The Latest Cheating (Non-)Scandal
According to this story in ChessVibes, untitled Bulgarian Borislav Ivanov (2227) was "strip searched" at the end of a tournament that saw him score 6/9, defeat 4 GMs and draw two more, and finish tied for third at the Zadar Open with a spectacular 2697 performance rating. The arbiter had him remove his shirt and empty his pockets, and finding nothing after examining Ivanov's pen, apologized and let his performance stand. (Calling it a "strip search" makes it sound rather salacious, even if it's (barely) true, lexically.)
Ivanov's result is of course pretty improbable, but as he's a fairly young player on an upward trajectory (even if it isn't the trajectory of a prodigy), it's not as unlikely as it might otherwise be. This looks like a good job for Ken Regan and his IPR tests. (Are you out there, Ken?)
Reader Comments (8)
Surely if you were cheating you would put in a more 'moderate' performance?
Surely if you were cheating you would put in a more 'moderate' performance?
I'm of the opinion that it will be very difficult to catch smart cheaters. Using a program for a few critical moves in a game for a few games in a tournament could make a big difference. However we've seen several instances over the years of people cheating in rather egregious fashion.
But it isn't really a matter of being smart so much as it is one of self restraint. So the question is this: Are cheaters lacking in self-restraint and we're catching most of them, or are we just not catching most of the cheaters because they're being more restrained in their efforts?
It seems to me that more than an unusually or implausibly good performance is required to charge someone with cheating. In a case like this, I would think that, at first, judicious observation of the player and his habits would be justified, and then, if anything seemed reasonably supsicious, that might warrant further action. It's not as if the performance in this case is miraculously or impossibly good, so I think his treatment was somewhat unjust.
I'm of the mind that there are very few people who would want to cheat at chess - my take is that most of these few will be caught by obviousness. Are there/will there be smart cheaters? Well given you can run Stockfish on a small smartphone it won't be long before it's on a watch or a wearable computer embedded in clothes or even skin...
I'm not surprised they were suspicious - just look at the list at http://www.crochess.com/turniri/zadar/12/index.htm - Ivanov is bottom of the rankings of 36 players and the only one without a title of some sort.
Yes, I've been on the case. Will have results tomorrow. In a cheating case the "IPR" is secondary, and currently has about 40% modeling deficiency on top of wide error bars owing to random variation. My primary test, for move-matching, is pretty sharp within 15% modeling error. I will cite both. Most important, in the absence of physical or behavioral evidence---things apart from testing games or dredging up prior accusations---the statistical results are subject to the caveats stated on my "Parable of the Golfers" page.
The last game against Saric is quite a brilliancy ! Leaving the bishop to get Nc4 and Qf1! is really deep....
The tactical blows against Kurajica are really nice too...
To me it's pretty obvious the guy was cheating. It's not a regular 2200 player game ! To dispose of grandmasters in 30 moves or so, most of the time thanks to deep tactics....... 2700 performance... and the guy is a programmer too... No wonder they searched him !
It'll be funny to follow the results of Borislav Ivanov from now on !
FM Lilov has done a 1 hour video analysing the games and explaining why he believes that Ivanov has cheated... deserves a close look.
The NY Times reports that Ken Regan has finished his analysis, and Ivanov's moves matched the computer engine's far more than any other player's in KR's database. His moves matched more closely than Carlsen's, or any other world champ's. While Regan considers the numerical analysis to be insufficient without other forms of evidence, I would regard Ivanov's incredibly weak 8th round performance (when no moves were broadcast in real-time on the internet) to be the "smoking gun" that proves Ivanov was cheating.
[DM: "Proves"? That's too strong. Without judging Ivanov's case one way or another, I think most of us would be very upset, if innocent, if we were accused of cheating and then searched, and were that to happen our performance would likely be adversely affected.]