When accusations fly

‘OK, there is a body with a knife there, and the police come and say nothing happened… You have to find who, why, what, but it happened, don’t pretend that it didn’t happen!’ Vladimir Kramnik deployed that analogy about the world of online chess, which he sees as riddled with cheating.

Cheating happens. Once in a while, I get demolished in a casual online game in a manner that leaves no room for doubt that my opponent’s moves were silicon-assisted. Chess.com, the platform that hosts the biggest online events, has a dedicated Fair Play team to monitor cheating, and their checks have led to over a million account closures in 2023 (0.6% of the total number).

Nevertheless, Kramnik clearly believes that Chess.com and the online chess community are blind to the scale of the problem. His concern is legitimate, since nobody really knows how much of the iceberg lurks below the surface. So the former world champion has turned to a kind of online vigilantism. His modus operandi is to post statistics which don’t directly accuse but insinuate that various players’ results are too good to be true.

Hunting for statistical anomalies in performance is a valid approach. Professor Kenneth Regan, a prominent anti-cheating expert, approaches that task with an underlying model of how likely a player of a given strength is to select certain moves. By contrast, Kramnik’s figures often seem cherry-picked, and when he claims to have consulted mathematical experts, the evidence is scant. And he blasts it out with a blunderbuss. In a recent interview with the YouTuber Levy Rozman, he claimed, ‘I just look at data, I don’t look at names or anything… There is always a chance that a player is playing fair, but it has to be examined.’ That’s a cop-out: by implicating widely, he convinces nobody. The Czech grandmaster David Navara (perhaps the most unimpeachable opponent I have ever faced) recently felt obliged to publicly defend himself after his name appeared in one of Kramnik’s half-baked lists of data.