Cheating in chess

[Matt Bridgeman]

I think the 2000 rating for an IM was for rapid play. We don’t go on chess.com very much, but isn’t it common for titled players not to bother too much with rapid play on that site? I remember Simon Williams in his YouTube long play series had to basically stop using auto pairings for rapid games and get people he knows to challenge him because he was constantly getting cheats who were subsequently banned. He’s rated something like 2180 after 400 odd rapid games on there.

[NickFaulks]

Using computer assistance for puzzle chess is an offence?

[Wadih Khoury]

Using computer assistance for puzzle chess is an offence?

My personal stance is that if anything is competitive, multiplayer or has a leaderboard, then no assistance should be allowed.

Puzzles, puzzle rush, puzzle battle all fall in one of these criteria so no assistance should be allowed.

[JustinHorton]

2850 puzzle rating does not equate 2850 Fide and hence Magnus+ level. Especially for juniors, puzzle ratings are routinely several hundred points above their play rating.

I see my puzzle rating has been as high as 2569, more than 500 points above my OTB FIDE rating

[Matthew Turner]

[*] An IM with 2000 chess.com rating is an anomaly. Most players at that level have a chess.com rating 100-200 higher than their Fide. 2000 blitz is a junior level in the 10-14 age bracket.

My chess.com rating is 1991 and my last three games have been against a strong FM, an IM and a GM with ratings 2007, 2123 and 2247 respectively.

[NickFaulks]

My personal stance is that if anything is competitive, multiplayer or has a leaderboard, then no assistance should be allowed.

I wasn't asking about people's personal stances. Either it is forbidden or it isn't and if it isn't then it should not be used as "evidence" leading to vigilante lynchings.

[Wadih Khoury]

My personal stance is that if anything is competitive, multiplayer or has a leaderboard, then no assistance should be allowed.

I wasn't asking about people's personal stances. Either it is forbidden or it isn't and if it isn't then it should not be used as "evidence" leading to vigilante lynchings.

It is indeed forbidden:

Puzzle rush and Rated tactics
--The below are NOT ALLOWED--

Engines - You may not use any engines to help solve tactics! This includes Chessmaster, Fritz, Komodo, Houdini, Stockfish, Chessbase with any active UCI engine, etc.
Tablebases - Tablebases are chess databases that contain every possible move for endgames of up to 7 pieces. These will always lead to a win when a win is possible, and are for that reason the same as asking an engine to analyse a position. Tablebases are not allowed!
Help from other people - You may not to ask for help from another person! Do not ask anyone for advice on your tactics! Do not let anyone else play on your account!
Books, opening databases, or any outside help at all! - For puzzle rush and rated tactics, on Chess.com, if you are looking at anything other than the game in front of you, you shouldn’t be! Focus on your game, and play with your own mind and strength!

Full rules here: https://support.chess.com/article/317-w ... -chess-com

[Robin Nandi]

Using computer assistance for puzzle chess is an offence?

Yes. It is forbidden in the same way as using computer assistance in a blitz game.

[KMcCarthy]

The point of my original post is the mapping of all the things together ...I am stronger in all forms of the game than those I have described so benchmarking all aspects highlights the extraordinary allowing greater focus to be put on their games etc....Someone achieving 2850 on the Chess.com puzzles is au fait with using the engines to solve ..at around 2500 the puzzles get very deep in places and beyond patzer-thinking....Yes there is often a difference in ratings across different time controls but our comparisons (using over 2000 example players) does not show any that are over 1000 different between Blitz and Rapid ...and 73% of the comparisons actually show Blitz as the higher...THEN we look at the gameplay ...over many days/games ..not taking bare % scores but really looking in to them until there is certainty ..only then is a report to Chess.com entered ..no lynch mob ..no Kangaroo court but a 'success rate' of banning in the high 90s so must be on to something

[JustinHorton]

no Kangaroo court but a 'success rate' of banning in the high 90s

...which is not all that good in the circumstancess

[NickFaulks]

a 'success rate' of banning in the high 90s so must be on to something

You must be so proud.

[Richard Bates]

Isn't it possible to get good at puzzle rush purely through the use of memory.

I sometimes wonder if i might end up with a lichess ban for falling foul of their rules on "deliberately" losing. The difference in my standard when i am genuinely attempting to get my rating higher, and when i am playing totally casually, after several drinks or half asleep in the middle of the night must be several hundred rating points. And often of course i may miss some absolutely basic tactic early on, only to fight back and draw or win the game.

The problem is that the rules of the sites themselves, which are almost written as if they are expecting people to treat every game as if it is a World Championship match, and the actual attitudes of players who may just be seeking to pass the time, try things out, have a bit of fun, prioritise creativity over calculation etc etc are often completely at odds if treated literally.

[KMcCarthy]

Lots of denial on here ...as in other parts of our chess community ...I would say 93% 'hit' rate for a casual sleuth is fairly successful ...not 'proud' but these cheats need rooting out or online chess competitively remains tainted...and yes some players on both main platforms(including an old colleague) have been wrongly banned in my opinion ...and also others for 'sandbagging' ...a stupid rule which I too strongly disagree with Richard...so let me make it clear ...I have never reported an opponent who beat me in casual chess or meaningless tournaments ...I only report clear and obvious cheats who are ruining our game...and will continue to do so...and only when they are doing it in prize events or NCL or 2020 or County Online...The powers that be do the banning and reverse results etc ...and if people do not like being found out then surely not cheating is the easiest way around that

[Robin Nandi]

Lots of denial on here ...as in other parts of our chess community ...I would say 93% 'hit' rate for a casual sleuth is fairly successful ...not 'proud' but these cheats need rooting out or online chess competitively remains tainted...and yes some players on both main platforms(including an old colleague) have been wrongly banned in my opinion ...and also others for 'sandbagging' ...a stupid rule which I too strongly disagree with Richard...so let me make it clear ...I have never reported an opponent who beat me in casual chess or meaningless tournaments ...I only report clear and obvious cheats who are ruining our game...and will continue to do so...and only when they are doing it in prize events or NCL or 2020 or County Online...The powers that be do the banning and reverse results etc ...and if people do not like being found out then surely not cheating is the easiest way around that

I support your aim. I think you should be proud of a hit rate >90%. It is a lot better than doing nothing. The level of proof could always be adjusted to >95% or >99%.

However, I would question some of the rating numbers contained in your previous post. Particularly the use of puzzle rating which can legitimately be totally out of sync with other ratings.

I would prefer an engine matching approach based on "% of moves matching Stockfish 1st choice" and "average centipawn loss". Comparing those numbers with the distribution from OTB games with the same rating. Can then come up with a "probability of getting such engine matching or closer" assuming no engine use. Also can compare with the same player's OTB games.

Would be good to test out on the players of the past e.g. Lasker - who was definitely not using Stockfish - how many of those would the method have identified as cheating?

I have been able to produce some engine matching stats for recent online games, but have not had the time to run them over the database to come up with the probabilities. I would like it if someone ran a public service that published these stats including the probabilities for every tournament.

[NickFaulks]

and yes some players on both main platforms(including an old colleague) have been wrongly banned in my opinion

So you regard them as acceptable collateral damage? Remember that in these times this can result in the end of a young player's career.