Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
The Latest Cheating Scandal
Anyone else seen the latest from the Zadar Open?

2200 Ivanov had some smashing results:

<!-- m --><a class="postlink" href=""></a><!-- m -->

but his own countryman, FM Lilov seems to think he played very "houdini-ishly" (by very, I mean 95 to 97% of the time!!):

<!-- m --><a class="postlink" href=""></a><!-- m -->

Quite a long video but very intriguing indeed! Lilov is suggesting organisers in big tournaments invest in some anti-cheating devices... certainly a sad state of affairs if such a measure is deemed necessary.

I would like to think that any player can significantly and quickly improve with effort, hard work and belief. And that an in-form and improving 2200 could play to 2600 hundred strength. In fact we have seen this with our very own Alan Tate (Congratulations again for your efforts on getting that GM norm).
However, after watching Lilov's video it appears that maybe Ivanov's "success" has not been merited honestly...

Any thoughts?
Would be interesting to find out how he actually did it (if it's true). I mean you'd imagine doing some sort of search on him would have revealed something suspicious, but seemingly not. Funny that when the live stream stopped he went on to lose that particular game. I presume then that the moves he played after that live stream stopped were not as strong as engine moves, and so there is a clear link. To prove his guilt (or innocence) it'd be a good idea to ban the broadcast of his games at future events to see if his performance remains the same (or similar).
<!-- m --><a class="postlink" href=""></a><!-- m -->

Alan Tate recorded a 2641 performance and GM norm while rated 2234.
Here is a well-written and interesting summary of Lilov's video, posted by Chris Rice on the ECForum... The entire thread is at

"Here is a quick and dirty translation of all the best bits of the Lilov article:

"The consensus is that Ivanov played almost perfect chess and followed Houdini in at least 9 out of 10 cases, for each move it was always one of Houdini's top three proposals.

Of course, one may ask, why Ivanov has not won all the games and lost in the second round against GM Ognen Jovanovic? Lilov notes that Ivanov lost to Jovanovic despite almost exclusively following Houdini's recommendations. It was because the position was closed after the opening and we know that in closed positions the computer has been known to play badly. Lilov reveals that Ivanov lost that game by making errors in a totally drawn position.

The following position (you'll have to refer back to the diagram from the ChessBase article <!-- m --><a class="postlink" href=""></a><!-- m --> - ed) Black played 115th ..Bd6? and with the simple white 116.Nf4 it was clear that Black had overlooked the threat of losing the pawn at d5. White then played 116… Bxf4 117.Kxf4 with an easily won pawn ending for White. Lilov explains in the diagram position that Black should have moved his king with 115 …Ke6 116 Nf6 d6 with an easy draw.
In other words, Black had been playing this game almost perfectly, not making any mistakes and then overlooked a simple threat, which forced him to transition into a lost pawn ending. Lilov explained this oversight was probably caused by lack of time and assumes that the correct continuation in the hectic closing stages could not be delivered.

But Ivanov hit back immediately and in round 3 he destroyed GM Bojan Kurajica with brilliant play on two fronts, which was crowned by a tactically precise mating attack. Lilov explains that Houdini is again almost always of the same opinion as Ivanov. The remarkable thing about this game is, however, above all, that Ivanov has played "chaotic, illogical trains of thought" based on hidden tactical options and precise calculation. Lilov also feels its conspicuous that Black, a 2520 GM, contributed almost no resistance.

Round four was followed by a draw against Davorin Kuljasevic, again with striking parallels to Houdini. However, the game was simplified right after the opening so that Black had little tactical ways to outwit his opponent.

In rounds five and six, followed devastating victories against Zelcic and Kozul, both were beaten by mating attacks. And again there was a strong agreement with the proposals by Houdini. Lilov remarks how amazingly accurate and precise Ivanov ‘s play is. Even the best players choose their moves intuitively and this intuition is prone to error.

After round seven, Ivanov is investigated by the tournament committee and following this in round eight the game against Borki Predojevic, in contrast to the previous games, is not broadcast live on the Internet. Lilov shows that Ivanov shows none of the preciseness and creativity from previous rounds and loses pathetically.

In the ninth and final round Ivanov wins in a game then rich with incident and tactically precise against GM Ivan Saric and ends up fourth in the tournament.

Lilov concludes that "more than 95 to 97% of Ivanov's trains of thought are trains of thought that would be suggested by very powerful engines which no man is capable of.

But Lilov is not content to analyse just Zadar, but digs deeper - and reveals more suprises. He looks at the chess career of Ivanov and notes that the Bulgarian had difficulties up to September 2012 of even winning one game against a player rated more than 2200 Elo. Take the first round of the Balkan Belogradchik Chess Festival 2012. Ivanov loses to a player rated 1916 Elo. But then comes the great metamorphosis: Ivanov is transformed overnight into a dangerous and strong player and wins 7/7.

Lilov has no doubt that Ivanov had computer help. At the end of this fascinating video, Lilov impresses with precise analysis and a deep understanding of chess and follows with a further surprise: Lilov shows what tools Ivanov could have taken advantage of: small cameras that can transmit the position to a partner anywhere in the world who is watching and analysing on his computer and a miniature microphone that can be hidden so well and deep in the ear, it cannot be discovered.

Lilov explains, both the camera and microphone in Bulgaria can be bought for about 50 Euro and are a popular tool by students in examinations. To prevent cheating in chess tournaments, Lilov suggests that tournament organizers should have a transmitter that emits noise and prevents the electronic signals."
And if Alan is a cheat, I'm the king of Scotland...
"How sad to see, what used to be, a model of decorum and tranquility become like any other sport, a battleground for rival ideologies to slug it out with glee"
Why would jamming signals be needed - could the moves not just be transmitted a move behind?

We might find out soon how it was done, if he was cheating I’d expect him to own up sooner or later - how could he function as a chess person, he couldn’t even go over the games afterwards, as dropping pawns and pieces everywhere would give the show away. Maybe he just wants a quick GM title to put on his CV.

It's just lucky he's not been able to set a more believable level to cheat at though :-(
WBuchanan Wrote:Why would jamming signals be needed - could the moves not just be transmitted a move behind?

We might find out soon how it was done, if he was cheating I’d expect him to own up sooner or later - how could he function as a chess person, he couldn’t even go over the games afterwards, as dropping pawns and pieces everywhere would give the show away. Maybe he just wants a quick GM title to put on his CV.

It's just lucky he's not been able to set a more believable level to cheat at though :-(

Interesting point Walter! What if a cheater set Rybka/Houdini/whatever to 2400 level? I'm guessing this could be done very simply?! Jumping from 2200-ish to 2400 IM level isn't as remarkable as 2200 to 2600+ but would still be very 'worthwhile' to many. The latest Houdini-related article on ChessBase has it rated 3300+ Sad

Some of the earlier cheating escapades showed up the stupidity of the cheaters - the 'von Neumann' guy apparently didn't even know how the pieces moved, he just followed the co-ordinates. Must have been amusing to see him move the horsey one square at a time like a child, following the L-shape, while beating GM's Big Grin
WBuchanan Wrote:Why would jamming signals be needed - could the moves not just be transmitted a move behind?

My understanding is that the cheater is directly transmitting, to an outside source, a view of the board after the opponent has moved. The organiser's transmission isn't playing any part in this case. Hence the proposal is to do something to interfere with the cheater's transmitting equipment.
My issue with all this is that nothing has been proven. The guy has played exceptionally well but it is not impossible for players to play 400+ above their rating as has already been mentioned Alan has done so.

So often in life we are easily deceived by false statistics - often because they are incorrect. If the odds PRIOR to an event are 1 in 10,000 that a specific player scores as this player did then if you are looking with hindsight you are really asking the question what are the odds of ANY 2200+ player putting in such a performance. If there are 1000 such players playing 10 events a year then the odds are 1 i.e. a certainty.

A similar situation arose in connection with the New York lottery when one lady one it twice in one year (or some short period) and the newspapers focused on it saying that the odds of anybody winning it once were so poor that to win it twice must be impossible and something fishy must have happened. As it happened a statistician investigated it and based on the simple fact that any one winning the lottery twice would be newsworthy i.e. it did not have to be a specified person who only played the game those two times, they forecast that the odds for a winner to win a second time were not insignificant and subsequent studies of past results proved this to be correct.

Someone has done an analysis of his games and, having found excuses to ignore the games he lost, states that a high percentage of his moves are the top candidate moves for a certain engine. This is meant to be proof of cheating. Has anyone done a similar comparison with other players to see how their move % compares and whether players who have a hot event appear engine like?

My concern here is that a player's name is being pulled through the mud based on nothing more than poor statistics. He may be guilty but produce some actual evidence, not gossip.
Although I agree to a certain extent about the gossip factor and the allegations being unproven (ChessBase can be sloppy this way sometimes) I think you'll probably find that the odds of such an event actually happening is much higher than the figures you have used (1 in 10,000) so his result is much more impressive 'statistically'?!

Almost all of the '2200-player producing a 2600+ result' occurrences come from players who have already shown significant improvements in previous events; strong juniors on their way up is an example. Our own Alan Tate had already shown signs that he was at least capable of such a breakthrough before his Croatian results, whereas Mr. Ivanov hadn't shown anything of the sort.

His engine match-up is way higher than would be expected from even strong GM's having a 'hot event' and having played through all of his games from the event it most definitely did not appear that this was a 2200 player on a lucky/improvement streak - the way he dispatched most of his GM opponent's was seriously impressive.
As a player of a similar rating to Ivanov, and one who follows/studies/prepares for opponents of 2200 strength on a regular basis, I really don't believe this was an unaided 2200 player. There weren't excuses given for his losses, just reasonable possible explanations. I should point out though that losing games immediately after being accused of cheating isn't a new thing (look up the well-known Mamedjarov-Kurnosov incident from a few years ago for a previous example).

The question of proof is a difficult one - the technology exists to do what he has been accused of, almost impossible to discover without an intimate search (not quite rubber gloves but that day might come around!)

What level of proof would be satisfactory?

Forum Jump:

Users browsing this thread: 1 Guest(s)