Not when the ELO is close. It is for these players.
| +800 | 0.99% |
|---|---|
| +750 | 1.32% |
| +700 | 1.75% |
| +650 | 2.32% |
| +600 | 3.07% |
| +550 | 4.05% |
| +500 | 5.32% |
| +450 | 6.98% |
| +400 | 9.09% |
| +350 | 11.77% |
| +300 | 15.10% |
| +250 | 19.17% |
| +200 | 24.03% |
| +150 | 29.66% |
| +100 | 35.99% |
| +50 | 42.85% |
| 0 | 50.00% |
This is a random chart I found for chess. It may not be exact but it’s the idea.
This makes sense. Thanks.
The evidence is entirely circumstantial and weak, and there’s nothing from the actual Hans v Magnus OTB games that suggests cheating so far to anyone’s knowledge. I also suspected maybe it was a Postle-type situation when I saw the headlines, but it’s really quite different. If this was actually analogous to Postle, guy would have been busted quickly. Instead, if there does turn out to be any cheating here, it’s going to be minimal and strategic cheating, like maybe taking an engine consultation not even at a rate of once per match.
The interesting concept here is that cheating just a little bit can result in disproportionately large advantages, and that’s where this diverges from Postle by a lot. A better poker analogy would be something like a 2 bb/100 winner cheating just enough to win 3 bb/100 or something to that effect. There’s a quote from Anand (somewhere in this article) that seeing one relatively simple engine evaluation per game could be worth 150 ELO points, and Ken Regan, the top expert on chess cheating, more or less agreed with this number. It raises serious questions about the degree that this could even be detected and how that might affect the legitimacy of the game. See @ChrisV’s bullet points about Caruana’s skepticism over Regan’s ability to detect this type of cheating.
The strongest piece of circumstantial evidence is that Hans recently admitted to cheating online at ages 12 and 16, but then Chess.com banned him and released this statement:
Difficult to read this as anything other than “You’re lying and here are the receipts, feel free to explain.” I think Carlsen’s actions and the fact he’s willing to go all-in on this should count as reasonably good circumstantial evidence as well, but it’s not entirely clear to me if he is alleging OTB cheating or simply objecting to the fact that this guy is a known lying cheater online. There’s a bunch of other stuff like bad analyses of match play and ELO rating timelines that are poorly designed and not really evidence of anything, but I also don’t think Ken Regan’s analysis works as counter-evidence for the reasons Caruana is saying. That particular comment starts around 41:45 in the interview below:
hes been a known online cheater for years, if it was only the online cheating that wouldnt explain why Magnus elected to play Hans OTB in St Louis and then immediately withdrew from the tournament after the game and refuses to play him at all now. i think its clear Magnus suspects more than just online cheating
After his game today Magnus said he will make a statement next week. When asked if he had any evidence, he said something like that will be addressed next week. Also said he’s been getting support privately since his statement yesterday or whenever it was.
Best guess I think Magnus knows Hans cheated a decent amount online, suspects that he has cheated OTB and objects to having to play a serial cheat and play guessing games about whether he’s really playing an engine. My feeling is that gun to head he’d bet Hans was not cheating in the Sinquefield Cup game but that he feels like his play was negatively affected by tilt from the situation.
right cheating in chess might be just used for one or two moves a game, ie, if postle only did his shit once a session he would’ve gotten away with it for sure
also, Hans isn’t cheating in every game, which in theory should make for easier to catch when he is, though I’m sure every GM has some game they played ridiculous just cause that’s how games go range wise from people.
0 voters
I think it was Svidler said something like that if he could merely consult the engine eval once a game, not even see the move, then he’d probably have been world champion. It makes sense, like you are trying to figure out whether to go into some sharp variation where it’s unclear if you have the advantage or settle for a probably drawn position, you consult the engine eval and it says +3 for you, that’s all you need to know.
Yeah I saw that too. The thing I wonder is how much this would allow someone to evade detection compared to choosing a top N computer move. I haven’t thought about it much but the point is that computer move matching models don’t really work here; instead, the inference problem becomes something like “abnormal play after change in evaluation status,” and it’s not clear to me if that’s just a different problem of similar difficulty or a much more difficult problem.
Magnus just said he’ll have a statement next week and that won’t be the last he has to say about cheating in chess or something like that
He’s approaching OUR HEALTH CARE PLAN IS COMING IN TWO WEEKS status.
Working hard, thank you!
I’d like to see old Manny wriggle his way out of THIS one. Ah! Well. Nevertheless,
Excluding the resignation against Hans, Magnus went 19 wins, 7 draws and 1 loss in the Julius Baer. Won the final against Erigaisi 4.5 - 0.5. Just untouchable all event. I am once again begging FIDE to do away with classical chess and make rapid the standard. The games are so much more interesting and it is a format that rewards creativity over the board without becoming as random and error-filled as blitz can be at times.
In other news, Magnus has said he is going to continue to refuse to play Hans. Danny Rensch said in a Reddit post that Magnus has not seen any behind-the-scenes stuff at chesscom, but it seems clear that Magnus is extremely convinced that Hans has an illustrious cheating career.
I still like tournaments like the World Cup where you play a small number of classical matches, then if it’s tied move on to rapid with increasingly (decreasingly?) aggressive time constraints. A much stronger player will win the classical match most of the time and then we players are roughly equally strong at classical then the shorter time formats produce more interesting matches.
Yeah I just don’t really see the point. I don’t really care who is stronger at classical. I just see the existence of the format as status quo bias, basically. Like should we give the players 5 hours each, that would be even more “real” chess and would determine who is “really” the strongest? Classical made more sense in pre-engine times but it is increasingly a war of preparation, which is making the games more sterile, and it’s not clear to me what I am getting in exchange for that.
In the final day of the Julian Baer, with Magnus already leading 1-0 on match score, he came straight out with the Pirc in game 1, just picking an opening which is known to be worse for Black but produces dynamic positions, and wrecked Erigaisi with it. Erigaisi, now needing to win on demand with Black, also went for the Pirc and Magnus dismantled it. In what sense is it “real chess” or “better chess” if they instead have to play out some position 25 moves deep in Ruy Lopez theory? Seems to me like what I saw in those games was a much better demonstration of chess talent.
I agree with all that.
I have a soft spot for longer format chess mainly because I think there are a lot of weaker chess players that never get any better because they never play slow enough chess to really think through their moves. I think it’s the biggest weakness in the current chess community, a lot of players that are starting to play now play nothing but blitz and rapid and it sets a low upper limit on their improvement. So to have the pros play nothing but that format is going to “ruin” chess for a lot of club players by underscoring the misconception long format chess is obsolete and irrelevant. If all the top tournaments go to rapid format, the local tournaments will follow suit. This is arguably good for chess as a spectator sport but IMO it is bad for chess as a participation sport. More people should play slow chess.
And it would have been 5 - 0 if he didn’t just need half a point in the last game on Saturday and go for a draw from a winning position
