Upgrade to Chess.com Premium!

Suspected Cheater 'Strip-Searched'

  • SonofPearl
  • on 12/30/12, 3:31 PM.

News has emerged that a Bulgarian chess player was "strip searched" after being suspected of cheating during a tournament in Zadar, Croatia held from 16-22 December.

There were 36 players in the tournament, including 16 Grandmasters, yet the 26-year old computer programmer Borislav Ivanov with a rating of only 2227 Elo finished with a 6/9 score, beating four GMs along the way to claim 4th place.

The top placings at Zadar

# Name Elo Pts
1 GM Predojević Borki  2600 6.5
2 GM Stević Hrvoje  2622 6.5
3 GM Sumets Andrey  2638 6.0
4 Ivanov Borislav  2227 6.0
5 GM Jovanić Ognjen  2538 5.5
6 GM Kožul Zdenko  2638 5.5
7 GM Šarić Ante  2533 5.5
8 GM Martinović Saša  2530 5.5
9 GM Cebalo Mišo  2402 5.5

.

Despite the "strip search" no evidence of cheating was found. 

Stanislav Maroja, chairperson of the chess union in Zadar District said, “After the eighth round we received a signal about Ivanov’s game and after his game with Borko Predoevic, who later on won the tournament, we decided to check on both of them. There were suspicions that Ivanov has some electronic tools to help him and in my capacity of a judge I decided to make a move in line with the FIDE rules. It is not true that we made him strip naked. He himself took off his t-shirt, while we emptied his pockets.”

The FOCUS Information Agency quoted GM Zlatko Klaric, “Ivanov is chess programmer...he made moves like a computer, which was obvious in the game vs Jovanovic. Technologies are so developed now that theoretically, since the games were aired live, Ivanov’s friends...could have sent him hints for his moves through chips, which could have been placed under the skin, in the ear, or in the teeth.”

Either Ivanov had the tournament of his life, or he found a way to cheat without being detected. All of his games from the tournament are below.

This disturbing story also raises many other questions, two being: a) what are the limits of an arbiter's powers to search an individual suspected of cheating? and b) do we now have to scan suspected players for implanted computer chips?

.

.
Games and results table via Chessgames.com.

14495 reads 167 comments
7 votes
more Misc

Comments


  • 19 months ago

    paladino2012

    I know what happened! Ivanov bought the DVD course of GM Igor, with all the secrets to win against GMs!

    Another possible explanation could be that he read De Lamaza, and practiced his method.

    Further if anyone search a game database, it is very easy to find 2200 players beating 4-5 GMs in a row. Even Kasparov is scared by 2200 players.

  • 19 months ago

    _valentin_

    dokter_nee:  The question is how to determine what constitutes sufficient evidence in a way that serves not only the judge (i.e., people like you) but also the judged (i.e., the players who've been incriminated).  
    At the absence of a full proof, including the so-called "smoking gun", such a threshold is not easy to determine from a policy perspective -- as much as a strong coincidence of move choices between a human player and an engine may seem implausible and exceedingly unlikely.  Possibilities always exist (e.g., psychics are often able to read other people's minds -- why not then read the engine's mind and is that illegal if they didn't actively seek it but just channeled it?), albeit perhaps outside the realm of what some people may be prepared to believe is possible.

  • 19 months ago

    dokter_nee

    Valentin, that's just a silly analogy.

    There just has to be enough evidence for there to be no reasonable doubt. That is the only criteria.

    Game from round 7, 100% (really 100% houdini top choice moves, from move 7 ! . I don't know of any game where this has been done before. From the opening until the end 100 %houdini top choices. 

    If you know how much better engines are than humans at the moment (Carlsen doesn't stand a chance against them). There really is no reasonable doubt that he 26 year old, 2200 player, could play like this.

    But I guess he could have been also very very very very very very very very lucky...  

  • 19 months ago

    _valentin_

    Just because you have some strong evidence (based on your perception) does not constitute "proof", especially when it comes to accusing a person of some nefarious action.  Even in criminal cases, people don't get sentenced until the "weapon" of the crime has been found and clearly associated with the presumed-guilty person.  In this case, no weapon of crime has been found, despite obvious attempts.  The evidence is still there, but this is not a proof, and cannot be used as such in any meaningful court case or similar.

    In the case of treating people with dignity, it must be the case that one is presumed innocent until proven guilty, not until suggested guilty.

  • 19 months ago

    dokter_nee

    Yeah Sittingpawn, you make good points about innocent until proven guilty.

    But I just went over games 1 2 and 3 with Houdini 2.0(starting from move 10), and in games 1 and 3, there are like 6 instances where he doesn't chose the main move, but choses the second main move (with very little difference), or 2 times the third (with also very little difference, which can be due to individual engine differences).  I often watch games of Carlsen, Aronian, Kramnik (the absolute top GM's) online, where you have live Houdini Analysis. These guys never play such computer perfect games. Never!

    So from the 3 matches I analysed, he played 2. That is real proof he had engine assistance. Really, it is.

    And in the second game, he also follows Houdini for a very large amount of moves.

  • 19 months ago

    NM Petrosianic

    Basic statistical AN of the facts (the moves) indicated an extreme likelihood of cheating.  Normal top GM games hit an engine's top 3 moves up to about 60 percent of the time.  The player in question hit much higher, exception of game 8, which is a curious game how the game played out in the first 16 moves and then in the remaining phase.

    In addition to the statistics, some moves may be red flags to computer use... moves that only a computer would play b/c it's not very good or b/c a human would never simplify or translate the position in such a way b/c of risks.

    I ran tests of his play against Houdini [engine matches of his moves to the engine's top move] - anybody can do the same.  You will find that the match-ups are very abnormally high.  I didn't present all that I found b/c it is a thankless task. Wink

    Anybody can present a position statement and use rhetoric and claim that is an argument too.  Innocent  I don't like to support cheats though in cases in that is extremely clear to me that this occurred.  It's good to know that people have the best intentions though regardless of their judgments and their feelings and their ability to navigate concrete and circumstantial evidence. Wink  Not the first time this sort of thing has happened in the chess world, sadly.  Too bad, but the world has much bigger problems (the propaganda machine Sealed) and much greater pleasures... so good-bye and enjoy your day.

  • 19 months ago

    ClavierCavalier

    sittingpawn, these young chess players who beat GM's are usually on their way to being a GM or in a simul.

  • 19 months ago

    ChocolateTeapot

    This debate has become far too heated, and far too irrational. All this random talk of embedded microchips is highly fanciful, but downright ridiculous. It would be nice to be able to hear from Mr Ivanov himself, but given the way he has been treated, I think that is highly unlikely.

  • 19 months ago

    sittingpawn

    @Rustyknife... NO! that's inane!

    1.) Being a computer programmer has NOTHING to do with being a cheater and just because you are a computer programmer should not in any way incriminate you.

    2.) What do you mean by makes houdini precision moves? This statement then can apply to most player above 2000, I guess anyone who makes a good combination that agrees with Houndini now (ahhh I remember the days of rybka) is some cheating chess philanderer.

    3.) So from now on if the lowest rated player plays better than his rating, lets accuse him of cheating... What's the point of even showing at a tournament if you're going to be the lowest rated player because if you do well then you're a cheater and I guess the flip is that you perform as expected and loose every match.

    4.) Beating a GM is the goal here, btw, the term GM does not mean you're god-like in chess and it doesn't mean you should automatically win. How many GMs get beaten by 12 year old kid prodigies, should we accuse them of cheating?

    5.) I'm glad you got that down to the 26th decimal place, if you'd only gone for the 25th I'd think, well that's not too bad.

    I don't know the standards for innocence in Croatia, but I should state that not only should a person be considered innocent until proven guilty, he also was shown innocent by every method possible... He was stripped search, the live stream was shut down and still he performed above his rating... Prove him guilty and maybe I'll agree with you on point 5.

  • 19 months ago

    P_G_M

    Honestly, I just want the Tata Steel Chess Tournament to start because I want to see if Carlsen can reach a rating of 2870 and maybe Anand will finish second place for a change, but most probably he will move down to #8 or #9 in the live rating list.

  • 19 months ago

    _36darshan--

    lol

  • 19 months ago

    sittingpawn

    @sisu~ I disagree, he described the move as against chess culture, which I find pure twaddle to say the least, but even if we were to interpret it as meaning against opening theory we still cannot deny that the move is considered a possiblity. His whole premise was that this move goes in the face of getting the knight to c4, but it doesn't the knight can still get to c4. The problem with this discussion is that we're nit picking over an opening choice which is by no means great but isn't some 1200 rated player move. How often do we see players of lesser strength try new or more obscure openings against higher rated opponents hoping to catch them off guard. I just don't see the relevance of picking that one move out of all the games and calling attention to it.

  • 19 months ago

    sittingpawn

    Also on this point of a computer chip under the skin. I know that it can be done and it of course has been done to animals for tracking but a chip alone isn't going to do anything, even if you got it to produce an electric stimuli to a person the chip would certainly not have enough power to produce any noticeable effect by the user. A larger power supply and some form of mechanical device would also have to be put in to communicate with the person well enough to send a message, likely as in a Morse code. To go through this to cheat in such a small tournament seems less likely than the idea that for one tournament he played above his pay grade.

  • 19 months ago

    sittingpawn

    @Petrosianic~ It's a shame you've given up on discussion so quickly and wont be able to hear the arguments against what you've pointed out. I guess that says something.

    To your suggestion that I do the research myself. I cannot consider sitting down with a computer and using it to analyze a game as research because it isn't. True research would be to investigate everything known about this individual, his past games, the direction his playing life has taken, the stability and rise of his ratings and his work as a "chess programer". Genuine research would be intricate and too much to vest for someone who has been shown to not have been cheating. I find it funny that someone on here points out that he lost the match that they after canceled live feed but then goes on to state he beat a GM in the next round which was also not a live feed but still says the guy cheated. Humans like to attack people who do well and hate to see people succeed. We have to tear them down when this occurs, it is our biggest flaw. 

    Petrosianic, you state "in addition to the statistical improbability of matching Houdini's #1 choice every single move for hundreds of moves." Which I believe you were referring to the match which you said he was 100% with houdini till move 116 but you fail to answer my question about move 69... Ke3 I believe it was. And also I must confess a bit of skepticism in your statement that he concurred with houdini 100% of all moves leading up to even move 69 and here is why. While computer programs assess information based off of complex algorithms, the truth of the matter is the hardware and software are equally important and chess is of a nature that even computers will not completely agree on a positions best move 100% of the time. If this were the case there'd be no reason to pit them against each other because the games would all be the same. If you play a computer you will find that even if you make the exact same moves they computer will not. This is not because it's learning but because at times many moves are equally as good as other moves and it really is just how deep the computer has looked and where it is in it's inspection when it makes it's move. If you set the parameters of an engine to analyze the game at say 10min a move and you run that same engine 100 times, you'll likely get some repeats but you will never get 100 identical analyzations. So, the probability that this guy is cheating and that your computer just so happens to coincide with the computer helping him along for 116 moves, especially when there were many inaccuracies and even a bit of a blunder played by that player well before the 116th move just seems to be astronomical. You two would have to have been running the same program (unlikely in his case if he was trying to test out a program he wrote in a real life situation, which was implied in pointing out he was a chess programmer), running on the same hardware while running the same background non chess software... in the same atmospheric conditions with equally similar electromagnetic interference and... well you get my point, I hope.

    You stated: "I don't spell it out for people point-by-point, move-by-move, why one move is a computer move, and one move is a human move; I assume people have the capacity of reason to do that for themselves." First off the ability to do so isn't really an issue of capacity of reason. It's an issue of knowledge. If it were so easy to point out a computer move from a human move than there would be no issue with whether or not someone was cheating. If we automatically could find a computer move in a string of human moves than this debate wouldn't be possible. What makes a move computer "like" is really what is the main point of contention. But the real truth is that if he played similar moves but then somehow blundered later and lost all of his games nobody would then go back and say he was cheating for a point of each match because those moves were obviously made by a computer.

    "And yes 9...Nc6 in game 9 consigns Black to a worse position than normal as several continuations suggested that was based on human, computer and strong human database analysis." Bent Larson made a career off of playing openings with moves that were "worse". Truth be told it's not a move that is horrible or even grossly off base with the najdorf. It might be worse that is hardly proof. Nakamura plays worse positions, I can ramble off many players and I'm sure there are many at a 2200 level that play openings with a move that is "worse" than standard book after the 9th move and yet still manage to pull of exceptional middle and end games. Regardless of the other evidence move 9 in that game is not evidence of cheating. Bad opening decisions with remarkable outcomes occur regularly.

    Again it's a shame you gave up the discussion when you said; "I'm done reading this thread; I find analysis of evidence (the moves themselves) more interesting than opinions founded on not-evidence (speculation)." This idea that the moves are the only factor which needs to be considered is confounding to say the least. If we put a players moves into a bubble and look at nothing else we only get a partial view of the situation and cannot come to any accurate conclusions. I think however my "opinions" do have some evidence, my pointing to move 69, my discussion about computers and of course there is that bit of evidence overlooked by your houdini 1.5, the guy was strip searched and shown not to have anything on him, the live cast was shut down and still he beat a GM and played a helluva game, lost or otherwise. There's more to this than his moves, there's his background, his play style his mood, the psyche of his opponents, etc...
     
    Good luck, have fun and don't cheat. And I apologize for this long winded and likely mistake ridden post.
  • 19 months ago

    sisu

    @sittingpawn: the move by itself is not enough to qualify someone as a bot, that is true. However it is what Petrosianic described it Smile

  • 19 months ago

    sittingpawn

    @sisu~ I understood this but I disagree with the premise that this move is dubious or in the least so bad that it warrants calling someone a bot. Openings aren't as cut and dry and people believe. Many a Grandmaster have made careers of reviving openings once considered "dubious". It also isn't completely alien to the knights movement to c4. Keep in mind the move is less common so maybe as to keep the rote memory of a GM off his toes and it can bring one into a middle game position quicker so if the player is less confident in his memory of opening positions but more confident in his calculation skills it could work; it's not a horrible move. I mean for goodness sake, the guy could easily have made a mistake in the opening that wasn't punishable but the idea of using that move to lend credence to an idea that he was botting is a horrible stretch.

  • 19 months ago

    az09

    My coach had me teach to a group of his junior students some basic openings to help improve their understanding of some basic ideas. Although I did try quite hard and eventually got preety good at playing several openings when I tried using them against reasonably strong opponents in friendly matches (at least 1600+, in -say- 30/20 matches) to get a good opening position or at least surprise them I blatantly failed. Using a bunch of different openings against a GM when you are a 2200+ player seems impossible to me. Then again, he might be the next Kasparov or Carlsen or Houdini 3.
    That's all I had to say. 

  • 19 months ago

    _valentin_

    A remarkable result against players 200-300 points above him.  Whether or not it's a cheating case, it's evidence that there are things we don't and can't know or predict even in an apparently logical domain such as chess.

    There is practically no possible way to detect small computer chips if they've been implanted carefully, without being overly disturbing to people's privacy, so at some point soon the searching method will have to admit defeat.  Alternatives exist though, such as not transmitting the games live, and/or enclosing players in a cube (which is doable for small tournaments) that hinders or scrambles electronic transmissions.  It seems like one of those so called "arms races" without an end.

    In information and communications theory there is a concept called a "covert channel", referring to the hidden transmission of signals via some apparently undetectable mechanism (e.g., scratching your head 3 times means queen, scratching twice means rook, etc.).  There is a proven result in that theory, which says that good covert channels are undetectable even in theory.  The only thing that can be done against a covert channel system is to force a slowed down rate of transmission (e.g., making it a lot of work for the transmitter to create and send his/her signals in a way that keeps it all undetectable).  In other words, if done carefully and diligently, this stuff is virtually impossible to catch or prove.

  • 19 months ago

    P_G_M

    Honestly, I just want the Tata Steel Chess Tournament to start because I want to see if Carlsen can reach a rating of 2870 and maybe Anand will finish second place for a change, but most probably he will move down to #8 or #9 in the live rating list.

  • 19 months ago

    tomxypdq

    I played a kid who's father used a computer in the guest room of the Mechanic's Institute of SF. UCSF rated. The kid went to the bathroom almost EVERY move after move 10 or so. Yes, he went into a stall so I couldent see what he was up to the two times me and then the TD checked on him. He played like a GM for about 24moves though rated 1600 or so. Think he was packin software n hardware in his baggies. Very frustrating and annoying. 

Back to Top

Post your reply: