"One alternative is to make each player put down a deposit for the game, so that going into CD looses the bet + deposit, as Alderian mentioned"

Would the deposits cause some people to drop below 100? I mean it would otherwise not deter new players from CD'ing.
My idea is that just something simple would work. Like: a score x/10, where x is the number of times that a player went into CD in the last 10 games he played. A player with a score of 6/10 would probably not be a good choice for a reliable game.
My point is more that it does not have to be some penalty reflected in a rating (GR or points), but rather some public display of 'trustwortyness'.
Be it by CD count or left games count, it would work very well to bring down the number of CDs i'd say. Especially when it's clear to everyone that 'just going into CD to leave' is frowned upon and i'ts very clearly publicly displayed in some way (as bad thing).

But we're going horribly offtopic with this ofcourse...

I'm not saying you can't figure stuff like this out now: you can check out their player page in detail (looking at games). But i'ts not easy to see how many times a player CDs, if it was recently or not, etc.

@sqrg - your suggestion would just get people to start new accounts.

Perhaps we should look at a more typical game than the following one:

"To give you an idea of G-Rating and rewards, if, at the point of the last Ghost Rating, MM had played numbers 2-7 in a game, these would have been the changes in ratings he’d have for various results:
Win: +191.669
2-Way: +73.9294
3-Way: +34.68286
4-Way: +15.05958
5-Way: +3.285619
6-Way: -4.56369
7-Way: -10.1703
Defeat: -43.8102
You should note that it is not linear from 0, and that 3 3-ways are not worth a win, indeed, with the reduction in deltas that comes from gaining rating, you gain more from one solo than you do from 6 three ways."

Let's say there's a game with the following ranked players: #25, #50, #75, #100, #125, #150 & #175. Let's say #100 GR ranked wins. Then, based on TGM math, we get:

Win: +98.7
2-Way: +41.6
3-Way: +22.5
4-Way: +13.0
5-Way: +7.3
6-Way: +3.5
7-Way: +0.8
Defeat: -15.5

A solo in Ghost's first scenario is worth 5.5 three-way draws, but in a more typical scenario, the second one that I noted, the solo is only 4.4 three-ways. Well, I suppose that is relatively similar, really.

Anyway, sounds like the consensus here is that the GR relatively accurately awards WTA solos as well as WTA draws, which is the question I was wondering about, so thanks all for your comments.

Reasons why I do not like the GR:

1) people have used it to determine their alliances [however, that likely happens through other forms of notoriety in FtF tournaments as well]

2) I feel that it is inaccurate. Chess is a game of 1v1. Barring some major life changing event a rank 2400 player does not lose to a rank 1200 player. Diplomacy is a game of 1v6. The ELO system of Chess does not apply. If we put the rank 500 - 505 players in a game with the rank 1 player, MM, and told them they get a dollar for every time they solo, but the 6 of them get a dollar every time MM loses, MM would never win. Yet he is ranked significantly better than they are.

3) More in accuracy. Chess is Chess. The same board, the same rules, the world over. This is not true of Diplomacy. There are variants. No, I am not talking about other maps, yet. Let's just talk standard map. There is the official variant, fleet Rome. There are numerous forms of press. There are the shift left and shift right variants. There is blind variant. And now there are the MULTITUDE of map variations (I thought is was at least 50). So, does playing one form of the game make you good at playing another form of the game? They all use one GR, but I would argue that being good at one does not make you good at another.

4) Tournaments warp the GR. They do this by changing the meaning of the word "winning." Some of the tournaments of this site promote meta-gaming, some allow for it, and some do not. Plus the "point" structure of the tournament affects whether or not solo'ing in a game is better than some other out come.

I agree that most forms of ranking can be manipulated. It only gets worse in our ever changing modernism. For example, you could play an anonymous game, but if you know who is in the game, not what countries they have, it is easy enough to PM them on this site. Or, just have your friend sign up for the same anonymous game at the same time.

Segue...if the true goal is to prevent players from going CD, I suggest a system similar to what the old Judges had. A player earns 1 point for each game he plays. He loses 10 for each game he CD's in. When making games there would be a mechanic for setting a minimum player rating to join. Admittedly, it would be less about how good a player is than about how reliable a player is, but it seems obvious to me that how good a player is is rather subjective while his reliability can be determined objectively.

1) I think this has been getting better, although I do think there has been some protection going on, with the 'min GR' games.
2) Agreed. Is also a reason for the protection noted above - totally understandable when one is set to loose so much.
3) I was lead to believe marked variants (ie ones that were definitely variants and could be identified from the database) were treated differently.
4) Don't even get what you're on about here?

Manipulated: Totally agree, but those cheating cases we just have to hope don't occur, and trust people.

anti-cd: Agreed that would be a better limit on CD, but the issue with CD is normally with new accounts, and in practice any system like this would just lead to people dumping the account to loose the bad reputation.

Now, despite all these comments, I am very suspicious of GR as well, and against it being used as religiously as it is currently. However, it does seem to be accurate in general, and so its judicious use is surely a good thing

@Figlesquidge

3) My understanding was that they were treated the same but that the formula was adjusted to account for the number of players as appropriate. Are you saying that there are separate GR's for the variants?

4) The GFDT is about advancing as far as you can while accumulating points. Meta-gaming is allowed and encouraged. So, you and I decide to split the first game. We draw. We get GR points based upon a 2-way draw. But, anyone else in the game who was playing to "win" would stand no chance against us (especially if we had decided on a 3-way with another player). So, the tournament changed what "winning" due to the fact that it is no longer about winning a game, but winning the tournament, and the rules are not the same.

This was one of the reason I came up with the TMG Masters to help eliminate the meta-gaming from tournaments which I found in the others I participated in.

Sorry for any typo's. Apparently, the option to edit them out does not exist, or have I just over looked it?

TMG, variants only count 50% as much or something like that. Possibly 25% for some variants. Variants includes nopress and public press. This was changed either a month or two months ago I think and is retroactive, again qualified with "I think".

I've joined some games where ghost ratings were used as a selection criteria. But I have never used ghost ratings (or points) to decide who to ally with or against. There are pros and cons to allying with or against "better" players.

No way to edit. You can only post afterward a clarification if necessary.

But on a diplomacy website it only makes sense that you can't edit your posts. You gotta live with what you put out there.

TMG, thanks for the in-depth response! Here are my thoughts:

1. Yeah, some do, but it happens with points too and isn't really avoidable.
2. Ranking diplomacy players is a daunting task that is impossible to do 100% accurately. I think the concept of the GR is pretty damn cool and does a MUCH better job than anything else out there and I'm very impressed with TGM for putting it all together, VERY impressed.
3. I think GR should just be for the classic game and not the World Map or the Ancient Med, and I do like the fact that gunboat and global press are weighted less.
4. I totally agree on the tournaments. If you play in a League with four or five games, it's pretty damn tough to not suffer at least one loss in those games solely based on (allowed) metagaming and as long as flashman isn't in your league the same goes for winning one of the first few games in league play! ;-)

I've never even considered PMing someone that I know is in one of my anon games and assume nobody else does that, but that's an interesting point, though I'd hope the person getting the PM would report it on the global tab so the group could decide to cancel the game.

So, TMG, where are you now going to get your diplomacy fix and how will these issues you have be solved/different? More to the point of this thread, do you think solos should be weighted more heavily or are you saying that GR is so broken that it can't be tweaked a little bit to make it better? Again, I think it's brilliant to base a ranking on the level of competition even if diplomacy is much different than chess. Brilliant. I get your point that diplomacy is different than chess, but that doesn't mean the same concept for ranking is invalid. Regardless of how you rank people that play diplomacy, the fact will remain that a bunch of lower ranked people will ALWAYS be able to take out a higher ranked person, I don't see why that rules out a ranking system that is based on the level of competition.

@figle: maybe, but doesn't that happen when you take points away aswell? I'm sure they care about their stats (why else would they go to the trouble of making a new account) so to just start over... most people won't do that.
Players with 0 games played are even more dodgy to play with (if you're looking for a really reliable game)
Anyways, i notice many people going CD because they "don't care anymore", they lost the game already or whatever. Seems to me it's just not clear enough that going CD is bad manners.

True enough. I don't know how you should deal with it - a difficult decision

I have to say that, for myself, neither points nor GR influences who a ally with. Early in a game it may be about a personality I happen to know to be trustworthy, but in the later stages, it's always about alliances of convenience: stopping a win, grabbing some SCs from a former ally knowing I have a new one backing me, etc.

But I do agree that tournaments and the leagues (especially the leagues) can affect GR wrongly. Hell, my one and only win was in a torunament game and came about as part of the grand negotiations so I and another player could get the top two spots in our particular league by stopping the guys who had been drawing and getting points that way. I got 12 for the win and they just got their 8 while my ally was in all three of the draws for 12 as well.

An epiphany: why not multiply the current GR by the win percentage of a player to come at a final GR?

Draugnar, initial alliances are pretty critical, so admitting to metagaming by latching onto a buddy you like early on is a serious matter, you should think twice about/before cheating like that.

LOL TMG!! Here I thought my idea of a WTA solo getting ~ a 5% bump up in relative rating got a cold reception, I can't imagine what yours would get if someone took it seriously.

So, because I have 1 win in 50 games, my GR is only 4? (it's just shy of 200 I think right now). That definitely doesn't reflect my skill. Just because I participate in a large number of 3 and 4 way draws doesn't make me a horrible player.

Um, I said I used experience and how they play. I don't "latch onto a buddy". Hell, I ally with people who I don't care for all the time but who I know play a certain way. That isn't cheating and is considered perfectly acceptable (go look at the rules page). Everyone does it, whether they admit it or not. You can't help but have prior ecperience influence your decisions. If you know somebody always stabs by Spring 05 or is a pathological liar, you keep your alliances short and don't trust them.

And what part of the word "may" eludes you MM. Also, please tell me where I said I ally with a buddy. I said I *may* chose my alliances based on *knowledge* of how another *plays*.

Hmmm, perhaps I should not have tried to get a reaction out of you, Draugnar, though it was a nicely controlled one at that. BUT, this is why I try to only play anonymous games, everything you know outside a game should be checked at the door in a new game, which admittedly is difficult, but that's why I talk so much to people in a game, to get a read on that person in that specific game.

Anyway, what do you do when playing a game as Turkey against a Russian and the agreement was made to DMZ Black Sea and that Sev would slide into Rum in 1901. As Turkey, regardless of who is playing Russia, it's pretty damn hard to resist not taking Black Sea and sliding into Arm even if the end result is being accused of metagaming against said Russia. Now, let's imagine this specific Russia has already accused you of metagaming against him in multiple past games, do you align with the Russia just to show the Russian you are not metagaming or do you stab the Russian and play the game on its own merits? Personally, I'd hope I'd play the game on its own merits.

Well, the initial reaction was nicely controlled, but you're slipping a bit Draug! ;-)

Draug, the fact you are considering "knowledge" that is outside the current game, even if you don't act on it, is the issue.

In that case, I play based on the merits of the move. But you also have to consider the future (within the game) and will pissing off Russia in 01 serve your long term needs for a short term gain?

I think you are right, we all do it to some extent, playing anon limits this for me.

Draugnar, thanks for taking the metagaming discussion to a new thread, let's end it in this one.

MM, I respect you as a player (unlike some) so I'd never go off an a barrage (well almost never). I have started a new thread on this specific topic so we don't derail this one. Let's move the discussion there.

@MM, regarding this:

"everything you know outside a game should be checked at the door in a new game,"

I strongly disagree. If people choose to play non-anonymous games, I feel knowledge of their play style is extremely relevant. If this was an exclusively in-person group, and I knew that person A invariably made short sighted stabs, then I'm going to be less willing to take risks with my border with them.

If people are *not* willing to have their past used as an aid to decision making by their opponents - an unwillingness to draw, a willingness to kamikazi if the #2 player stabs them while trying to set up a stalemate line against #1, etc - then IMO they should choose to stick to anon games, rather than become annoyed if someone isn't willing to trust them

PLEASE POST METAGAMING COMMENTS IN OTHER THREAD

Strat:

"Rules:

2. No Meta-gaming
You can't make alliances for reasons outside a game"

Strat, this reply is also in the other thread, let's take it there.

Let's get back to TMG: "An epiphany: why not multiply the current GR by the win percentage of a player to come at a final GR?"

And my response: "LOL TMG!! Here I thought my idea of a WTA solo getting ~ a 5% bump up in relative rating got a cold reception, I can't imagine what yours would get if someone took it seriously."