But is it teaming?

@GustavKlimtPaints I am not sure I qualify for WTA games. Last time I checked my rating was around 1550, maybe more, maybe less. My understanding of the game could be outdated if WTA had completely replaced the original rating method, but it hasn't.

Would you like to share your experience of WTA games? Has teaming vanished/reduced/stayed the same/became more subtle? Has the problem been solved?

By the way, WTA games are available for players of any rating (as I've elsewhere in this thread). There are any rating, 1300+, 1500+ etc. WTA rated games in the list, as well as options that include anonymous, dead king walking etc.

I'd say WTA has completely eliminated games like the one in the post you linked; while it is still advantageous to eliminate a side player for obvious reason (they are closest to you and pose greatest threat, it opens your flank up for queening), people need to be a lot more careful with material, since it may be more important in the three person game. I see more often opposites not allowing queening from players across them. The biggest change though I'd say is the three person game, which has become a lot more interesting and complex. Since WTA rating works like this: 1st gets 4 wins vs. average rating, 2nd \ 3rd are equal (this is key, because there is no incentive for 2nd over 3rd) and get 1 loss vs. average rating, and last gets 2 losses vs. average rating. Oftentimes when 1 player is eliminated it will be obvious that person will be last, so the remaining 3 have no incentive other than to play for 1st in all cases. Depending on both points and material balance and how that changes throughout the rest of the game, there is no two person alliance that can work for the rest of the game; any two person temporary alliance favors the strongest of the two who would win the heads-up match, so this is something players have to keep constantly in mind (can I win 1v1 vs. that other guy or have enough points to claim?) and the material will stay a lot more even between the 3 players because no one will allow any other to get ahead too much (two won't let another get away with the game by just fighting each other when it doesn't make sense). Anyway, I played some FFA before WTA, but always quit in frustration because of the incentive to play for 2nd (opposite is eliminated, then you are next because people are happy with either 1st or 2nd) but I'm really having a blast with WTA.

Interesting! From what you describe I see some similarities with Last-Man-Standing. Do you know of any cases in WTA where the last standing player didn't won the game?

There are lots and lots of games like this.

i.e. https://www.chess.com/4-player-chess?g=879929

Hi @gustav,

I can sympathize with what you and your team have to deal with. Was just reported by red for teaming with my opposite in this game. https://www.chess.com/4-player-chess?g=896239-89

All I did all game was actively attack him, then my opposite won his queen when I checked him to open up his position. Consider this a game example of not teaming.

>>Now in order to resolve a report for teaming, instead of having to review a single game you must review multiple games, after painfully retrieving them from the game archive.
Well, umm, no. Now, if I personally receive a report for teaming, I look to see if the reporter, themselves, has done the work of sending, to me, the administrator, multiple games where they believe that these two players have teamed together. If they haven't... I don't bother either... and just circular file the report.
The problem with Skeftomilos's definition is that it is a definition that doesn't define. It describes but it does not define.

Reading some of these reports is very funny. Sometimes, for example, a player lists all three of the opponents as 'teaming'. Recently one guy said 'he doesn't know how to play'.

Do you prefer this one? Teaming: the activity of working together as a team.
This is the general definition, not constrained to the context of 4-Player-Chess.

Again, it is a definition that, besides being circular, does not define.
In one sense that definition would prevent nothing at all. In FFA there are no teams, therefore one cannot 'work together as a team'.
But of course what is meant is 'work together as if they were a team'. But in that case, again, there would be almost nothing prevented. In team 4pc the players cannot take each others pieces (even when it might be good to do so) nor can they place each other in check, they have separate chat and can draw private arrows. None of this can apply in 4pc FFA. Thus they can never 'work together as if they were a team'.
What is really, really meant is, "Play in such a way as that others, watching, might understandably be confused as to whether FFA or teams were being played." But that definition, by itself, is purely subjective. I have had players point out 'Look, they didn't take that queen'... when I, myself, didn't even see the queen was under attack... so focused was I on some other piece or some check or something.
In order for it to be truly objective it would need to be something like anti-chess... ie it would have to make taking a piece mandatory. Under certain condtions. In which case, as Gustav points out, this objective rule could be mandated by the computer, leaving administrators nothing to do. (As, again, in anti-chess, where you literally cannot move when there is a piece you can take). But that is not the game that FFA people want to play.
In the end the problem is thus, given the current nature of FFA: Blue could have taken Green's queen, without even losing their own piece. Was that teaming? Or did Blue miss it? Was blue setting up some other strategy? Did blue not want Green weakened just then because Red and/or Yellow would have been too strong?
And how is an administrator, signaled for teaming, supposed to know all of that???

@VAOhlman this is the reason I avoided the general definition, and proposed instead a domain specific definition: What constitutes teaming is for the teamed players to coordinate their pieces, and form a combined army. They support their teammate's pieces, instead of capturing them.

This is a definition that everybody understands, and the administrators would not have a hard time enforcing it. Being descriptive is an advantage I think.

That, again, is not a definition, and is not objective. Altho it does include more details:
1) Coordinate: if this is by open chat, then this is already ilegal, and enforced. I routinely chat ban a dozen people or so a day for this.
2) A combined army: This is either meaningless literally impossible given the rules of FFA as enforced by the computer. I would need more definition to tell which.
In military terms a 'combined army' is one which shares joint command and control, which uses combined logistics, etc. (That would be a cool variant of teams, that last... where you could give a rook to your partner!)
3) They support their teammates pieces: Actually all three players in FFA 'support' a piece in the two meaningful senses:
a) They make it dangerous to take a piece  if they can take it afterwards. and
b) A king cannot take a piece which is 'supported' by another player.
4) Instead of capturing them: This is by far the biggest issue that turns up in reports and discussions and the one I deal with above.

So, sorry, you haven't helped.

So you say that the proposed definition is not enforceable? That you couldn't use this definition to judge if teaming occurred in this or this game for example?

The Rules & Reporting topic includes this rule of conduct:
7. In the FFA variant, players shall NOT go out of their way to get into games together and consistently play in a way that benefits one or both of them (prearranged teaming / collusion), nor shall players use the chat to collude.
Do you think that this rule is objective, properly defined and enforceable?

 It is not objective when it says 'go out of their way', obviously, as that would involve mind reading. Several other parts are more objective, when properly defined, as I did with the chat rules:

It is forbidden to request or accept a request to team in four player FFA or solo chess. This applies for a given move or for the whole game. It also includes suicide talk, such as, “I’m so mad at red I’m just going to go whole hog against him for the rest of the game.” (It isn’t forbidden to be mad at red, or to go against him. It is only forbidden from announcing that in chat, and thus giving the other players the advantage of knowing your moves will be against red.)

It is forbidden to suggest a move in FFA; even if it is not a matter of teaming. This includes pointing things out, such as ‘Hey, Red’s queen is just hanging there’. Why don’t you resign?’ Or ‘claim victory’ are allowed but not encouraged.

It is allowed to say ‘It’s your turn!” Or “Hurry up and move!”. Moves may be fully discussed after they are no longer relevant, particularly after the game.

(Indeed the last part of the one section, "Why don't you resign" etc was debated, since by definition it goes against the rules, but people wanted it allowed, so I wrote it as a frowned upon exception.)

>>So you say that the proposed definition is not enforceable? 

What I said was that it was not a definition and not objective. That is a far, far different thing from saying that good players could not 'enforce' it. But any such enforcement would have the problems listed above. And quite frankly they won't have the time. I was sick for a couple of days and we racked up a 120 or so reports.

Aha. There is already a non objective rule in place. So the proposed anti-teaming rule would not be the first one. Would be just one more subjective rule in a long list of subjective rules, because all rules of conduct are subjective, by nature. To make the rules more objective, examples of violations have been added, but the examples cannot cover every case. And suddenly someone writes «you could choose» in the chat, and the administrators must judge if this phrase denotes collusion or not, without even having at hand a proper definition of collusion! (they made a judgement nonetheless)

What I propose is to replace the rule against prearranged teaming with a rule against ... well ... teaming. The reasons are:

1) Most players dislike teaming in general, not prearranged teaming only. I think it's evident from the kind of reports you receive.

2) Submitting and resolving a report for teaming consumes less time than submitting and resolving a report for prearranged teaming, because you have to review a single game, not many.

3) Distinguishing between teaming and not teaming is no more complex than distinguishing between prearranged teaming and "normal" teaming, because you must judge if it's teaming before judging if it's prearranged anyway.

(1) is certainly true.
(2) The difference between 'submitting' and 'resolving' is a gap the size of the grand canyon. I can easily 'resolve' all of the teaming reports given to me if they are 'submitted' with a list of relavent games. (I can 'resolve' the ones that aren't even more quickly, but in a less satisfactory manner.)
3) Wrong. Someone is not up on his statistics.
If someone says to me, "Joe and Frank cheated me in this game of poker" and I look and see that, at the end of the game, at the big pot, Joe won by a hair when Frank was dealing... I may or may not be able to resolve if they were cheating. However if one gives me a list of games, and in each one either Joe or Frank won the last big pot when the other one was dealing, my education in statistics, scant and far away though it were, can still help me to resolve that.
3b) And if you refuse to give me an objective definition of teaming, then I will not be able to enforce it. Just as I do not enforce the subjective 'go out of their way' in the current rules, but instead enforce the statistically relavent 'consistently play'.

It is simply nonsense to say that rules of conduct are subjective by nature. If there is a house rule that says that a player must say 'check' when they put the other player in check, or they will have to do 300 pushups, then, assuming that someone is watching the game and listening, that objective rule can be enforced. If, however, they are merely encouraged to 'play nice', then we live at the whim of the enforcer.
To say that 'examples cannot cover every case' is to confuse the issue. If the rule is subjective, such as 'play nice', then one might argue that. If the rule is objective, such as 'announce check', then examples, while helpful, are not needed. Merely good, solid, definition. What is check? What words are needed when announcing?
Mostly players dislike losing. Many players do not understand game theory. The combination of these two means that many players report 'teaming' when what they should have reported is 'the other guy knows how to play the game better than I do'. (Altho I have had players report that, too.)

Hmmm. So I could search the game archive for pairs of players that have played together 4-5 times, and one of them was always the winner of the game, and report these players as teamers providing the games as evidence. Statistically I should found many such pairs. You say that you would impose penalties to these players based sorely on the results of the games, without looking at the actual moves they played? I hope you wouldn't do that!

About the objectiveness of the rule "say check when putting the other player in check", here are some examples showing that it isn't:
a) Is saying échec au roi instead of check a violation of the rule?
b) Is whispering check at less that 20 decibel (not audible by your opponent) a violation of the rule?
c) Is saying check 15 seconds after the fact a violation of the rule?

About understanding game theory, this phrase has been extensively used to justify teaming, so I am skeptical about it. According to the game theory, prearranged teaming is the best strategy in WTA, so why is not allowed?

About the administrators been overwhelmed by the sheer mass of the teaming reports, it is not ideal but you could process only a sample of the reports, and discard the rest unresolved, adjusting the penalties accordingly. For example process a third of the reports, and triple the penalties for the violations you discover. It would still be a deterrent for teaming, and that's what this is all about.

>>You say that you would impose penalties to these players based sorely on the results of the games, without looking at the actual moves they played? I hope you wouldn't do that!

No. What I said was:
if one gives me a list of games, and in each one either Joe or Frank won the last big pot when the other one was dealing, my education in statistics, scant and far away though it were, can still help me to resolve that.
And that would, indeed, be much better evidence of cheating than you will be able to provide me with in practically any game. The odds, if you read what I actually wrote, are non-existant that it is not teaming. I would be interested in your lists of players and games, however.