Home > Publishing Schedule > Cheating in Strasbourg?

Cheating in Strasbourg?

GM Rausis has been declared caught by FIDE’s anti-cheating commission. Essentially with a phone in the toilets. It is my understanding that there is a clear admission of guilt. (read more on Chess24).

His six year rise to 2685 thus comes to a halt. Apparently you cannot go from no. 500 in the World to no. 50 in your 50s after all. Or at least not in this case.

We can expect him to get a 5-10 year ban, as life-time bans are not allowed under CAS.

This is of course if he is convicted by the Ethics Commission.

We cannot expect the police to get involved. There was a big US case of cheating in a major sport where they failed to get the police involved. And when my shed was burgled last year, the police did everything by email.

Here are my questions:

Do you find it appropriate of FIDE to announce an alleged cheaters name and indeed guilt in the way it has been done?

Do you have personal experience with cheating happening against you?

Do you think FIDE should roll the dice with relation to statistical accusations, such as Aldama in Chicago Open 2018, despite CAS in general rejecting statistical evidence. In case you do not know of this case, you can see the following game, where 16.a4/a3 and 22.Nd2/d3 shows a problem with the signalling process, rather than deviation from the source of the moves, Stockfish on Chess24. (And yes, I would love to be sued by Mr Aldama for him to clear his name. You can check his other games in the same tournament for corroboration).

Categories: Publishing Schedule Tags:
  1. Steve Mayer
    July 12th, 2019 at 21:33 | #1

    There are some strong Cuban players who are known to cheat in US tournaments. That said, I’d prefer not to get sued

  2. An Ordinary Chessplayer
    July 12th, 2019 at 22:31 | #2

    “And when my shed was burgled last year, the police did everything by email.” When will we see that episode of the police-procedural drama on the television?

    “Do you find it appropriate of FIDE to announce an alleged cheaters name and indeed guilt in the way it has been done?” It depends on the case. Primarily it comes down to – was the announcement factual? If so, I don’t have any issue with how it was done. See your last parenthetical comment. When the evidence is so clearcut that the *alleged* perpetrator can only *further* damage their reputation by suing or even by merely protesting, then put the cards face up on the table.

  3. An Ordinary Chessplayer
    July 12th, 2019 at 22:46 | #3

    “We can expect him to get a 5-10 year ban …” First, he needs to be invited to a 10-player round robin with similarly rated players. Make him suffer to scrape a few draws, *then* ban him.

  4. July 12th, 2019 at 23:10 | #4

    I believe in justice, it is difficult to talk about these topics, but if you are going to accuse somebody of cheating you should have some evidence, It is not possible to say that he was cheating because he played the moves of the computer, maybe he had trained a lot for weeks or months or even years, you can see the games of Carlsen, he plays like a true computer, but I am no saying he had used the help of one of them, no saying the opposite either.

    Best wishes!!

  5. Jopi
    July 13th, 2019 at 08:45 | #5

    What a bizarre accusation on Aldama. Purely on statististics and some possible move order issues?
    I think you should have something more concrete than that before making such an accusation.
    Can you just accuse people and they will have to actively defend themselves to clear their names?
    I find that unbelievable. It’s Bad Quality.

    Here is a famous dutch statistics case
    http://luciadeb.nl/english/summary.html
    Short summary :
    Nurse gets convicted because of statistics (as in too many people die with her around as a nurse).
    Many years later it is proven that the statistics just don’t hold up.

    Not all people with exceptional results cheat. Some do, some don’t.
    I think it is really hard to prove and without hard evidence it is better to give them the benefit of the doubt and congratulate them with their great performance.

    Cheating is tough. I suspected once someone of cheating (He was away multiple times during the game and nowhere to be found in or around the playing hall) and the position on the board was quite tense. It can make you quite paranoid at the board. I didn’t know quite what to do with it.
    Go to the tournament director? With the fear of being wrong and making a complete ass out of yourself?
    Stalk the guy when he leaves the board?
    In the end I lost the game. My opponent was around 200 ELO stronger, so I let it go.

    I do think that if you get caught cheating you should get a life long ban.

  6. Cowe
    July 13th, 2019 at 18:59 | #6

    How sad – and unexpected – from Rausis. He could sink into deep concentration and come up with amazing moves and concepts, of course without assistance. Not the kind to suffer against fellow grandmasters.
    From personal experience I remember a Corsican player who was 1900-ish and started winning prizes with 2400+ performance. He was quickly convicted but no prizes nor points were given back. Also, fraud need not involve technology: in the glorious days of Budapest norm tournaments, you could buy results for a modest price in the first rounds, but the closer you came to the norm, the higher the price for a half or full point 🙂
    There were also the ubiquitous arrangements in the last two rounds of open tournaments between prize contenders, a real shame almost inherent to our game. But the saddest part for me is to see true champions falling for cheating (Rausis, Feller and some others), like in the Tour de France where they need both talent and cheating to be on top. For chess though, just cheating could bring anyone on top as long as it is undetected. That can be a problem with Stockfish 12 implants.

  7. Michael Hehir
    July 14th, 2019 at 12:44 | #7

    Since the guy confessed I have no problem with FIDE announcing his guilt in such a manner, it should actually help to discourage more cheaters.
    I once had a master level opponent go into the bathroom after I made my 3rd move as Black and come out 15 minutes later knowing all the moves to a very off-beat opening! After 10 more moves it was time for another 10 minute bathroom break on his time. This guy also suddenly gained over 100 points in his late 30’s-early 40’s.

  8. Jacob Aagaard
    July 18th, 2019 at 17:19 | #8

    @Jopi
    The evidence against Aldama is as convincing as the photo of Rausis. Remember, the Rausis photo could have been doctored.

    And the nurse case is not comparable. Where you can see that a player is cheating is when he plays ALL the same moves as the engine. Then when he is put in a room without his assistant (many people spotted the helper), he plays hardly any moves like the computer. It is especially interesting when people play like the engine when there are many decent moves. Carlsen never does, nor do any other top player.

    In this case we even have the source of the moves, Stockfish on Chess24.

    The alternative is that this player is randomly playing at a 3000+ level when you see all these suspicious things and at 2300 level when he is isolated.

    No judge will ever be convinced about this after expert advice.

  9. Pinpon
    July 18th, 2019 at 17:43 | #9

    By the way , Sébastien Feller was by the end of May 2019 , definitively sentenced to six-month jail suspended by the Thionville Court

  10. Topnotch
    July 24th, 2019 at 20:20 | #10

    Photos/Video of patrons occupying the loo is a slippery slope to say the least. Seems like electronic cheating in chess is here to stay, and it’s not clear what effectively can be done about it.

  11. Seth
    July 24th, 2019 at 21:55 | #11

    Topnotch :
    Photos/Video of patrons occupying the loo is a slippery slope to say the least. Seems like electronic cheating in chess is here to stay, and it’s not clear what effectively can be done about it.

    I know the what the correct PUNISHMENT for it should be, but probably no one would actually sign on to the death penalty. I wouldn’t sign on to it either, but it would certainly deter cheating some.

    It really grinds my gears that someone would even think about using outside assistance. Like, what the hell is wrong with you.

  12. Benjamin Fitch
    July 26th, 2019 at 00:33 | #12

    According to Star Trek, by the 23rd century no one will ever have to use the bathroom. That will solve this particular electronic cheating issue. (Sometimes you need technology to fight a problem introduced by technology.)

  13. Patrick
    August 12th, 2019 at 21:10 | #13

    To answer the three questions in the article:

    1) NO! Convicted Cheaters, yes! Alledged? Absolutely Not! That’s defamation of character if you ask me. If he’s convicted, by all means announce then!

    2) It depends on your definition of cheating. I was at the World Open in 2009, and from round 3 onwards, if EITHER player had a 70 or 75 percent score (dont’ recall which it was) or better, NEITHER player could use headphones for anything (even music) or leave the floor the tournament was on, which was the second floor. My opponent, who had only 1 of 2 but I had 1.5 of 2, went down a level and outside. Nobody knows to do what, and he may not have “cheated”, but he violated the rule. It was a 10-minute deduction from his clock. The game ended in a draw.

    3) Statistical accusations are complete and utter BULLSH*T! I was rated 1999 (USCF) in August 2010, and faced a 2447 (USCF) as Black an won fair and square. The game of my life! So you are going to tell me I cheated because statistically a 1999 player shouldn’t find the moves I found? I have also gone 5-0 in Palatka, Florida that same year as the 5 seed (not 1 seed) of a tournament literally 4 weeks before that win over the 2447 (which put me at 3-0 at that point in that tournament as well). You going to tell me that a superior stretch of 8 games shows cheating based on statistics? I have never cheated in my life! And I’ve also failed numerous times since! Chicago Open is only 1 more game than that…

  14. Patrick
    August 12th, 2019 at 21:11 | #14

    (cont) 8-game stretch of mine going 8-0 and facing many higher players, one of which 450 higher! Anything can happen in a short stretch. A 9-round event is a short stretch!

  15. Bulkington
    August 13th, 2019 at 08:03 | #15

    @Patrick
    Let`s get the statistics straight. The cheating suspicion is not based on a probability to win against a higher rated player. This probability can be derived from the ELO formula and it might be higher than one expect. Lower rated player winning against higher rated player happens everyday and is appreciated and part of the system.
    Instead, the suspicion comes from a test developed by Kenneth Regan, based on a comparison of the played moves to top moves the engines suggest. This test appears to be very reliable. It is extremely unlikely that your wins would have been detected as suspicious.

  16. Jonathan O’Connor
    August 13th, 2019 at 13:12 | #16

    I heard recently about a case in Germany, where a player was suspected of cheating. He refused to allow the arbiter to search him, but said that the police could. However, the police after arriving to find that it was a case of cheating in chess, they just laughed and left the building.

    Now all players who wish to to play in the German chess leagues, at all levels, must sign a form stating that the arbiter may search their person if there is suspicion of cheating. I know of at least one player who refuses to sign it.

  17. Patrick
    August 13th, 2019 at 14:21 | #17

    @Bulkington
    So let me make sure I have this straight. If you ran the following game through a computer, and too many of Black’s moves match the computer move, it must be cheating? The following was that game from round 3 of the 2010 Columbia Open. White is Alexander Matros (2447):

    1.b3 e5 2.Bb2 Nc6 3.e3 d5 4.Bb5 Bd6 5.f4 Qh4 6.g3 Qe7 7.Nf3 f6 8.fxe5 fxe5 9. Bxc6 bxc6 10.Nxe5 Nf6 11.Nd3 Qe4 12.Nf2 Qg2 13.Qe2 Ng4 14.Qf1 Qxf1+ 15.Rxf1 Nxh2 16.Rh1 Bxg3 17.Kd1 O-O 18.Nd3 Bg4 19.Kc1 Rf1+ 20.Rxf1 Nxf1 21.Ne5 Be6 22.Nxc6 h5 23.Nc3 h4 24.Ne2 h3 25.Nxg3 Nxg3 26.Be5 h2 27.Bxg3 h1=Q 28.Kb2 Qg2 29.Be5 Rf8 30.Nxa7 Rf1 31.Rxf1 Qxf1 32.Nc6 Bd7 33.Nb8 Qb5 34.Bxc7 Bf5 35.Be5 g5 36.a4 Qc5 37.c3 Qb6 38.d4 g4 39.a5 Qh6 40.Bf4 Qh1 41.Na6 g3 42.Bxg3 Qg2 43.Ka3 Qxg3 44.Nc5 Qxe3 45.Kb4 Qe7 46.a6 Bc8 47.Kb5 Qa7 48.Kc6 Bxa6 49.Kxd5 Qf7 50.Kc6 Qg6 51.Kc7 Be2 52.c4 Qg3 53.Kc6 Bd1 54.b4 Qc3 55.d5 Qxc4 56.d6 Bf3 57.Kc7 Qxb4 58.Nd7 Kg7 59.Kd8 Qxd6 0-1

    Now let’s say 46 of the 59 moves by Black match the top choice by artificial intelligence. Does that make one suspicious based on what you are saying?

  18. Bulkington
    August 13th, 2019 at 18:06 | #18

    Regan`s algorithm is more complex and requires quite some interaction with the engine to get the metrics it needs which apparently is an absolute strength of the move and how much the move is weaker compared to the engine`s first choice. The implementation required 35000 lines of code in C++. Which is something. I have this information from the article “How to catch a chess cheater” explaining the maths behind it in an comprehensable way, also provides a portrait of Regan, quite an impressive character with an impressive vita. You find it in the Internet. But besides all those details, it is easy to test such an algorithm if you think of the millions of games from the last century where it should not complain.

  19. August 13th, 2019 at 18:21 | #19

    @Patrick not a mathematical/statistics expert but the idea is if had 50+ games where non database moves were matching more than a top level GM pre computer times for similar number of games, than this would be suspicious. One game is not enough, there may be factors in the game, which are making the lines forced in some way.

  20. Topnotch
    August 14th, 2019 at 04:39 | #20

    @Patrick

    I only glanced at your game for about a minute or two with Chessbase and Stockfish as I was curious to see how much of it was book and how you played once the game left theory. 11.Nd3 is a losing move according to the Engine, and your move while not bad isn’t even rated in the top 3 by Stockfish which considers 11…Bg4 as strongest, to play it you have to work out that 12.Bxf6 Qxf6 loses for Black but the intermezzo 12…Qe6 followed by h5 is winning for him. These kind of calculations are tricky so are raw statistics, many factors have to be taken into consideration, the main one being that it is near impossible for a Human being to produce top engine moves consistently in a non forcing position, and if he does its 99.99% certainty he is cheating. Even the finishing moves chosen by an engine are often ones a Human would never choose in a game, Humans seek the simplest most practical path to victory, while the Engine seeks the shortest.

    Honestly it always surprises me when very strong players gets caught cheating with engines during a game, since one would assume that the engine would only need to be consulted once or twice at a critical juncture to tilt the balance decisively. However it seems that chess strength is irrelevant as greed and insecurity takes over and the Engine is consulted compulsively like an addict seeking his next fix.

    What really grinds my gears though are how many, for want of a better term, non cheaters, go on the defensive and…

  21. Patrick
    August 14th, 2019 at 15:34 | #21

    @Topnotch
    To answer your question about “book”, it was all book until that 11th move. The normal “book move” is 11.Nxc6 and after 11…Qe4, 12.O-O with an equal game.

  22. Jacob Aagaard
    August 16th, 2019 at 12:38 | #22

    @Patrick
    What if the statistical evidence is 100% correlation between a players moves and Stockfish on Chess24? In a whole tournament?

    Of course it becomes more difficult when you have players that have 100% correlation with top 3 or top 5 choices of the engine. One Youth World Champion was alleged to be cheating in this way.

  23. Patrick
    August 16th, 2019 at 15:29 | #23

    @Jacob Aagaard
    I see what you are getting at, but there also has to be other factors used to determine if cheating is occurring. Take a game of mine from 2000. I was black, and it started 1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 e6 5.Bg5 dxc4 6.e4 b5 7.e5 h6 8.Bh4 g5 9.Nxg5 hxg5 10.Bxg5 Nbd7 11.exf6 Bb4 12.g3 c5 13/d5 Qb6 14.Bg2 O-O-O 15.O-O b4 16.Rb1 Qa6 17.dxe6 Bxg2 18.e7 Bxf1 19.Qd5 Bxe7 20.fxe7 Bd3 21.Ne4 Bxb1 22.Nd6 Kc7 23.Bf4 Kb6. At this point, White played 24.Nxf7 and lost in 11 moves. This can lead to multiple questions:

    1) If White had played 24.Nxc4+ and went with the perpetual, the entire game would be a well-known Book Draw. Not sure if artificial intelligence would say every move was best, or if it prefers say, the 19.Kxf1 line, but here you have a “perfect” game between two 2000 players.

    2) White resigned on move 35, so Black had to make 11 moves, moves 24 thru 34. If all of those moves are top candidate or top 3 candidates according to artificial intelligence, did I play perfectly for 34 moves, or 11 moves? The first 23 were all book and played in probably less than 5 or 10 minutes by each player.

    3) I have not seen the games, but if everything he plays is highly theoretical and known to a draw (Dragon, Botvinnik Semi-Slav, etc), did he exactly play all that many “correct” moves?

    I’m just saying it takes more research than just statistical facts. Research further and maybe you have a valid case with the…

  24. Patrick
    August 16th, 2019 at 15:30 | #24

    (cont.) statistics, but sometimes you really don’t! That’s why I say it can’t be used as a valid standalone “blanket process”.

  1. No trackbacks yet.

 Limit your comments to