Chess.Com getting sued.

[Mick Norris]

I have now been reinstated at chess.com and have received an apology.

[Matt Mackenzie]

I'm disappointed, had assumed Plaskett and Woffindens book Bad Game 2 would be with us imminently

That would be tricky, unfortunately.

[Jacques Parry]

Congratulations Justin! I don't suppose you got an explanation too?

[JustinHorton]

Kind of, though not an adequate one (i.e. it vaguely explained where they might have been coming from, it didn't explain any of the problems with their procedures).

[JustinHorton]

Reddit

[John Moore]

Justin - pleased to hear you've been reinstated.

[Roger Lancaster]

I have now been reinstated at chess.com and have received an apology.

Pleased to hear it but, of course, this simply begs the question of how many others - few of whom will have been as articulate and well-connected as Justin - have been wrongly accused of cheating by Chess.com and similar platforms.

[Roger de Coverly]

Pleased to hear it but, of course, this simply begs the question of how many others - few of whom will have been as articulate and well-connected as Justin - have been wrongly accused of cheating by Chess.com and similar platforms.

I could suspect that if these platforms are being used for higher profile events involving top Grandmasters, that their anti-cheating methods are going to come under greater scrutiny. Justin was accused of playing a high number of book moves. That was a correct deduction by their software but they were seemingly oblivious that the competition was being played under rules where this was permitted. Equally they may have failed to take into account that authors would use very likely use engines to look for suggestions and check the reliability of games and analysis.

Most top GMs appear to have a good memory for theory and variations, so if they've analysed an original and unpublished line with an engine, they are likely to be able to successfully recall it and would want to be able to play it without facing accusations of cheating.

[JustinHorton]

Pleased to hear it but, of course, this simply begs the question of how many others - few of whom will have been as articulate and well-connected as Justin - have been wrongly accused of cheating by Chess.com and similar platforms.

I could suspect that if these platforms are being used for higher profile events involving top Grandmasters, that their anti-cheating methods are going to come under greater scrutiny. Justin was accused of playing a high number of book moves.

Well maybe, but just also generally playing at a strength a bit higher (but not enormously higher) than they thought I should have been, a problem that could have been averted by their waiting a little bit or by properly investigating the content of my games.

[Roger Lancaster]

Well maybe, but just also generally playing at a strength a bit higher (but not enormously higher) than they thought I should have been, a problem that could have been averted by their waiting a little bit or by properly investigating the content of my games.

That sounds crazy in that, apart from the fact that many players sometimes play "at a strength a bit higher [but not enormously higher]" than might have been expected from their ratings, it's almost inevitable in the case of rapidly improving juniors - and Chess.com doesn't record DoBs.

[Ian Thompson]

Justin was accused of playing a high number of book moves. That was a correct deduction by their software but they were seemingly oblivious that the competition was being played under rules where this was permitted.

I believe Justin was playing Daily chess.

The rules say:

"In Daily Chess (turn-based games with several days per move), you may consult any resource which is not engine-based. This includes books, opening databases (including the Chess.com Explorer) for [www.chess.com/openings|opening moves], and thematic games (though not their engine analyses). Tablebases are NOT allowed. You may not consult an engine, or another human, to provide an opinion on the opening database, tablebases, self-preparation or analysis that would relate to a particular game-in-progress on Chess.com."

The rules are not very sensible because they put the onus on the player to ensure that they are not using a reference source that was computer assisted. If it's a printed book or magazine you're consulting, how would you know whether the author used engine assistance to write it, other than restricting yourself to very old books written before engines existed.

They also rather strangely say that you can use someone else's previous opening preparation (providing its not computer assisted), but you can't use your own prior opening preparation - "You may not consult ... self-preparation or analysis that would relate to a particular game-in-progress on Chess.com."

[JustinHorton]

What the guy said, among other things, was the following:

In correspondence play, the use of legal tools (books/theory/otb analysis/endgame studies) combined with the sheer amount of time players get to analyze, have (sic) pushed clean human players into a region of accuracy that can make the numbers very hard to determine the difference between clean and guilty.

My view of this is that it's not actually "very hard" at all if you look at the games and observe the large number of egregious errors in them, and that I find it "very hard" not to conclude that the policy was to issue bans in a peremptory manner without at any point thinking about the detail.

[JustinHorton]

Well this is interesting

[Paul McKeown]

Well this is interesting

Urgh. Banned for beating a Grandmaster (or for that game only, Grandpatzer). chess.com can be thuggish at times.

[Paul McKeown]

Reddit

Well done. Persistence and protesting loudly do sometime finally result in justice. I'm pretty sure that that were to happen to me, I would just say f**k it and move on, but your principled approach has something to be said for it, too.