December 30, 2012 14:04

Bulgarian player strip searched after suspection of cheating

During the Zadar open, held 16-22 December, a Bulgarian chess player was strip searched after being suspected of cheating. The arbiters didn't find anything and apologized, but the news did reach Croatian main stream media.

Borislav Ivanov, untitled and rated 2227, had the tournament of his life at the Zadar Open which concluded last week. The 25-year-old Bulgarian, who works as a programmer, scored 6/9 and a 2697 rating performance which included victories against GMs Bojan Kurajica, Robert Zelcic, Zdenko Kozul and Ivan Saric. At some point the arbiters decided to search the player. Suspected of getting help via an electronic device, Ivanov had to take off his clothes.

After the eighth round there were suspicions that Ivanov had some electronic tools to help him and in my capacity of arbiter I decided to make a move in line with the FIDE rules,

Stanislav Maroja, the chairperson of the Zadar County Chess Federation told daily newspaper Jutarnji List.

According to the newspaper Ivanov voluntarily took off his shirt, had his pockets emptied and gave his pen for inspection, but the arbiters didn't find anything and then apologized to him. A number of Croatian newspapers reported on the incident, e.g. the Croatian Times, Focus and the aforementioned Jutarnji List.

It's unclear whether the current FIDE regulations give an arbiter the right to search a player. The official Laws of Chess don't mention anything specific, so perhaps Maroja referred to the following paragraphs:

13.1

The arbiter shall see that the Laws of Chess are strictly observed.

13.2

The arbiter shall act in the best interest of the competition. He should ensure that a good playing environment is maintained and that the players are not disturbed. He shall supervise the progress of the competition.

Two months ago a German chess player was suspected of cheating at the German Bundesliga. He refused to be searched, whereupon his game was declared lost. This decision wasn't based on FIDE regulations though, but on specific Bundesliga regulations. The regulations for the Zadar Open don't mention anything about (penalties related to) cheating or the possibility to search players.

The tournament had two separate groups, and the Bulgarian participated in the top group. This A group was for players rated above 2300 FIDE but players rated between 2200-2300 could choose whether they wanted to play in "A" or "B". Both were 9-round Swisses with the time control 90 minutes + 30 seconds increment. Borki Predojević and Hrvoje Stević tied for first with half a point more than Ivanov.

Borislav Ivanov's games

PGN file

Peter Doggers's picture
Author: Peter Doggers
Chess.com

Comments

Bartleby's picture

Stay fair. This is a U18 player on his way up.

Both big jumps in playing strength and strange effects of the rating system can happen when a young player makes rapid progress. I think we all agree that it would be nice if Ivanov is just a case of delayed development. Do such cases exist? The anecdotal evidence so far is a magnitude or two below what Ivanov achieved. But to me it's good enough to accept it's possible.

Thomas's picture

I don't think Frits Fritschy meant to be unfair ... . Yet, Elo 2086 at the age of 17 (and just 29 rated games in 2012 before Groningen) doesn't suggest that Ivo Maris is a very ambitious and very talented rising star from whom such results could be expected sooner rather than later. Hence, the Groningen tournament webpage correctly calls his result a big surprise (without asking any questions, also correctly).
At/starting from a much higher level, the most prominent relative late bloomer is probably Levon Aronian (just Elo 2522 in January 2001 at the age of 18).

brabo's picture

As for many years been a teammate of Tigran Gharamian, I can tell you that his rise was more spectacular. In 2002 at 18 his rating was only 2239 fide !!! In 2011 he achieved at 27 years a rating of 2676 so an increase of 437 points which is absolutely a stunning performance.
I also know the reasons of this but out of high respect I am not going to disclose this here (nothing to do with cheating).

Bartleby's picture

...was meant as reply to Frits Fritschy

Frits Fritschy's picture

I'm pulling out of this discussion, not because of a lack of fresh arguments, but because it's going on too long. I made my opinion clear enough, and respect that of anyone else (except for the occasional shouters)

I wish everyone all the best for 2013.

Anonymous's picture

Having read all the posts up to this point, I guess we can conclude there is no real evidence whatsoever Ivanov cheated, right? All. the stats in the world "prove" nothing in this matter.

Anonymous's picture

No real proof ... of course he's been accused at least twice before according to GM;

"Croatian gross-master Zlatko Klaric, on the other hand, said that Ivanov was cheating, because he was already accused of this at chess competitions in Bulgaria and Serbia. “Ivanov is chess programmer, who since mid-2011 until now had won only one rating point, while at the Zadar tournament – 60."

Maybe he's just misunderstood?

Anonymous's picture

Any concrete proof forthcoming at those earlier instances of Ivanov being accused of cheating? I want to know how he did it! Don't you?

RG13's picture

IF he cheated then I want organizers to know how for sure. However as someone seemed to imply before, he will either fade out and be irrelevant or continue to rise and eventually either get caught or confirm his new level.

Thomas's picture

So the "evidence" is: one weak GM well past his prime comes up with vague and (see below) internally inconsistent statements. Maybe Zlatko Klaric just wanted GM Klaric to be quoted in a newspaper, this may well be the first time in many years?
If Ivanov cheated before in other events, how come he wasn't successful? So "it seems he cheated before" (read: 'someone' said so) and "first success since mid-2011" (actually wrong, see my previous post) seem to be contradictory statements.

BTW if Ivanov is indeed a _chess_ programmer he has become some sort of chess professional just not a professional _player_. Then his job, including advice from more experienced colleagues, will also make him play better over the board? It could also explain, at least to some extent, why he "plays like an engine"!?

RG13's picture

@Thomas, playing "like an engine" is not just a stylistic thing but it is the ability to analyze the tactical nuances of a position much quicker than even the best human players. If the rate at which his moves correspond to the top choices of a strong engine is more than the GM's then that would be serious grounds for suspicion. Not finding evidence after a search MAY only mean that the suspicious party has evaded detection.

Anonymous's picture

Excellent points...it's rather disturbing someone is considered to be cheating during a chess game without any concrete proof whatsoever...if cheating he either had access to an electronic device on an ongoing basis, or he had very small earpiece with someone following the game and telling him what to move and where. Maybe he had an accomplice tape an electronic device behind one of the toilets in the men's room and would use a secret signal in the tournament room when Ivanov should go and check it. IMO, until concrete evidence is found of cheating, the apology from the TD at the tournament should stand.

RG13's picture

Also in a tournament hall with at least 64 boards an accomplice can 'innocently' stand in front of a board that corresponds to the target square on a chess board. If a strong player knows the target square then he can usually work out which piece goes there. Errors in spite of cheating can be explained by the ambiguity which can arise with this method. This could be easily done in a tournament hall that allows spectators to roam freely.

dokter_nee's picture

I've checked some of his games from this tournament with houdini 2.0
In the games of round 1 and 3, there are only a few instances where he didn't take the top line my houdini suggested(started looking from move 10), but took line 2 or line 3, (each time there was only a slight difference in the line).

In game 2 (which he lost) he also follows houdini for large streches of moves.

In game 7, he plays 100% all the top lines given by Houdini from move 7 ! At that point he plays the top line which gives 0.00 while the other lines are like -0.20 for a couple of moves (which suggest to me he was slightly worse in the opening). Even at the end, when there were several completely completely winning variations he still made the top Houdini lines. I've never seen such 'luck'. I follow the super GM's(Carlsen, Aronian,...) online with live houdini analysis, they never ever play like this.

dokter_nee's picture

It's not game 7, but the game of round 5 against
Robert Zelcic (2560) where he plays from move 7 until the end at move 32 on every move the houdini top move...

brabo's picture

I didn't do the check with an engine but just replaying the games and you feel the hand of an engine.
Anyway can we condemn somebody if we see 100% or close to 100% of his moves corresponds with a topprogram or not? Some people say no, you need hard evidence in the sense of finding out how he cheated, others say yes. I remember in Belgium we had recently a case where some lady was condemned for murder without any real evidence, just circumstancial. It was a big scandal although the verdict was done by a jury.

Thomas's picture

Off-topic as it refers to a different (ongoing) event, but rather on-topic as it also deals with cheating or cheating allegations: A Basel newspaper quotes Peter Erismann, president of the Basel Chess Festival:
http://www.schachfestivalbasel.ch/images/Presseartikel/BAZ30122012.pdf (bottom of scanned page, in German)
"We assume that only fair players sit at the board. Some professional players have a bad reputation in the scene. We simply do not invite them. ... As we select our players, we can to some extent boost the spirit of fairness."

Sounds good at first sight, but many "huuus" at second sight? To what extent can a Swiss Open 'select' the participants? What is the 'scene'? Fellow professionals, other organizers, journalists, the blogosphere?? Are some professionals excluded just because they were victims of (unfounded or unproven) cheating allegations at earlier occasions?
It also seems at least a bit naive to think that only professionals might cheat (there are also amateurs in the field), and only those who were - right or wrong - suspected at earlier occasions.

BTW this article is much longer than an earlier one announcing the field (Vachier-Lagrave, Grachev and eight other "world top" GMs rated above 2600) - says something about what appeals to mainstream media? And the journalist doesn't seem to be up-to-date to put it mildly - later he writes that a cheater can be caught if one discovers that some moves coincide with the first choice of "the well-known chess program Fritz" (for how many years has Fritz lost its benchmark status?).

dokter_nee's picture

Yeah, I went over 5 of the games, with houdini 2.0 on my laptop. games 1, 2, 3, 5 and 6
in 1, 3 and 6, there are only a few moves(maybe like 5 per game, starting from move 8 or 10) that are not the top line in my houdini, but instead they are the 2nd or 3rd choice, always within 0.10 of the top evaluation..
In game 2 which he lost, he also follows houdini for big parts of the game.

Combining all factors, I no longer have a reasonable doubt about this.
(Off course you can doubt everything, and it could be the most extreme case of 'luck' this planet has ever seen. That is possible)

dokter_nee's picture

And game 5 was off course 100% the top line starting from move 7.
Carlsen in topform would have had no chance to win against that. (a draw possibly, Carlsen is pretty good)

Reader's picture

In my opinion, it is improbable for a 2200 rated player to play at 2700 standards but on the other hand 2200 players are reasonably strong and can beat multiple GM's in a tournament, even if the odds are 1 in 10,000 that the GM's would play into something he can win, it doesn't mean he can't be lucky

Reader's picture

In my opinion, it is improbable for a 2200 rated player to play at 2700 standards but on the other hand 2200 players are reasonably strong and can beat multiple GM's in a tournament, even if the odds are 1 in 10,000 that the GM's would play into something he can win, it doesn't mean he can't be lucky

HR Sadeghi's picture

Chess is in quite a sorry state !

the champ's picture

I have again gone through all nine games with Stockfish on my laptop. Ivanov makes a total of 290 moves (I did not evaluate the opening moves that could be considered to be theory). 256 of these are the first choice of Stockfish. That makes 88%. It would have been higher if he hadn´t started to lose his magic in round 8 after the 15 first moves. In the last 19 moves he commits several mistakes, small and big. It is reported that the internet relay of the games went down during this round.

So how good is 88%? I took nine top games of nine world champions to compare with. Here is a summary of the results.

Lasker - Capablanca, St Petersburg 1914, 1-0. Lasker reaches 81%.
Nimzowitsch - Capablanca, St Petersburg 1914, 0-1. Capa reaches a record 84%.
Botvinnik - Keres, Moscow 1953, 1-0. Botvinnik makes 79%.
Fischer - Spassky, Reykjavik 1972, game 6, 1-0. Fischer makes 61%. Considered to be Fischers best effort in the match by many including Spassky - who applauded Fischer on stage after the game.
Karpov - Kasparov, Wch match 1993, game 17, 1-0. Karpov reaches 64%.
Karpov - Kasparov, Linares 1993, 0-1. Kasparov makes 76%.
Topalov - Anand, Wch match 2010, final game, 0-1. Anand makes 76%.
Carlsen - Anand, Master final 2012, 1-0. Carlsen makes 57% in this brilliant game.
Kramnik - McShane, London CC 2012, 1-0. Kramnik makes 68%.

A total of 249 moves when the known opening moves are subtracted. 181 are first choice moves for Stockfish. That makes 73%.

Ivanov is not playing world champion chess. He plays 15% better.

So how unlikely is it to play that well by pure chance? Let´s do some math.

If we simplify and say that in every position there is only two moves a good player (or computer) has to choose from. An oversimplification for sure, but the results are staggering. Then the chance to reach the world champions score 73% is in the order of one in 10 to the power of 21 (a one with 21 zeroes after it). To reach Ivanovs 88% you have one chance in 10 to the power of 53.

If we grant Ivanov world champion strength his chance of reaching 88% would be one in 10 to the power of 32.

To sum it up: there is conclusive proof in the moves themselves that Ivanov cheated. The proof is so overwhelming that it dwarfs any DNA test that is currently used in courts around the world.

Ivanov has left the fingerprint or DNA of a computer at the scene of the cheating.

dokter_nee's picture

Great research champ, thanks.

Thomas's picture

Funnily, this research could rather suggest that Ivanov is innocent? If he indeed did cheat on every single move after the opening, he could just as easily play the second-, third- or fifth-best move of Stockfish or any other engine whenever this is sufficient to keep a clear advantage.

If Ken Regan's approach was implemented for tournament games and accepted as sufficient _sole_ proof of cheating, what could happen? An actual cheater can "double-cheat" in such a way to make sure that his level of play is high enough to be unlikely, but not "impossibly high". It's impossible to manipulate a DNA fingerprint (which BTW in a recent case in the Netherlands would by itself still be insufficient for conviction in court), it's easy to manipulate an engine fingerprint. On the other hand, an innocent player would run a risk (slight but non-zero) to be wrongly convicted of cheating ... .

As to Carlsen-Anand, Bilbao 2012: this is just evidence that this game wasn't "too brilliant" by engine standards. It might have been hyped because many people like Carlsen, and many (often the same ones?) currently dislike Anand.

brabo's picture

Thomas you can't admit that the evidence is growing against Ivanov. Today I already see 3 strong pieces of evidence which are certainly together usable in court.
1) The testimony of the grandmaster which refers to Ivanovs cheating history
2) The calculated chance to produce for a 2227 player a result of almost 2700
3) The calculated chance to produce the number of computermoves compared with number of moves played for a player of the caliber of 2227.

DNA has 1/7000 as probability to be wrong. DNA on itself is not sufficient for conviction but combined with another piece of evidence, people get sentenced and put in jail.

Here we are talking about probabilities much lower to be wrong. We also possess several pieces.

I believe in court Ivanov could certainly lose the case against a cheating allegation. However is anybody willing to invest his money, time for such minor cheating? In the end we are talking about maybe maximum 1000 euro that he took out of the pricemoney and of course some ratingpoints. Personally I never go to court for 1000 euro even if am 100% sure that I am right. It is of course not good for chess that people can get away with cheating as it will cause the end of chess.

Thomas's picture

What exactly is the evidence?
1) Someone wants to be mentioned in a newspaper. Recently a policeman told me (in a rather different case, and nothing where I was personally affected or hurt) "we cannot investigate based on rumors circulating in pubs!". I assume and hope this also applies if a VIP spreads a rumor - and to me Klaric is hardly a VIP just because he got the GM title some 30 years ago. And it seems that even he only said that Ivanov was _accused of_ cheating, not that he cheated let alone how.
2) What are the odds that an amateur football team beats a strong professional team? Still it happened more than once. You may say that's just one game, but a football match can be considered a series of mini-games: each time one team attacks with the other one defending is a new game (then football is a very drawish sport).
3) More difficult to refute, but it still matters how difficult it is to find engine moves and here opinions diverge. I have games where Houdini likes about half of my moves (what's your percentage?), and at a recent occasion an opponent had to play amazingly bad chess to "find" several moves that weren't in Houdini's top10.

Altogether the evidence is, IMO, like finding someone guilty of murder because one person said "he did it!" without giving any details, he would have a reason and doesn't have an alibi. It would work for a show trial in a dictatorship where the culprit might be tortured until he confesses.

Of course one never knows what a regular court would decide. But Feller was successful in court despite stronger evidence against him (SMS messages and the exact method how he/they cheated).

brabo's picture

Thomas:
1)If Klaric says the same in a court then it will be considered as a piece of evidence. It is not sufficient but can be used as an important addition.
2) We are speaking about odds which are infinite small like winning euromillions or hit by lightning. Such kind of odds are used in a court as evidence.
3) The earlier mentioned study gives a clear correlation between number of moves corresponding with an engine and somebodies rating. From everybody a number of moves will correspond (like exchanging queens) but it is the percentage which is counting and the deviation in the games of this percentage. Again it is all about probabilities.

Thomas, you clearly want 100% evidence and this doesn't exist. It never exists even in murdercases.

On my blog I mentioned the following case http://schaken-brabo.blogspot.be/2012/07/afgesproken-resultaat-in-open-g..., The reaction of the organizers was that if Feller can't be punished with such strong evidence then why should they even bother to start up an investigation. I fully agreed with the organizers. It is a hopeless situation but not serious.

Thomas's picture

Klaric: "He was already accused of cheating at events in Bulgaria and Serbia"
Judge: "Which events?"
Klaric: "I don't remember"
Judge: "Who accused him?"
Klaric: "I don't remember"
Judge: "How did he cheat?"
Klaric: "No idea"

If this counts as evidence in court (even as "important addition") I lose faith in the legal system. And Klaric hasn't been more specific so far ... .
But OK I give up, I won't convince you.

P.S.: When I talk about similarities between my play and Houdini's moves, I obviously do not mean forced recaptures or anything a player with Elo 1200 could also play.

brabo's picture

Sure if GM Klaric can't back up his statements then it is indeed pretty useless. However I believe a grandmaster will not make such statements without any reference. What is the point of doing that? Afterall he didn't participate so didn't lose money neither rating. You maybe will tell me that he wants attention well in such case he will never of course make the same statements in a court.

Anyway we have to wait and see what will happen next. Many questions are left open.
1) Will GM Klaric come up with clear references of his accusations?
2) Will Ivanov keep playing chess?
3) Will he lose the 115 points won compared with his previous rating in the next tournaments or he will further improve?
4) If he keeps improving the rating, will he be finally caught?

Nobody can answer this and time will (probably) tell who is right.

Reader's picture

It does seem high improbable for such stats but according to research by Dan Ariely, he suggests that people often cheat by only small amounts as much as our conscience allows us, your analysis shows that he is cheating in huge amounts. If i were a player attempting to cheat i would definitely hide my cheating by playing masking it with bad moves.

Reader's picture

Perhaps we need the opinion of a statistician, i remember i once read that statistician can tell between a real series of coin tosses and a generic series of coin tosses by analysis because people often underestimate the improbable(e.g:5 heads in a row) and try to make it so realistic that it becomes obviosly fake.

Anonymous's picture

it is easier to say he played good.
are chessplayers cheaters?

Reader's picture

Well, i think you are over-generalizing. Every human being on earth has the capacity to cheat given the right incentives. Even if 99.99% of chessplayers play fair there will always be that 0.01% that do cheat so we must keep an eye out, although i do agree that we shouldn't be slandering without evidence

Chris's picture

the way of disputing that matter here suggests somehow that chessplayers are cheaters or that cheating in chess is usual

Septimus's picture

You are making a huge assumption that Stockfish's first choice is indeed the best move. If anything, I would say Carlsen's choices are better than Stockfish, so in fact inferior moves are being played by the so called "cheat".

To sum it up: It is dangerous and stupid to brand somebody a cheat based on NO EVIDENCE, but mere CONJECTURE.

Anonymous's picture

CAPS LOCK IS STUPID TOO

Septimus's picture

The anonymous idiot hath spoken!

RG13's picture

"I would say Carlsen's choices are better than Stockfish"

Really? Then I guess you would bet on him in a match against Stockfish? I'm glad you're not my broker.

Anonymous's picture

I completely agree.

Phil's picture

You convinced me that Capablanca had access to Stockfish too

Chris's picture

good point :)

RG13's picture

Verily I say unto you, one who is greater than Capablanca has come!

HR Sadeghi's picture

Forgive my political "incorrectness": chess is no sport and given the cheating possibilities, one should simply ban live online transmission of moves. That is but one of a,number of measures to avoid cheating with (or without!) outside help.

Raj's picture

Whatever be the reality, I enjoyed going through this well-played game. His performance in future tournaments would certainly indicate his strength. Tournament organizers can check whether they can install mobile signal "jammers" (used by security personnel for VIP security) that block mobile phone signals and thereby prevent cheating and any hand-held metal detector detects mobile phones on the body - no searches needed and no embarassing situations. "Prevention is better than cure."

Mic78's picture

Thanks the Champ for the conclusive analysis showing a huge statistical evidence for cheating

Chris's picture

by stattistics you can 'prove' whatever you wany,
statistics is not giving the prove but the probability of sth

RG13's picture

I think if the attitudes of people in this thread are representative of ogranizers and arbiters then chess as a serious 'sport' is finished. People want so much to believe that 'the next big thing' has come that they are willing to dismiss all kinds of circumstantial evidence that something is amiss. People cheat and whether or not Ivanov cheated in this case is immaterial to the importance of assuming people will if given the chance. The so called 'search' by the arbiter amounted to a 'cursory pat down'. His pockets were checked and he took off his shirt. I read nothing about getting him to move his long hair out of the way so that his ear canals could be peered into. Too embarrassing? Not if chess ever wants to be taken seriously as a sport.

ref:
old tech used to cheat way back in 2006
http://www.chessbase.com/newsdetail.asp?newsid=3292

Chris's picture

following your philosphy 'people are cheating' everybody shall be sent to the jail.
Do you cheat, too?

RG13's picture

I don't believe a human being has walked the earth who hasn't cheated to some extent at something or at least told a 'white lie' (which is a form of cheating). It all differs on the degree and the subsquent amount of damage.

Pages

Latest articles