Cheating in chess

[Roger de Coverly]

When the the full test is performed all theoretical moves ( as defined by moves played by 2300+ players) are removed, so things should look a lot better for John.
Of course, if John has done lots of computer analysis and expanded on existing knowledge, remembering lots of lines with all his opponent's responses then it is possible that he'll fall foul of both Regan tests, but in reality how likely is this to be the case for a decent club player?

ejhchess triggered chess.com's detection system merely by following lines contained in the books on his bookshelf. This was at "daily" chess, where such consultation of opening theory is allowed. If you define theory as having to be played by 2300 players, doesn't that exclude all sorts of unfashionable or plain dodgy stuff? I'm thinking of theory in things like the Kings Gambit or Ponziani as well as other obscurity.

[JustinHorton]

ejhchess triggered chess.com's detection system merely by following lines contained in the books on his bookshelf.

Just to add as I usually do that we think this but we do not know it. (Also, the account name was and is Justinpatzer.)

[Matthew Turner]

When the the full test is performed all theoretical moves ( as defined by moves played by 2300+ players) are removed, so things should look a lot better for John.
Of course, if John has done lots of computer analysis and expanded on existing knowledge, remembering lots of lines with all his opponent's responses then it is possible that he'll fall foul of both Regan tests, but in reality how likely is this to be the case for a decent club player?

ejhchess triggered chess.com's detection system merely by following lines contained in the books on his bookshelf. This was at "daily" chess, where such consultation of opening theory is allowed. If you define theory as having to be played by 2300 players, doesn't that exclude all sorts of unfashionable or plain dodgy stuff? I'm thinking of theory in things like the Kings Gambit or Ponziani as well as other obscurity.

What you are suggesting is that moves are considered by the detection systems which should be excluded, well yes I am sure that happens on occasion, but if 'plain dodgy stuff' with questionable move matching rates is erroneously included is this really going to have a huge impact on the test result?

[Roger de Coverly]

but if 'plain dodgy stuff' with questionable move matching rates is erroneously included is this really going to have a huge impact on the test result?

What I am thinking is that if only games played where both players are above 2300 are included, there's whole areas of opening theory that players of that standard never venture into. There's a line in the Kings Gambit Declined which runs 1. e4 e5 2. f4 Bc5 3. Nf3 d6 4. Bc4 Nf6 5. Nc3 Nc6 6. d3 Bg4 7. h3 Bxf3 8. Qxf3 Nd4 9. Qg3 . Play can continue with a rook sacrifice on a1. That is or was sufficiently book to have featured in MCO 10 as well as the 2006 Olympiad but you would be unlikely to see players rated above 2300 with either colour. If I were to speculate as to why, it's because both 6. .. Bg4 and 7. h3 aren't considered the best lines even if they look obvious enough.

[Matthew Turner]

but if 'plain dodgy stuff' with questionable move matching rates is erroneously included is this really going to have a huge impact on the test result?

What I am thinking is that if only games played where both players are above 2300 are included, there's whole areas of opening theory that players of that standard never venture into. There's a line in the Kings Gambit Declined which runs 1. e4 e5 2. f4 Bc5 3. Nf3 d6 4. Bc4 Nf6 5. Nc3 Nc6 6. d3 Bg4 7. h3 Bxf3 8. Qxf3 Nd4 9. Qg3 . Play can continue with a rook sacrifice on a1. That is or was sufficiently book to have featured in MCO 10 as well as the 2006 Olympiad but you would be unlikely to see players rated above 2300 with either colour. If I were to speculate as to why, it's because both 6. .. Bg4 and 7. h3 aren't considered the best lines even if they look obvious enough.

A simple check of Chessbase online will show that there are a number of games with 2300+ players from this position.

[Nick Burrows]

Raunak Sadwahni's lichess account flagged and now closed.

[Paul McKeown]

A simple check of Chessbase online will show that there are a number of games with 2300+ players from this position.

Indeed, 7. h3 has even been played once by Kasparov, albeit in a simul against Oxford United football team. And by Fischer...

I can see several games in Chessbase from the 1990s and later in which IMs and GMs have ventured this line against other titled players, and John Shaw devotes four pages to it in his magnum opus. Shaw gives a few illustrative examples by titled players which don't appear to be on the databases. Chessbase gives White a 64% score based on the games it knows, although every serious author on the KG, including Shaw, admits that Black's defences are perhaps better than adequate.

Of course, Roger is right in the sense that 7. Na4 gives White somewhat the better chances whilst avoiding all the nonsense and without taking major risks.

[Matthew Turner]

The chess.com article about the Osmak case has been highlighted on a Bridge Forum and received a rather interesting comment from my teammate John Cox (who is, it may be worth noting, a lawyer).

"As a member of FIDE's Fair Play Commission, I would like to correct a slightly misleading impression given by the article linked to above.

First, the figure of 20 players refers to the entire World University Championships, which comprised women's team rapid, individual rapid and blitz, and the same for the men. There were almost 1000 players altogether of whom 20 were disqualified, not 20 out of the 100 or so who played in the women's rapid.

Second, the statement in the article that FIDE stated there was no proof of cheating is not quite accurate. The position is that the regulations for this event provided that players would be disqualified if there was reasonable suspicion of cheating. In practice, this either means being seen regularly glancing furtively offscreen on the camera which players are required to have on them, or else playing moves which resemble rather too closely the moves which various computer engines suggest as best, according either to the platform's algorithms or FIDE's own investigations (or of course quite often both). If a FIDE spokesperson said there 'was no proof of cheating', what he or she meant was that the authorities had not proved and did not need to prove whether there was cheating, merely to establish a reasonable suspicion that there was. Consequently a disqualification is not a statement by FIDE that there was sufficient proof to establish cheating; that is not the question which the investigating panel needed to concern itself with. That is not the same as a statement that there 'was no proof'. It would be more accurate to say that the question of proof did not arise.
March 31"

[NickFaulks]

So is her reputation damaged or not?

[Angus French]

The chess.com article about the Osmak case has been highlighted on a Bridge Forum and received a rather interesting comment from my teammate John Cox (who is, it may be worth noting, a lawyer).

"As a member of FIDE's Fair Play Commission, I would like to correct a slightly misleading impression given by the article linked to above.

First, the figure of 20 players refers to the entire World University Championships, which comprised women's team rapid, individual rapid and blitz, and the same for the men. There were almost 1000 players altogether of whom 20 were disqualified, not 20 out of the 100 or so who played in the women's rapid.

Second, the statement in the article that FIDE stated there was no proof of cheating is not quite accurate. The position is that the regulations for this event provided that players would be disqualified if there was reasonable suspicion of cheating. In practice, this either means being seen regularly glancing furtively offscreen on the camera which players are required to have on them, or else playing moves which resemble rather too closely the moves which various computer engines suggest as best, according either to the platform's algorithms or FIDE's own investigations (or of course quite often both). If a FIDE spokesperson said there 'was no proof of cheating', what he or she meant was that the authorities had not proved and did not need to prove whether there was cheating, merely to establish a reasonable suspicion that there was. Consequently a disqualification is not a statement by FIDE that there was sufficient proof to establish cheating; that is not the question which the investigating panel needed to concern itself with. That is not the same as a statement that there 'was no proof'. It would be more accurate to say that the question of proof did not arise.
March 31"

Can you provide a link Matthew? Is that all that John Cox says on the matter?

[Matthew Turner]

Here is the link
https://bridgewinners.com/article/view/ ... -possible/
I have quoted John's comments in full. I respect his insights as a member of the fair play commission, a lawyer, an expert player and a friend. That does not mean that I agree with him.

[Angus French]

Here is the link
https://bridgewinners.com/article/view/ ... -possible/
I have quoted John's comments in full. I respect his insights as a member of the fair play commission, a lawyer, an expert player and a friend. That does not mean that I agree with him.

Thanks Matthew. I was wondering if John, I suppose as a lawyer and member of FIDE's Fair Play Commission, had anything to say about the fairness of the process by which decisions were taken - in particular the apparent absence of rights to make representations and to appeal. Was he at all concerned that a member of the Fair Play Panel for the event in question had publicly distanced himself from the decision taken by Panel and had seen no evidence of cheating in the case of Iulija Osmak?

[NickFaulks]

If a FIDE spokesperson said there 'was no proof of cheating', what he or she meant was that the authorities had not proved and did not need to prove whether there was cheating, merely to establish a reasonable suspicion that there was.

How does John know what he or she meant?

Can they just claim to have "established a reasonable suspicion that there was cheating" and then run away?

[Leonard Barden]

this either means being seen regularly glancing furtively offscreen on the camera which players are required to have on them,

Another detail that might have raised suspicion was that she had looked away from her screen a few times during games, for which she has an explanation: the limited vision of one eye, which provides for just 16 percent of normal vision.

"In cases when I'm nervous, I have to not focus on the board and try to rest my eyes and look to the side," Osmak said.

[JustinHorton]

Indeed, 7. h3 has even been played once by Kasparov, albeit in a simul against Oxford United football team.

Well not as such. (I can't remember why the club was involved or who represented them, but I do recall that they were just one of a number of organisations represented, and of course it didn't involve the team as such.)

Can they just claim to have "established a reasonable suspicion that there was cheating" and then run away?

Well John's a lawyer and we're not, but I would be surprised if from the point of view of defamation law, a successful defence could operate on the lines of "I didn't say she was definitely cheating, I said she might be".