-way alliance status?

What about them? I know of no way to determine the impact of these items on the overall finish. What I *do* know is that the winners won.

> Do we count a 9th place as 2nd just because he allied with 1st place?

No. We count it as a win. See below for an example.

> What does the real 2nd place player get?

3rd.

Of course, for an alliance of 1st and 9th to win would be an extreme stretch. I've seen 2nd/3rd, and even 1st/4th, and I can imagine 2nd/5th. What these players have demonstrated is that they can work well with each other, and that the lower placed player can provide something to the higher placed player that's critical to the win.

> what does all this score juggling because of alliances do to the concept of scoring balance or zero sum?

With the exception of the mention of extra points for a solo win, nothing. It works fine.

@Kelmain,

> Players who don't ally should get a solo victory bonus.

For an 11-player game, the scores mentioned ~20 days are:

1st - 103.5294118
2nd - 73.41176471
3rd - 46.43137255
4th - 22.58823529

So, a solo win would get these 103+ points, and 2nd would get 73+ points. I think that's a fairly good win bonus.

For a 2-player alliance, as currently documented, the winners would both get 88.47058826 points, and the 3rd-place player would get 46+ points.

Things get a bit tighter in a 30-player game.

@Mjs68508,

> In fact, not only did I come in fourth, there were only 4 of us left in the game, so I came in last.

No. The last-place person was the first to drop out.

1565 days, 5 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I just realized that I was more of an ass than usual here, for which I apologize. I'm slow.

In addition, I'd concede that there exists an instance where not position or win condition but rather end-game planet count alone (relative to in-cluster planets, perhaps? or perhaps not?) would have some value as a solitary ranking.

I don't see it as having value in a ladder system, but as an isolated statistic it would be of some interest, as would mean destruction scores.

1565 days, 4 hours, 38 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Winning... 1st place we can all probably agree on.

But as there can (usually) only be one 1st place, that would indicate all others are losers. OH... unless they manage to use diplomacy to ride a better player's coattails. Nonsense to that, I say!

The whole idea of keeping track of everyone's game position and transferring those results to a ladder is to give the rest of the players something to fight over - all the way to the bottom of the ranks and all the way to the end of the game.

You can call this a loser's ladder if you like, but I think it gives everyone a better reason to go on fighting even after they realize they can't take first place. And that would mean a better game for everyone. Somehow, talking someone else into carrying them across the finish line lacks a certain brag-factor to me. Maybe getting a little recognition for helping someone else finish in 1st is OK. A badge maybe? But, actually getting to count the game as if YOU took first place? REALLY? I am the only one that sees the hypocrisy of this after all the zero-sum talk?

But, I know I can't fight the tide by myself and as nobody seems to agree with me, I guess I can just sit in my corner and be sad :(

1565 days, 4 hours, 13 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"Somehow, talking someone else into carrying them across the finish line lacks a certain brag-factor to me. "

In that case all alliance victories should be considered defeats. Coming in first place with 150 planets is 5 times as easy as coming in first in a solo victory. Accepting an ally is an admission of failure in the quest for a solo victory and should be penalized points, say equal to last place, since there was the same failure there.

I have come in first with solo victories. I have come in first with alliance victories. The alliance victories were when I realized I would be unable to win the game outright. Solo victories should be worth 5X - 10X more than an alliance victory. I have more respect for a player with 1 solo victory and 4 last place finishes than I do for a player who has 5 alliance victories. The first player is certainly a better player and should be ranked higher on any legitimate ladder.

Max Ally = 0 victory is 5X more impressive than a solo victory in a max ally = 1 game, which is 5X more impressive than a victory in an alliance victory.

1565 days, 4 hours, 4 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The current Achievement system ranks them as:
6x - solo win
4x - win with help
2x - the help

I don't think those numbers are terrible, gents.

I wouldn't go so far as to say that "Ally=0" games are much harder than "Ally=1" games, and besides, it's a balanced difficulty, equally applicable to all players. But a solo win is harder to get than an allied win.

I dunno... Personally, I'd think it would be a lot easier to simply track end-of-game achieves awarded by game and adapt a sliding logarithmic scale to a curve based on the ratios. Get a best-match estimate solution using means, smooth it per-player, and make it semi-zero-sum to normalize it. Just use that for your ladder basis. Much simpler to code and there's no meaningful difference between that and the scoring system that's used to measure victories and losses, which means the ladder would be normalized to the site.

1564 days, 20 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I agree with Mule here (see...you're NOT alone). The whole point of the ladder was to show where, on average, a player finishes his games. Why are we trying to make this simple idea so complicated? I'd be interested in how many games end in solo victories, but I suspect it's under 5%. Sure, it's impressive to win without a 'declared' ally, but NOT 5x more difficult, and these wins are nicely rewarded in the other ranking system.

The games where my ally came in 3rd or 4th I would guess they would have finished lower if not for our teaming up, so they get a higher finish as a reward, and that 'losing' player deserves the "2" for the well fought effort.

Keep it simple. If multipliers are added for difficulty of competition...fine, but that's it. And PLEASE don't award a "1" to both allies. Beyond screwing up the 'zero sum' aspect, there's a difference between a team game and an alliance. Some of us love this game, but wish it didn't take 1 or 2 years to finish a game so we ALLY to just end it in 10 months instead of 15 months. Personally I'd feel guilty taking a 1st where my ally in the winning effort had 180 planets to my 70.

1564 days, 19 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Question:

If a player lasts until the end and finishes with 10 planets and another player quits on turn 50 with 40 planets and his replacement ends up with 40 planets, what is the order of their ranking of these 3 players (Finisher, dropper, and replacement) (Game lasts 100 turns, none of them wins)?

Who was the most successful of the 3 players and will be ranked highest?
Who was the least successful of the 3 players and will be ranked lowest?

1564 days, 18 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Tom+N,

> The whole point of the ladder was to show where, on average, a player finishes his games.

Actually, I thought the point is to show how skilled each player is. That should include ALL the aspects of the game.

If the players with 2nd and 3rd places in planet count have formed an alliance to win over the player with 1st place in the planet count, they've won. Ignoring the win and giving them 2nd and 3rd intentionally ignores the diplomacy and teamwork aspects of the game.

> The games where my ally came in 3rd or 4th I would guess they would have finished lower if not for our teaming up

Sometimes yes, sometimes no. It depends on circumstances which are far too complex to have software look for. Sometimes the player with the low planet count has contributed something other than planets that allowed the win. A few examples of this are Cloakers, web mines and big ships.

> PLEASE don't award a "1" to both allies.

There is no intention to do so. The current plan (being discussed here) is to split 1st and 2nd for a 2-player alliance win. Furthermore, the current plan is to split the points from those two finishes evenly. It is possible to change the split without changing the 1st/2nd position.

> Beyond screwing up the 'zero sum' aspect,

The current plans don't cause any problems with zero-sum.

> Personally I'd feel guilty taking a 1st where my ally in the winning effort had 180 planets to my 70.

In those circumstances, would you feel guilty taking 2nd?

In the above example, the combined points for 1st and 2nd are about 177. This could be awarded as any of 103.5294118/73.41176471, 88.47058826/88.47058826 or 95.09647061/81.84470590.

> Why are we trying to make this simple idea so complicated?

Because it's far easier to put complexity in early in the design than to design it in after the fact. The complexity being currently discussed is not really all that complex, and well within what can be done. The question we're currently considering is what is the BEST way to handle it. If it gets too complex to actually implement, one of the people with programming experience will mention it.

@J-Zan,

> As an "outsider" watching this conversation

There are no outsiders. The scoring will have to be applied to all players for it to have any validity.

There was some discussion previously about "opt-out", but that won't be able to be done. At most, a player might be allowed to have their score not be displayed. The score will have to be calculated.

> some of the possibilities and suggestions raised will question (for me) the value of the whole thing

At this point, the thread has become something of an idea review. Questions should arise, be written up here, and gather responses. Some of these ideas are likely to be bad, and others good. At this point, the goal of this thread should be to improve the quality of the scoring.

If someone thinks that an idea is ludicrous, please mention it. If someone has a new idea, bring it up.

@Mjs68508,

The relative position of these three players (low to high) would be Dropper, Finisher, Replacement.

1564 days, 18 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

One of the nice things about chess is that many tournaments (where ratings usually occur) have "classes." That is, players of similar ratings play together for a class trophy. For example, those rated 1000-1199 will play together, etc. There is also usually an Open section which is for everyone plus newcomers.

Since most of our games are open games, most games, especially Melee, have very few finishers. Thinking of people I have played with, the list will end up with a lot of regular players with huge negative totals, in effect, telling them they are rotten players. I wouldn't be surprised to see them leave.

The ladder should be opt-in (so should the regular officer system) so that those who just play for fun aren't officially ridiculed by a list.

"Hey dude, your rating is -2000 and it is only going to go down. Sucks to be you."

Those of us who do our due diligence and study peoples' records when we start a new game or they are a replacement don't need and won't use such a ladder. All it will do is humiliate a lot of players publicly.

1564 days, 18 hours, 42 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

And all this time I thought we were discussing a ladder that would count up and average everyone's game finishing positions!

We may hope this will somewhat equate to skill level, but we all know that this is actually impossible given all the variables. We hope the variability of any individual game will be overcome by averaging between a number of games.

If we start micromanaging every game trying to remove variables, I think we screw up the data worse than if we just report actual results.

Let's get back to the KISS principle. Letting a programmer (no offense intended) that loves complexity design a ranking system is like letting an attorney design a legal system. It ends up needlessly complex.

1564 days, 18 hours, 41 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Thinking of people I have played with, the list will end up with a lot of regular players with huge negative totals, in effect, telling them they are rotten players. I wouldn't be surprised to see them leave.

I think this will need to be checked once the first set of numbers are in. I don't think they'll be as bad as you believe.

1564 days, 18 hours, 40 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The scoring system doesn't make sense. In ALL other games that have a ladder, the scoring system is based on the ratings of the participants. They put points in the pot according to their current ranking and they receive points according to their finish.

So, actually, you shouldn't call it a ladder, since it is different than any other ladder.

1564 days, 18 hours, 29 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"And all this time I thought we were discussing a ladder that would count up and average everyone's game finishing positions! "

I like Mule's idea. Instead of falsely calling it a ladder, let's simply call it Player's Average Relative Finishing Position List. It is a nice name and accurate. Part of the repugnance of the idea of calling it a ladder is that it is not designed like a ladder and does not give accurate results.

But, if you simply call it:

"Player's Average Relative Finishing Position List"

I think there will be a lot more acceptance and approval. It even has a nice acronym, PARFPL, or PARF for short (easy to type and easy to pronounce).

1564 days, 18 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Btw, since we are using computers, it is just as easy to generate 12 lists as 1. There should be a list for each race plus one cumulative list.

1564 days, 18 hours, 19 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> In ALL other games that have a ladder, the scoring system is based on the ratings of the participants.

That's included. Please read the document. It's called the "Average Strength Modifier", and it (possibly with a different name) was used in at least two of the three previous Ladder examples for VGAP and Planets.nu.

http://help.planets.nu/Ladder

> They put points in the pot according to their current ranking and they receive points according to their finish.

Hmmmm. Interesting concept. I suggested something like that in Gnerphk's thread on improving the Achievement scoring, but the idea received no support.

1564 days, 18 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Btw, since we are using computers, it is just as easy to generate 12 lists as 1. There should be a list for each race plus one cumulative list.

I believe that's in the document as well, along with a 13th list that's JUST race.

1564 days, 18 hours, 16 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer,

"> As an "outsider" watching this conversation

There are no outsiders. The scoring will have to be applied to all players for it to have any validity."

Outsider to your conversation; I said nothing about opting out.

To state it more carefully: I'm not actively contributing to most of your discussion because I find most of inane. That's important to remember as you do an "idea review" - many of us find these ideas bad and want a simple ladder, but will not engage these detailed conversations for a variety of reasons. I might wager that the original poster fits that profile.

Also,

"> The whole point of the ladder was to show where, on average, a player finishes his games.

Actually, I thought the point is to show how skilled each player is. That should include ALL the aspects of the game."

You won't offer a quantitative metric that is usable and noncontroversial to measure this qualitative issue (skill). This is a fool's errand to try to do as precisely as you are attempting.

A simple average ranking of finishes with no other variables will do about as good a job of measuring something as slippery as "skill".

Both systems (the overly complex one) and (the simple one) will still be "sub-perfect", and I doubt that whatever precision you try to bake into the overly complex one will be much more useful than the crude, but simple metric. Especially as people try to figure out why they are ranked a certain way.

1564 days, 15 hours, 18 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If you wish to keep the conversation alive, we need a new thread with a good summary in the initial post. I suspect I'm not the only one who lacks the stamina to follow a 100+ post thread, especially with no paging.

That said, whatever system we create I think J-Zan above has a very good point, it being pretty much impossible to make a system that automatically measures skill in any meaningful way.

Which brings me to the fundamental question: Why? What are the objectives we are trying to create here?

Personally I know exactly what I want the new system to achieve:

1) Some motivation for players to keep playing even if things do not look good. At least for the start and mid-game.

2) Ensuring the senior officers games are truly senior officers games. Those who want tough competition should be able to get that without resorting to private games.

3) Ensuring that only truly competitive players get into the championship games.

Anything that does not address these three goals is of lesser importance to me.

A ladder will achieve the first one and depending on the implementation it could help determiming the second and third. But if things go too complex or risk being punitive to new players, goals 2-3 could be achieved very simply by victory requirements (win normal game for senior officers, win senior officers for champs game).

1564 days, 14 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm with you up to the last sentence or two, @SMN.

How do we address "If things... risk being punitive to new players"? Because that strikes me as of vital import.

1564 days, 10 hours, 27 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I like that Smn sets out goals. It helps to stay focused. And, like GP, I agree with most of what he says.

#1 is real high on my list too.

#2 / #3 .. I have no problem setting up or playing in custom games to get good competition. But, so far we have not had the tools to do this without doing a bunch of extra effort. Custom games allow us to use the current ranking system, but IMO that is less than satisfactory and we cannot set a minimum tenacity.

IF we have a "Player's Average Relative Finishing Position List" as MJS says, or a ladder as most say, we will still have to use password control and check player ranking manually. Because NU has not indicated any indication this might become official. Same reasoning applies to Championship games.

As far as, "if things go too complex or risk being punitive to new players" - I agree about keeping it simple.

What is "being punitive to new players?" People that join and drop very often? Yah, I'm OK with being somewhat punitive to them. But, new players are not necessarily them. If new players have a place to polish their skills before jumping into the deep end and being measured, I think we can handle that.

1564 days, 10 hours, 7 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

RE New Players: The beautiful thing about the current Difficulty Modifier is that it does this for us. Beginner games run from 0.5 up to just under 0.7 max out of a possible 3.0 score. If we weight results by the DM, we'll see something that neatly parallels the site both for beginner games and those restricted to Senior Officers. Likewise, it should automatically decrease (relatively) the numeric value of most melee games.

1563 days, 19 hours, 55 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Regarding, "Current difficulty modifier" Nice, that is, but it still exists in a system for which we are seeking to provide an alternative.

Regardomg, New Players under the proposed ladder... What if we ignore any games that have a "Maximum Rank: Sub-Lieutenant" that beginner games now have? Naturally, we ask NU not to automatically put new players into any other games and new players receive an information email that explains (among other things) the existence of our scoring system?

1563 days, 19 hours, 3 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> Regarding, New Players under the proposed ladder... What if we ignore any games that have a "Maximum Rank: Sub-Lieutenant" that beginner games now have?

Then you delay a valid score by a year or more. In addition, you will have similar inaccuracies while the score is stabilizing. I see no valid reason to not score Beginner games.

I've gone over this time and time again. Due to the relatively low number of games most players have, we can't ignore ANY games. Each and every game played that can be scored should be.

Due to the unique operation of Training and MvM games, they can't be scored in a valid manner. Beginner games should be able to be scored just fine, and will provide some initial separation between the players.

Until a player has 3-5 completed (i.e. the player survuves to the end)games, or has conquered a few players with valid scores, their score will be listed as Provisional.

1563 days, 18 hours, 57 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Whisperer: I'm OK with provisional status as you describe it. I was just trying to respond to the concerns posted here about new players. AND, to move on from GP's comment about the current system.

That said, where are we (or you as programmer?) on avoiding modifiers for things beyond strength of opposition? Have you been convinced that we don't need to micromanage for every variable?

1563 days, 18 hours, 55 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Additional...
By every variable, I guess I also mean inflating an ally's score. :)

1563 days, 17 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Gnerphk,

> beautiful thing about the current Difficulty Modifier is that it does this for us. Beginner games run from 0.5 up to just under 0.7 max out of a possible 3.0 score. If we weight results by the DM, we'll see something that neatly parallels the site both for beginner games and those restricted to Senior Officers.

That would be fine, except that the Difficulty Modifier is based on the player's rank, which is derived from Achievement. The Ladder system has a better solution called the Average Strength Modifier. This is described in the documentation, and hasn't changed (except in name) since the RCC implementation.

1563 days, 17 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> where are we on avoiding modifiers for things beyond strength of opposition?

I agreed with Gnerphk that a modifier for cluster complexity/difficulty wasn't worth the effort.

> I also mean inflating an ally's score.

Define "inflating". See my response to Tom N above.

1563 days, 17 hours, 13 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@Whisper: > Your replies >

Actually, I thought the point is to show how skilled each player is. That should include ALL the aspects of the game.

If the players with 2nd and 3rd places in planet count have formed an alliance to win over the player with 1st place in the planet count, they've won. Ignoring the win and giving them 2nd and 3rd intentionally ignores the diplomacy and teamwork aspects of the game.

> Sometimes yes, sometimes no. It depends on circumstances which are far too complex to have software look for. Sometimes the player with the low planet count has contributed something other than planets that allowed the win. A few examples of this are Cloakers, web mines and big ships.

> The current plan (being discussed here) is to split 1st and 2nd for a 2-player alliance win. Furthermore, the current plan is to split the points from those two finishes evenly. It is possible to change the split without changing the 1st/2nd position.
----

I take from this that if a 2-player full alliance wins the game as per the diplomacy or military rules, regardless of their actual individual finishing positions, the 1st and 2nd place points would be added together and divided by 2 and awarded to each player. Yes?

A 3 player alliance would add points for 1st, 2nd, 3rd / 3 and give to each player?

The same for team games?

I think I could live with that. :)

1563 days, 17 hours, 10 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Additional thought...
This isn't JUST about who wins the game.

Alliance #2 and alliance #3 that do not win the game... what formula is applied?

1563 days, 16 hours, 38 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Alliance #2 and alliance #3 that do not win the game... what formula is applied?

I hadn't thought about that. I'd be tempted to use the Team rule for them (average of the points for their earned places).

1563 days, 16 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mjs68508 said: > "And all this time I thought we were discussing a ladder that would count up and average everyone's game finishing positions! "

I like Mule's idea. Instead of falsely calling it a ladder, let's simply call it Player's Average Relative Finishing Position List. It is a nice name and accurate. Part of the repugnance of the idea of calling it a ladder is that it is not designed like a ladder and does not give accurate results.

But, if you simply call it:

"Player's Average Relative Finishing Position List"

The more I think about, why not scrap ALL this complexity? Can we safely assume NU will not be scrapping the current Achievement Leaderboard? Let THAT system handle all the extra goodies, multipliers, modifiers, etc, etc. Give us a simple contrasting list of where each player's average finish is, PERIOD. It can have an overall, per race, and per game type (Classic, Campaign, Melee, Team). In those rare occasions where a player finishes 1st in planets, but doesn't win the game...award him a "3", and the winners get a "1" and "2" respectively. KISS.

1563 days, 16 hours, 16 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

RE: > Alliance #2 and alliance #3 that do not win the game... what formula is applied?

whisperer: I hadn't thought about that. I'd be tempted to use the Team rule for them (average of the points for their earned places)...

Isn't that the same formula that would be applied to alliance #1 (winners)? - If so, I would like that.

1563 days, 12 hours, 43 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Isn't that the same formula that would be applied to alliance #1 (winners)?

Not quite. For the winning alliance, if they're not #1/#2, they're moved there. Other teams/alliances are left where they are.

Only the players who actually finish get this. The players who don't make it to the end are left where they end up.

1563 days, 12 hours, 27 minutes ago

kelmain

RE: I am starting a campaign to lobby the gods of this site to...

I'm in favor of just applying the team adjustment to the winners, and that by just floating the two to first and second place (Or 1,2,3 for 3-way alliance)and otherwise retaining the score distribution. If you try to apply the team model to non-winning positions you will simply get people breaking their alliances at the end to avoid being pulled down by a partner. This will have the effect of hiding partnerships when we look at player histories.

All of these decisions on the ladder will probably impact how people choose to ally or not for those that care about the ladder position. Retaining something that parallels the achievement system minimizes the conflicting interests between this and the "official" system.

1563 days, 12 hours, 17 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Kelmain

> If you try to apply the team model to non-winning positions you will simply get people breaking their alliances at the end to avoid being pulled down by a partner.

Good point. I had no special reason for the above suggestion (as I wrote, I hadn't thought about it), but I believe this reason is good enough to scrap the above, and leave non-winning alliances where they end up. Thank you.

For Team games, I think the "average" method should be used for all teams. The reason for this is to promote the team to work together to achieve the win, or as close as they can.

1563 days, 12 hours, 15 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Since a 2 player alliance needs 50% of the planets I am thinking that in most cases the winning alliance will both be pretty close to the top anyway. Although I guess one player could have 49% and the other only 1% and actually have a very low position. And what you are proposing is that these two players get to share the combined and averaged points of #1 and 32 and not the average of their actual finishes?

This is what I was calling "inflating" the score of the lower player. And to keep the scoring zero balanced wouldn't this also push other players into lower positions than they actually achieved?

I believe there were others that expressed opinions similar to mine on this subject. So... are we gonna get a majority opinion somehow or does the programmer just get to make a command decision?

1563 days, 11 hours, 10 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> what you are proposing is that these two players get to share the combined and averaged points of #1 and #2 and not the average of their actual finishes?

Correct. Also, any players that were above the winners in planet count would be moved down. The #1/#2 position is due to the combination of taking the planets AND using diplomacy and teamwork to win the game.

> And to keep the scoring zero balanced wouldn't this also push other players into lower positions than they actually achieved?

Define "achieved". By not having adequate diplomacy/skill/resources/abilities, the losers lost.

> are we gonna get a majority opinion

I see a majority in favor of this, with a few undecided. I'll leave it as undecided for now, and wait for more input. If I see no additional information by the time the code gets to that point, we'll run a few tests to see what looks best.

I gave three different possible sets of numbers to Tom N above. These were 1st/2nd, 2nd adjusted by planet count and average. This one is probably too close to call without running some tests.

1563 days, 11 hours, 7 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Sounds fair. :)

Naturally, I hope anyone that agrees with me will speak up on this and everyone else can just not worry about it.

1563 days, 6 hours, 20 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

There are quite a few items that are listed as "Undecided". One of the things that will have to be done once the initial data is accumulated, and the code is written, will be to run the data through the various combinations. We'll then look for the differences in player order, and try to find the one that makes the most sense, based on wins and losses.

1558 days, 14 hours, 11 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I've started work on the new rating system. So far, the effort has been almost entirely on retrieving information from the PlanetsNu API. Dotman mentioned that this had been the hardest part of his earlier efforts, but I was still surprised.

I presented this work-in-progress at a programming meetup last Wednesday. The presentation is very focused on technology (due to the audience) but you might still find it interesting.

https://youtu.be/MgdRQMNdtKA

Here is the code for this project. As I mentioned before, almost all of it is dealing with the API. Until the API issues are all sorted out, I'm starting with a very simple rating formula: wins/(games+1).

https://github.com/ericlavigne/planets_assistant/blob/master/src/clj/vgap/scores.clj

I'm also thinking of creating my own API that provides all the information needed for a rating system, so that other players can jump straight into the interesting part of creating a rating system. Let me know if anyone is interested in that.

1558 days, 2 hours, 28 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Are you able to know how many sectors a player won?

1557 days, 23 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"in the past that message there wasn't."

Even the first alpha test game, Rebirth 1, had such a message:

http://planets.nu/#/sector/815/events

However, it's true that sometimes the win message is missing. Other times the join message is missing. These are some of the issues I mentioned about difficulty with the Planets Nu API. I am still working on this problem. For now, my approach is to only include games for which I can find at least one win event and at least one join event.

1557 days, 21 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

It would be tremendously easier if admins would add, for each sector played, a flag "win" 0/1.

1557 days, 18 hours, 25 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> However, it's true that sometimes the win message is missing. Other times the join message is missing. These are some of the issues I mentioned about difficulty with the Planets Nu API. I am still working on this problem.

I believe that if any games are encountered with missing data, the entire game file will have to be downloaded, and the information extracted a turn at a time. This will be a very time (and bandwidth) consuming task, and should be delayed until later. On the other hand, the list of games with missing data should be assembled ASAP.

Examples of "missing data" are:

1. A player finishes that has no join entry in the Events log.
2. A dead slot exists that has no death entry in the Events log.
3. A player join entry with no drop/resign/death entry, and is not in the finishers list.
4. A player with a drop/resign/death entry but no join entry.
5. A missing win entry.

> I'm also thinking of creating my own API that provides all the information needed for a rating system, so that other players can jump straight into the interesting part of creating a rating system.

Interesting. I hope that you're restricting the games in the DB to just those that are listed in the document. I also hope that you're gathering at least the information discussed in the document.

1557 days, 10 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"I believe that if any games are encountered with missing data, the entire game file will have to be downloaded, and the information extracted a turn at a time."

I'll do that if necessary. Sometimes even that isn't sufficient. Try replaying game 815. You'll find that turns 1-52 are not even accessible in the time machine.

http://planets.nu/#/sector/815

"I hope that you're restricting the games in the DB to just those that are listed in the document."

Each game will have sufficient information to easily filter based on the criteria in that document: game type, number of players, etc.

"I also hope that you're gathering at least the information discussed in the document."

Yes, except for game difficulty modifier. Mule made a good case that the new rating system should not rely on an output of the old rating system. A new rating system can create its own game difficulty modifiers by averaging the ratings of players in the game.

I'll try to get an initial version of the new API running as soon as possible, so that I'm not holding anyone else up. Probably it will include a small portion of the games in the beginning, gradually adding games as I find alternative sources for the missing data.

1556 days, 4 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> Try replaying game 815. You'll find that turns 1-52 are not even accessible in the time machine.

If there were any player changes during that time, we'll probably never get good data for it. Unless there's good data, it shouldn't be included.

1355 days, 14 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Just a quick update. I recently received a message from @Ericlavigne. Work continues on creating usable raw data from the Planets.nu records.

1355 days, 13 hours, 58 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I am wondering if we had an unofficial DH type ladder, would there be some way we could use it to limit players joining custom games? If not a rank, perhaps some numerical value? Something other than the host having to research each player before giving out a password, but without being too draconian?

1355 days, 13 hours, 8 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

If its unofficial and non sanctioned by the site the I think no.

The site used player rank and tenacity to limit game access.

If its not official for the site then it will have no effect on how you join the game.

1355 days, 12 hours, 47 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"I am wondering if we had an unofficial DH type ladder, would there be some way we could use it to limit players joining custom games? ... Something other than the host having to research each player before giving out a password, but without being too draconian?"

As long as there is an easy way to do the research, I think that is not such a big problem. Imagine we have a separate website that lists unofficial ranks for all the players. The host says that the min/max ranks on that unofficial ranking are X and Y and points players to the website. Players check their own rank before asking to join. Host does a quick confirmation before sending them the join password.

1355 days, 12 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> would there be some way we could use it to limit players joining custom games?

While @Martinr's response is correct, even if it's offsite, it gives one place to look to see if the player is what you're looking for. The manual check becomes simpler.

1355 days, 12 hours, 14 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

That's solid; I approve.

Bear in mind: I've done a couple of privately hosted games and I did manual screenings with a password. It's worked fine so far.

1355 days, 12 hours, 6 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

While the raw data collection is far from complete, I thought I'd share a few details about the progress. Here are two examples of game files.

NQ-PLS-70 (2014)
https://github.com/ericlavigne/planets-rating/blob/master/test/vgap/game_examples/nq-pls-2014.txt

Madonna
https://github.com/ericlavigne/planets-rating/blob/master/test/vgap/game_examples/madonna.txt

I started with games that I had played in, because it's easier to notice mistakes in a game that I'm familiar with. The next step is an improvement in calculation of the "winners" property. Madonna was an allied victory, but the current "winners" calculation just says that the planet-leader had a solo victory. I'll fix that calculation to consider alliances, but probably continue to ignore the various alternate victory conditions for now, such as max military score.

The biggest issue remaining is to sample a larger number of games and deal with any quirks or edge cases as they come up. I have not started on that yet.

While there's still a lot of room for improvement, we're getting close to the point where someone could start to build a rating system on top of it. Imagine this level of quality, but with the number of example games expanded to a few hundred. Do we have any volunteers?

1355 days, 12 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Can anyone pull the data and start giving us useful information?

Set up a sub site and export the data for say 100 people and see what numbers we get?

I bet after when we go through the data there will be strange results that will give you similar extremities of the ranking system you have currently.

You will have games that people will say don't count so should not count to the ladder as they give a weird result as they don't fit.

Unless you start from scratch like DH with specific games with set rules and you will need 4 or 5 years to get it all enough data to start giving you good results.

But then would it be any different from the current ranking system?

If you just use the "new" ladder system it will either be as flawed as the old system as its not had enough time to settle in or the potential game differences over the past several years will lead to ladder flaws.

Or in a private game you ask for everyone to ask to join. You review that person from their forum reputation (really bad ones you don't allow as its your game). You review their race rank then compare the games they have been in and who they teamed with and how good they are. You review their tenacity and how may turns they have missed and games they have dropped. Then you decide if they are either too good or too inexperienced depending on your own preferences you allow them to join or not.

Just my thoughts of this currently theoretical ladder system and our equally flawed rank system.

1355 days, 11 hours, 32 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

> I'll fix that calculation to consider alliances, but probably continue to ignore the various alternate victory conditions for now, such as max military score.

As I know the most about the scoring system we've been trying to put together, I'm the best choice for this. As soon as the winner determination is correct, please send me the data in CSV format.

1355 days, 11 hours, 23 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Martinr,

> would it be any different from the current ranking system?

Yes. The current ranking system is purely additive. The Ladder system is "zero sum", allowing both increases AND decreases in position.

The primary difference is that a mediocre player with a large number of games can work their way up to a relatively high rank in the current system, but would end up in the middle of the pack in the proposed ladder system.

1355 days, 11 hours, 11 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I looked at Eric's link and all the numbers made my eyes cross. :) I am not sure exactly what data should be relevant and what should not, but some things occur to me.

Finishing position is paramount, certainly.

But, modified by what? Strength of competition? What about number of players (max and min) in a game? Certainly that affects difficulty. A game in which all but a few players drop early cannot be as difficult as one in which everyone sticks it out until the bitter end. Alliance modifier?

As W indicates, a ladder which provides a single sum/score/value that can move up AND down is almost imperative for this to have value! Do the numbers available to us reflect any/all of this? Is there anything other than a modified value of finishing position being considered as additive to the formula?

1355 days, 10 hours, 36 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

Most of your questions are answered in the Ladder documentation (http://planets.nu/#/howtoplay/Ladder). It's probably been a while since you read it. It might be time for you to go through it again.

The information that @Ericlavigne provided is extracted from the Planets.nu scoring system, but there's little information on what it actually is. Some of it can be determined (player information), but most of it is just numbers. Before that information is useful to anyone, it will be necessary to know what the individual numbers represent. With any luck, he's providing the information that's requested in the Ladder documentation.

1355 days, 10 hours, 34 minutes ago

tigga

RE: I am starting a campaign to lobby the gods of this site to...

Lisp? Gah.

The most naive method is simple Elo with no metadata. The game difficulty is measured as the mean Elo score of all participants. All changes are then modified based off the difference between a player's current Elo and the game difficulty. Because the mean is the difficulty, these changes should sum to zero.

The real trick would be to determine the factor by which this change is modified for winning/losing. Coming in 2nd in a single-winner game, for example, shouldn't lose you points, but dying always should. These also need to sum to zero.

There's a whole bunch of ways of doing this. I'd do this:

Every player get the following score:
5 - final position (ie. first gets 4, second gets 3, 11th gets -6)

Added to this is a bonus for the winners. Lets say +11 for a solo winner, +6/+5 for double winner. A solo winner would therefore have 16 score, dual winners would have 10/8 score for first/second. Sum of all scores would be zero.

I'd actually want to program this to figure out if I've missed something ( I think it's zero sum...) as I've only ever implemented two-player Elo before, and I've not thought about this very hard.

1355 days, 10 hours, 33 minutes ago

tigga

RE: I am starting a campaign to lobby the gods of this site to...

I totally should have read the ladder documentation as what I've suggested seems fairly similar to something on there ;)

1355 days, 9 hours, 57 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

I looked at the data for the Modanna sector, and I believe there's an error in it. The winner is listed as the Cyborg, but they have only 144 planets, which isn't enough to win. I believe there should be two winners listed (Cyborg and Privateers). Is this the issue you mentioned?

The information I was able to extract from that file is as follows:

Danneskjold played Cyborg from turn 1 to turn 167, completing with 144 planets
Karaknor played Privateers from turn 1 to turn 167, completing with 120 planets
Ericlavigne played Colonies from turn 17 to turn 167, completing with 97 planets
Gr75 played Robots from turn 163 to turn 167, completing with 54 planets
Dragon22 played Crystals from turn 139 to turn 167, completing with 45 planets
Progressor played Feds from turn 1 to turn 25, completing with 43 planets
Sabretoothphoenix played Rebels from turn 1 to turn 167, completing with 37 planets
Saarlodri played Empire from turn 1 to turn 167, completing with 1 planet
Dotman played Robots from turn 75 to turn 149, completing with 72 planets
Degotia played Crystals from turn 22 to turn 134, completing with 99 planets
Sebiroth played Lizards from turn 1 to turn 124, completing with 45 planets
Snord played Robots from turn 1 to turn 73, completing with 66 planets
Mr. Shell played Birds from turn 1 to turn 46, completing with 63 planets
Lord Tomas played Crystals from turn 1 to turn 20, completing with 31 planets
Hanoy played Fascists from turn 1 to turn 17, completing with 10 planets
Plichi played Colonies from turn 1 to turn 7, completing with 3 planets

That's sixteen players (listed in score order). The scoring system needs to accommodate that, and I think that the system being looked at will do that.

@Tigga,

We spent a LOT of time trying to put that documentation together. It's the sum knowledge of quite a few people with Programming and Systems experience, along with several experienced Planets.nu players.

1355 days, 9 hours, 52 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Vall of vext removal! (activity-feed nazee)

1355 days, 9 hours, 47 minutes ago

tigga

RE: I am starting a campaign to lobby the gods of this site to...

Huh, replacements, I hadn't thought of replacements.

Incidentally, on inactive players, decay is bad as it breaks zero sum. The solution with a zero-sum system is (IM0) to reduce the weighting of games as they age. For an active player this shouldn't have much impact as Elo naturally weights newer games more. For an inactive player they would tend toward the mean.

Maybe a game could have a half life of one year? 18 months with a root(2) stepdown at 9 months?

1355 days, 9 hours, 46 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

Eric, Whisperer, etc... Hardly a wall of text.

Yea! Let's go for it. :)

1355 days, 9 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"As I know the most about the scoring system we've been trying to put together, I'm the best choice for this. As soon as the winner determination is correct, please send me the data in CSV format."

I'm happy to convert to a different format. Since there's not a clear translation from the hierarchical data I'm producing so far to the tabular data in a CSV file, please provide details of how the file should be structured.

"The information that @Ericlavigne provided is extracted from the Planets.nu scoring system, but there's little information on what it actually is. Some of it can be determined (player information), but most of it is just numbers. Before that information is useful to anyone, it will be necessary to know what the individual numbers represent. With any luck, he's providing the information that's requested in the Ladder documentation."

I'll use the Madonna game as an example to explain each section.

There are sections to represent some of the basic details of the game. Most of these should be self-explanatory. The win-condition of 1 means diplomatic planets. There are other codes for total planets, military score, and fixed turn. I'd like to replace the numbers with words that are easier to read, but this is Nu's representation and I haven't given this part much attention yet.

:id 94061,
:name "Madonna Sector",
:short-description "Beginners",
:win-condition 1,

The winners sections indicates that the player in slot 6 (the borg) won this game. It really should have been [6, 5] because the borg and privateers won as an alliance.

:winners [6]

The bases section shows how many bases each player has on each turn. The first number on each line is the turn number. The other numbers come in pairs indicating player and number of bases. On turn 1, all 11 players had one base each. By turn 43, player 2 had 18 bases and player 4 had 0 bases.

{:bases
{1 {1 1, 2 1, 3 1, 4 1, 5 1, 6 1, 7 1, 8 1, 9 1, 10 1, 11 1},
...
43 {1 0, 2 18, 3 2, 4 0, 5 13, 6 13, 7 11, 8 11, 9 12, 10 11, 11 4},

There are similar sections called :freighters, :planets, :military-score, and :warships showing how each of these changes from one turn to another.

The slots section tells us who played in the game. I show here two of the slots, 1 and 7. Slot 1 was played by progressor from turn 1-25, and no one replaced him when he dropped. Slot 7 was more interesting, with three different players and small gaps in between those players. We can see which turns and dates each player started and finished on. In this example, slot and race numbers are the same because this is a standard 11 player game. Slots refer to position in the "players" tab of a game, and race indicates feds, lizards, birds, etc.

:slots
{1
{:players
[{:account-id 19685,
:account-name "progressor",
:end-date "2014-05-01",
:end-turn 25,
:start-date "2014-03-26",
:start-turn 1}],
:race 1},
...
7
{:players
[{:account-id 20148,
:account-name "lord tomas",
:end-date "2014-04-19",
:end-turn 20,
:start-date "2014-03-26",
:start-turn 1}
{:account-id 20912,
:account-name "degotia",
:end-date "2015-01-10",
:end-turn 134,
:start-date "2014-04-24",
:start-turn 22}
{:account-id 21727,
:account-name "dragon22",
:end-date "2015-03-27",
:end-turn 167,
:start-date "2015-01-21",
:start-turn 139}],
:race 7},

The turns section just indicates the date on which each turn occurred.

:turns
{1 {:date "2014-03-26"},
2 {:date "2014-03-26"},
3 {:date "2014-03-28"},
4 {:date "2014-03-29"},
5 {:date "2014-03-30"},
6 {:date "2014-03-31"},

All of this is more information than was requested in the wiki page, because I would want more information if I were writing the ladder implementation. The only information requested in the wiki page that is not present in these files is "ending conditions". I indicate the last turn of the game, and the last turn that each player plays, but do not indicate the difference between resigning and dropping. It's not clear to me whether that distinction is really important to anyone, and that information is only available in Nu's "events" API which I've found very unreliable.

1355 days, 9 hours, 31 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"I looked at the data for the Modanna sector, and I believe there's an error in it. The winner is listed as the Cyborg, but they have only 144 planets, which isn't enough to win. I believe there should be two winners listed (Cyborg and Privateers). Is this the issue you mentioned?"

Yes, this is the issue I mentioned. The current calculation completely ignores alliances and just declares the planet leader to be the winner. Will be fixed soon, but not tonight as I need to finish up a planets turn. :-)

1355 days, 9 hours, 3 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Glyn,

Anglais s'il vous plaît. Je sais peut peut le faire.

If you have nothing meaningful to contribute, please don't post. This is an active thread. Someone else will post something shorter soon enough.

@Tigga,

As you can see from the post times, it's been 8 months since this thread was active. The following is from memory, so it might be a bit off.

I believe we determined that games won't decay or half-life. This is because it takes so long to play them (sometimes over a year). In addition, some people like to take breaks. We determined that a player would become "inactive" with a certain amount of time off, removing them from the rankings, but they'd keep their score until they come back from their break.

@Ericlavigne,

I looked at the second file. It appears that the turn numbers you're using are for before the turn. I think it would work better if the numbers were for after the turn (zero-based - one less than the current data has).

Here are the results of that analysis:

NQ-PLS-70 - Fixed Turn
The winner is the Rebels (Yahoud)
Yahoud played the Rebels from turn 1 to turn 102, completing with 153 planets
Mule played the Crystals from turn 1 to turn 102, completing with 107 planets
Lord Pollax played the Colonies from turn 1 to turn 102, completing with 99 planets
Scotty2beam played the Privateers from turn 86 to turn 102, completing with 64 planets
Regicide played the Fascists from turn 33 to turn 102, completing with 30 planets
Battle Toast played the Lizards from turn 43 to turn 102, completing with 4 planets
Anaconda played the Birds from turn 1 to turn 102, completing with 0 planets
Macros The Black played the Feds from turn 1 to turn 100, completing with 8 planets
Incideous played the Privateers from turn 1 to turn 85, completing with 57 planets
Gloria Stitz played the Lizards from turn 1 to turn 42, completing with 73 planets
Tim Wisseman played the Robots from turn 1 to turn 41, completing with 0 planets
Kokunai played the Fascists from turn 1 to turn 31, completing with 31 planets
Ragnor played the Empire from turn 1 to turn 23, completing with 16 planets

1355 days, 8 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavagne,

After looking at the data for a bit, I was able to figure it out. I can probably accept these files, but have no need for the Bases, Freighter or Warship sections (reduce amount of data transferred).

> The only information requested in the wiki page that is not present in these files is "ending conditions".

That information should be in the Events section. Looking at the data, I don't think it matters all that much, but it would be useful if you can figure out a way to get it. Of course the most recent entry lists the winners.

1355 days, 8 hours, 42 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The NQ-PLS-70 file indicates that Yahoud has 153 planets on turn 102, which matches the scoreboard on the website.

http://planets.nu/#/sector/100282/planets

The scoreboard also shows his score increasing to 157 planets on turn 103, which my file does not show.

Unfortunately, this missing data is a reflection of missing data in Nu's load-all API. In a game like this where just one turn is missing, I think we can push forward with what we have. I've seen another example (can't remember which game) that was missing around 30 turns. Such games will need to be excluded from the rating system unless Nu admins can help with that. Fortunately, all the extreme examples I've seen were very old games that won't have much impact on current rankings.

1355 days, 8 hours, 27 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

http://planets.nu/#/sector/100282/events

1355 days, 8 hours, 0 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I'm familiar with the events API, and used it a lot the first time I tried collecting game data. Unfortunately, that API is both unreliable and difficult to use.

1) Events are randomly missing. In the particular example you linked to, Gloria resigns twice with no join in-between. I've seen other examples where a player had no join events in the entire game.

2) Events are pure text, just like what you see on that page, and the formatting of that text varies such as whether the username starts with an ampersand, whether spaces are replaced by pluses, congratulatory text at the end. It's designed for people, not computers, and it's a lot of work to deal with that.

3) Sometimes there are events like "Yahoud and open have won." because someone dropped just before winning.

4) Usernames can contain the word "and". Event boilerplate can also include the word "and". Such ambiguities can be difficult to program around.

1355 days, 6 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"After looking at the data for a bit, I was able to figure it out. I can probably accept these files, but have no need for the Bases, Freighter or Warship sections (reduce amount of data transferred)."

The larger of these two files is 70kB as text or 16 kB zipped. With 2400 games, that's about 38MB total. Unless you're on dialup, I don't see data transfer as a big issue.

These files use EDN formatting. Here's a list of parsing libraries for the format.

https://github.com/edn-format/edn/wiki/Implementations

Let me know if you have any trouble with the EDN parser for your preferred language. It's easy enough to convert to JSON which is more widely supported. I mainly prefer EDN because it allows map keys to be any type (such as integers) so I don't need to represent turn numbers as strings.

1355 days, 4 hours, 49 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

> missing data is a reflection of missing data in Nu's load-all API

I can figure something out. If a player doesn't join/leave in the missing moves, it doesn't matter. If they do, we can probably interpolate from the data we have.

> Unfortunately, that API is both unreliable and difficult to use.

OK. That just means that we treat Drops and Resignations as the same as being killed. It's not optimum, but if that's what we have, it's what we have to live with.

> Let me know if you have any trouble with the EDN parser for your preferred language.

If I can't get it to work, I should be able to write something to translate it to what I need.

> With 2400 games, that's about 38MB total.

Actually, I was thinking about the time to parse at your end and mine. Data size in this area doesn't matter.

1354 days, 7 hours, 30 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

I have been considering @Ericlavigne's information that it is not possible to always differentiate between drops, resignations, FoFs and kills. This changes the sections of the Ladder Implementation Plan (http://planets.nu/#/howtoplay/Ladder) that describe how Eliminations, FoFs, resignations and drops will be handled.

The closest that I can come up with is that, when multiple players exit in the same turn, they will be sorted by planet count. Each of these players, starting with the one with the lowest planet count, will be placed into the lowest available score slot. In the case of two players leaving in the same turn with the same number of planets, they will share the same slot.

Is this acceptable?

1354 days, 7 hours, 22 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

If this is an "open" question: That works for me.
---

I'm just happy you are still willing to keep working on it.

1354 days, 5 hours, 46 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

> If this is an "open" question: That works for me.

It was. Thank you.

Considering the FUBAR state of the current score/rank system is, and the fact that there's no improvement in sight, something has to be done. As Joshua et. al. don't appear to be willing to do anything, we must.

1353 days, 20 hours, 34 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

"The closest that I can come up with is that, when multiple players exit in the same turn, they will be sorted by planet count... Is this acceptable?"

Whatever you decide is acceptable, especially for a minimal first version. I'll give my opinion anyway about the direction I would take this.

If a player "leaves" on turn 50 with just 15 planets, I couldn't care less whether it was FoF, resignation, or dropping. I'm far more interested in the result they achieved over those 50 turns. If two previous players were wiped out, then this person did better than those two but worse than everyone else.

If a player leaves on turn 50 with 100 planets, and finds a suitable replacement that goes on to win the game, then that player did very well. They and their replacement share credit for a victory.

If a player leaves on turn 50 with just 15 planets, and an amazing replacement manages to win the game anyway... The original player should still be near the bottom of the rankings, while the replacement receives more than the usual credit for victory because they had to overcome a difficult starting position.

How to accomplish this? We need a way to estimate, for a given turn and player, the odds that the player will go on to win the game. For a simple version 1, I would go with:

(victory odds) =
(max allies + 1) *
(player's owned planets) / (total owned planets)

This is also the equation I would modify later to take game difficulty into account, perhaps like this:

(victory odds) =
(max allies + 1) *
((player's planets) * (player's skill)) /
sum of (planets * skill) across all players

A player's credit for the victory is the change in odds during the time they played, divided by the portion of the game that they played. If the player lineup changes during the game, then everyone's credit needs to be calculated separately before and after the change, so that no one is "blamed" for the fact that a stronger player showed up.

Due to inaccuracies in our calculation of odds, a player might decide to abandon a losing position when their number of planets does not yet reflect their hopeless situation. Therefore, when calculating the change in their odds during their time in game, I would replace (odds at the end) with min(odds at the end, odds 20 turns after end). This also takes care of penalizing players who leave without finding a suitable replacement.

This is also why I'm interested in other scoreboard properties besides planets. Even if final victory is based on planets, bases and warships tend to improve the odds of getting or keeping those planets later. For later versions of the ranking system, I would improve the victory odds calculation by considering more subtle factors such as how well defended a player's territory is, how many enemy ship positions they are aware of, whether they have acquired cloakers, lokis, and star destroyers...

I know this is too much for a first version. But I like to plan ahead to more advanced versions to ensure that the work I do today will put me in a good position for the work I'm planning later.

1353 days, 20 hours, 6 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

The Ladder document has been updated with the above information. If anyone has issues with it, please post it here, or PM me.

http://planets.nu/#/howtoplay/Ladder - Player Early Exits

1353 days, 19 hours, 58 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Vote added. I lack the time to read the whole thread, but like the ladder Azzazello asked support foridea. If this a thing of the past and you are discussing completely other issues let me know or just ignore this post.

1353 days, 19 hours, 50 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Ericlavigne,

> We need a way to estimate, for a given turn and player, the odds that the player will go on to win the game.

That's not something I'm going to either design or code. The reason for that is that I don't believe the players, as a group, will ever agree that any such algorithm is close enough to be used to score their games when they leave early. In addition, I believe your formulae are oversimplified. The number of starbases, warships, military score and total ship slots all contribute to the chances of winning.

On the other hand, leaving a player where they drop is an incentive to complete the game. Look at the NQ-PLS-70 game above for an example. Macros The Black left a turn early with 8 planets, and is placed below both Anaconda and Battle Toast, who finished the game with 0 and 4 planets. While that may not be completely fair, it's something that everyone can understand and accept.

I will leave this estimate for the next phase of the project, which will provide you with time to refine your data, based on actual game results. I'll add it to the document as a possible future enhancement, but I won't have time to do this for a few days.

1353 days, 17 hours, 57 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

I think we should get too sidetracked by replacement issues.

If someone needs to drop and gets a good replacement, he get's an "attaboy." But, I don't think how well the replacement does should impact the dropper either way. His score should be based only on what he did before he dropped.

The replacement can be judged by his position relative to what he took over, just as I think it is done now. It's not a perfect solution in all cases, but the replacement player knows what he is getting into and can make his own choice.

1353 days, 16 hours, 53 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

@Mule,

While I like your idea about replacement players, it's beyond what is currently planned to be implemented. I think it would be relatively simple to implement, but i'd want to see some feedback from others before I change the plan.

1353 days, 16 hours, 41 minutes ago

mule

RE: I am starting a campaign to lobby the gods of this site to...

@Whisperer: I reviewed the http://planets.nu/#/howtoplay/Ladder - Player Early Exits and can find no problem with any of that. I am much more concerned with the process being delayed or stopped by trying to over complicate things or get sidetracked by minor issues.

I will keep calm and let you carry on. :)

1352 days, 8 hours, 21 minutes ago

RE: I am starting a campaign to lobby the gods of this site to...

Another question for the community:

Would it be acceptable to ignore any game with the string "begin" in it's name? This would ignore the fake beginner games that some players have recently created to farm for points.