Cheating in chess

[Roger de Coverly]

I think given the metrics that I outlined most people would conclude that the player was cheating.

I would not be so ready to jump from the factual information that moves matched an engine to the opinion or hypothesis that the player had consulted an engine whilst in play,

[Jacob Ward]

I think given the metrics that I outlined most people would conclude that the player was cheating.

I would not be so ready to jump from the factual information that moves matched an engine to the opinion or hypothesis that the player had consulted an engine whilst in play,

I think the point being made, though, is that you may not need to make that jump if you have all the additional information that the platforms can gather.

The jump seems very large to me, if it is on move matching alone; not nearly so large if the sort of additional information Matthew has suggested can be used.

A player whose moves often match the engine might be suspicious. A player whose moves only match the engine in complex tactical positions, at which point they have a pattern of receive move - minimise screen - 20s pass - return to screen - input move? That looks far more suspicious to me.

(To be clear, I don't think anyone is suggesting that this is exactly the information the platforms are using - it's just an illustration that information available to the platforms other than the moves can be highly probative.)

[JustinHorton]

- it's just an illustration that information available to the platforms other than the moves can be highly probative.

How sure are you of that

[JustinHorton]

On the Facebook posting I mentioned here, somebody called Wolff King Morrow comments as follows:

I remember 10 years ago playing a game on chess-dot-com where I used an opening book (which was totally legal) in a cc game. As soon as my opponent broke book, he blundered on the spot and I jumped on it to win in 20 moves. People thought I must have cheated because of the 100% accuracy ratio, but I explained the situation and did not get banned for it. Then I stopped playing there for many years. Then out of the blue, I get told my account has been banned for cheating. This is EIGHT years after the fact. When I once again explained the situation, only this time they refused to accept anything I had to say about it. Instead, they demanded I admit to cheating. I was very insulted, but people were referring to my banned account link as a mark of shame, so they put me in a position where I had to falsely confess to cheating when I didn't just to get the mark removed. From then on, I simply refuse to play cc on there any more and only play live blitz. Using books for cc chess, even ones written by GMs, can make it look like you're matching computer play. It's just not worth the mark of shame when you don't deserve it.

[Jacob Ward]

- it's just an illustration that information available to the platforms other than the moves can be highly probative.

How sure are you of that

I think the general statement I made (ie that non-move data can be probative) should be pretty uncontroversial. It's not far above saying "evidence is evidence". I'm not claiming any huge insights here!

The devil, of course, is in the detail, and it may be that the way the platforms use the data is unsound. Unfortunately they do not share those details, so I have no idea. It is clear that whatever process they use is not perfect, but that is not a huge insight either!

[Gerard Killoran]

One answer could be that, in long form chess online, players have to declare any opening books or databases they are using, as the game progresses.

They could also be able to declare 'last book move' when they diverge.

[David Sedgwick]

I don't know about later reappearing, but disappearance is what seems to be happening. It's not helped by lichess and chess.com issuing bans or incriminating flags without the say so of the bodies notionally in charge of the events they are hosting. Organisers seem to be reacting by reversing the process of making results and games readily available. Team events are easier to cover up than individual ones, since all that's needed is to turn wins or draws into losses. Making game scores difficult to find helps the cover up as well.

On the face of it, events held for junior players appear to have been particularly affected, although rapidly improving players are also vulnerable to the accusation that "on past results, you aren't good enough to play that well".

Here is another example of a selective cover up policy. I could scarcely believe it when I first heard it, but it checks out from the limited research which I have done.

On this Forum we all know about the infamous red circular Mark of Shame on Chess.com, there to condemn you as a cheat for evermore.

What I didn't know was that the mark doesn't appear in the case of titled players. Their accounts are closed quietly without any public indication.

So, if I am falsely accused of cheating on Chess.com, the dreaded mark will appear, with all the consequences of my good name and reputation as an arbiter.

If IA Stewart Reuben or FA Kevin Thurlow were to suffer a similarly baseless allegation, there would be no such consequences for them. Their CM Titles would save them.

[Kevin Thurlow]

"If IA Stewart Reuben or FA Kevin Thurlow were to suffer a similarly baseless allegation, there would be no such consequences for them. Their CM Titles would save them."

I have just added CM to my profile... This seems to be a rare case where CM actually counts for something.

[Nick Grey]

So why use this approach for untitled players. with all the consequences of our good names and reputations.

[NickFaulks]

I have just added CM to my profile... This seems to be a rare case where CM actually counts for something.

I think this shows proper respect and I may review my own decision not to use these platforms.

[Ian Thompson]

I have just added CM to my profile... This seems to be a rare case where CM actually counts for something.

Does it count if you added it yourself? Anyone could do that and say they had a title they did not.

When I registered at chess.com as a titled player I had to provide proof that I was who I said I was and they then created the titled player account for me.

[Kevin Thurlow]

I notice that Chess.com have awarded me rating points as they have identified an opponent as a cheat. This seems rather unfair, as in the game in question, we played a normal opening and I just overlooked a one-move threat on about move 15...

[Jacob Ward]

I notice that Chess.com have awarded me rating points as they have identified an opponent as a cheat. This seems rather unfair, as in the game in question, we played a normal opening and I just overlooked a one-move threat on about move 15...

My understanding is that if they identify a player as a cheat, all their games get reversed (even if there is nothing suspicious about that game in particular), on the basis that cheating in one game casts doubt on all your others.

[JustinHorton]

I think that's right, though of course one consequence of that they are directly telling what may be alarge number of people that you'tre a cheat, which is problematic if they're wrong.

[Ian Thompson]

That's different from lichess then. They only tell you you got some points back, but not from which game.