Yeah this is bad for Hans, I would guess he has cheated a lot more than he said including in paid events and that Magnus knows this.
I don’t find that “suspicious games” video convincing. If you select tournaments which people win, or attain GM norms in, you introduce a selection bias. It’s like looking at playoff series that NBA teams win and being like “this is suspicious because they hit more threes than they normally do”. Hans’s accuracy in the games was high but not unheard of, and when I looked through a couple of the games I didn’t find any of the moves suspicious. They all looked logical and totally findable by a human. Although the findings are certainly consistent with a theory of cheating, I didn’t feel that they constituted evidence.
The first Carlsen Caruana championship match in 2018 had an average CP loss of less than 10 over 12 classical games, all of which were drawn (Carlsen won on faster time controls). One could expect a run of sub 10 CP loss games from a 2800ish player in a tournament, but for a 2500 rated player I’d guess it’s basically impossible; then again oreo’s link above, ascii shrug.
It would be nice to have all the games, as it should be relatively easy to find if there were any other players at Hans’ strength who had bouts of similarly accurate play, whether it was when going for a norm or just in general. Like my sense is that a 2500 player is almost never going to play that accurately in 5-6 games, but I could be wrong.
I watched a different ~hour long stream with this guy earlier and wasn’t at all convinced that his model is tuned for detecting high-level cheating. He was explaining that for expert cheating, you have to weight heavily toward critical board positions, and then alluded that this part isn’t such an exact science and sort of fumbled when pressed on it. Which, to be clear, I’m not saying he’s wrong, but my confidence level isn’t too high that his methods are high sensitivity for expert cheaters. It’s a similar critique I have of baseball WAR models: We already know they perform well in the aggregate over large samples, but that isn’t extremely relevant for evaluating outliers and high quantiles.
Yeah this is what I wanted to see comparison data on. Need some control for how often this happens in general, and also need to weight it on importance like normed GM matches I think, since that’s what the specific accusation seems to be.
Yeah same point, need to see all of the data with a control group of assumed non-cheaters. Just looking at his games alone makes no sense.
There’s evidence Niemann has cheated. We don’t know how good he is at it. But idk maybe if you’re an expert player, you’re automatically an expert cheat if you decide to do it.
I mean after Ken Regan, the #1 expert in the world on chess cheating, comes out and says he looked and didn’t see evidence of cheating, I am giving exactly zero credence to Some Guy doing ad hoc analysis on his computer. But also the video was just not convincing even prior to that imo.
Still streaming so can’t get a timestamp but it starts about an hour 20 minutes in
If you just want results chuk played very unusual lines (didn’t castle and had his King on f2 for most of the game) survived into an end game that led to a draw after 50 moves without a capture or pawn move. Game was a little over 120 moves total
Really like the 15 + 10 time control. Matches are actually exciting to watch, and lots of blunders of winning positions and blunders that lose. And with 4 rounds a day you can’t do endless prep for each opponent. Just way more fun to watch live than classical.
Magnus performed at 2939 in the prelims despite forfeiting to Hans. They interviewed Magnus after and asked him about the Hans drama and he said he would probably say something after the tournament
Caruana said a bunch of interesting stuff in an interview, summary stolen from Reddit here:
that rumors about Hans were swirling for a long time but he never found them credible
that Magnus wanted to drop out before the tournament when Hans was chosen as replacement
that Magnus made the wrong decision to drop out and Fabi thought it was emotional
at the same time, Fabi views Magnus’s motives as a selfless protest, willingly damaging his own reputation to speak out about an issue he believes is harming the community
that Fabi doesn’t trust Regan’s methods to detect cheating because they exonerated a player he was sure was cheating
he thinks Regan should prove he can detect the cheating Hans admitted to as a minimum test
that cheating at the highest level is a serious threat to the game and he worries chess will become like cycling where everyone feels the need to cheat
that online tournaments with $200,000 prizes provide serious incentive to cheat and that the money is life changing to most people in chess
they discuss the issue with chess.com bans not being public and not leading to OTB sanctions
Fabi doesn’t think FIDE can sanction Magnus. Players are allowed to withdraw from tournaments, and they can’t sanction him for the early resignation because that tournament isn’t FIDE sanctioned
he also doesn’t think Magnus has much he can say in a statement, doesn’t think he has proof. Expects another nonstatement
suspects Hans’s silence is because he lawyered up
talked about his experience with Petrosian cheating against him. He felt his game against him was weird at the time but didn’t say anything to anyone except his coach. It only came out because Wesley was willing to speak out publicly a few days later -said he’s become increasingly pessimistic about the ability to catch strong players who cheat since that experience. He said he’s thought about the problem of cheating a lot since it happened
he doesn’t think there is adequate punishment for players who cheat online. Partially due to the secrecy of online bans. Says he knows of players in the top 50 who have been banned online but faced no repercussions in their career
says chess is half functioning on an honor system right now. Says rating or being an established player does not equal trust, and top players do sometimes believe other top players are cheating. Said there are some he trusts absolutely and others who he thinks could be capable of cheating
Interesting. I thought the very fact he lost was sort of evidence of cheating. Don’t these people know who is very likely to win before game even starts.
