Tiger Lilov's video about the Zadar Open 2012 motivated me to look at the relevant data in some detail. The issue involves a non-grandmaster player (NGM) who played against grandmasters (GM) with such surprising strength that many people wondered whether he had an unfair advantage, possibly in the way of computer assistance.

I took a sample of moves from games 58, 75, 87, 100, 117, and 136 in which the NGM played GM. The NGM won three of these games and lost one. The other two ended in draw by mutual agreement. The question raised by Tiger concerns the use of computer generated moves - specifically moves recommended by Houdini 3 Pro. Since I don't have access to Houdini 3 Pro, I used Houdini 3 and collected data from the tournament PGN file provided by chessbase.com. The sample data are summarized below.

To exclude rote opening moves, I tallied the data from move 10 in each game. I counted 321 moves in these six games - 160 by the NGM and 161 by the GM players. Of the 321 moves I identified 195 as "computer moves" because each move coincided exactly with the move recommended by Houdini at a depth of 19 to 20 half moves. The other 126 moves are likewise identified as "non computer moves". So we find that almost 2/3 of the moves in all six matches, 60.7% to be exact, are coincident with Houdini recommendations.

Of the 160 moves made by the NGM 115, or 71.9% , coincided with Houdini moves while the 161 moves made by GM contain only 80 Houdini moves at the considerably lower rate of 49.7%. We can carry out a statistical test to determine whether the observed difference in the rate of computer moves could have happened by chance in the small sample of moves taken.

The Chi-Square test shows that if the GM and the NGM had the same propensity to make Houdini moves, the probability of realizing the observed difference in this sample is 0.3%. In non medical statistics, we normally consider a probability of less than 5% to be something we would not expect to happen. So in this case, the sample data show that there is in fact a real difference in the computer move rate between the NGM and the GM.  In particular, we conclude that the computer move rate of the NGM is higher than that of the GM.

The additional question raised by Tiger is whether the NGM's computer move rate is unusually high. To answer that question I took a sample of moves from the 2012 world championship between Anand and Gelfand counting from move number 10 to move 29 or the end of the game if the game was shorter than 29 moves. Of 398 moves in my sample, 280 or 70.4% were computer moves with Anand slightly higher at 73.5% and Gelfand somewhat lower at 67.2%. In this context, we can see that a computer move rate of 71.9% is actually as unusually high as Tiger had suspected as it is comparable to the rate of the reigning world champion.

Incidentally, the positive relationship between player rating and rate of computer moves shows that computer moves are not an aberration but consistent with good chess and the high rate of Houdini moves by the NGM may imply only that he was simply playing well.

Another question raised is the the occurrence of long sequences of consecutive Houdini moves by the NGM. The longest sequence of Houdini moves by the NGM in my sample is 14 moves. There are two of these sequences - one starts from move 15 in game 75 and the other from move 13 in game 87. The longest sequence of such moves by the GM is 4 moves and these occur in all the games. These sequences often occur because of a series of exchanges. I found 40 instances of consecutive Houdini moves in the sample, 20 by the NGM and 20 by the GM. The average length of these sequences for the NGM is 5.15 moves. The average length for the GM is almost half that at 2.75.

The t-statistic for this difference is more than 12 which implies that the observed difference could not have occured in our sample if there were not a real difference in the propensity to make sequences of Houdini moves. We conclude that the data provide sufficient evidence that the NGM has a higher propensity to make sequences of Houdini moves - just as Tiger had suspected.

By way of comparison, the average length of Houdini move sequences in the Anand-Gelfand 2012 world championship match is 4.17 (Anand 4.19, Gelfand 4.15). The NGM's average sequence length of 5.15 exceeds the length achieved in the world championship by about one move.

The data seem to show that the NGM tends to play a lot like Houdini. As to the question of how the NGM developed this ability, we really have no actual data to go on and thus all we can do is speculate. It could be that the NGM has played so many games with Houdini that he developed a Houdini-like pattern in his moves. It is also speculated that the NGM used devious high-tech methods to access Houdini during the match and that therefore he enjoyed an unfair advantage over the GM.

Yet, although there is insufficient evidence to conclude that the NGM used any of the high tech cheating devices described by Tiger, the important message in Tiger's videos may be that these devices exist and they are not very expensive; and that alone should be sufficient reason to install the cell phone jammers in tournament halls that Tiger has proposed.

Cha-am Jamal

Thailand

• 2 years ago

seems ivanov has been banned from playing for 4 months by bulgarian chess federation whilst they use some scientific means to check if he did cheat. the wheels of chess justice turn slowly but is on the move now it seems :).

• 3 years ago

frankBgambit

brilliant

thank you

• 3 years ago

prof regan has published his Analise of  Borislav Ivanov zadar games  you can find here:

http://www.cse.buffalo.edu/~regan/chess/fidelity/data/Mercurial.txt

his estimate for inanovs performance is between 3089 - 3120 IPR(elo)

with my estimate with my own system of 3107 ipr performance im quiet pleased to see that my figure is inline with prof regans estimate of this performance.

and thought ppl might want check out prof regans work on this case.

• 3 years ago

hello frankBgambit, what an excellent piece of work you have done. thank you for sharing your findings. i will go over your report more carefully tomorrow and write to you with some hopefully helpful feedback. i would like to learn more about the ipr. let's stay in touch.

• 3 years ago

I ran these games through my system which is based on prof Regans model and summary for all games performance based on moves played is as follows

Ivanov, Borislav    Top moves    (156/258)    60.5% mean error    0.054    IPR(ELO) 3107
Ivanov Opponents  Top moves    (104/259)    40.2%     mean error 0.234 IPR(ELO)    2226

this is for all moves other than first 8 moves and moves that players forced to play or in drawn positions.

IPR is what prof Regan calls intrinsic performance Rating.

I would also like point out game number 8 which is game where live braodcasting of the games was stopped by the tornement directors,

Ivanov, Borislav (White)- Accuracy 93% Blunders 0 (0%.) Mistakes 1 (4.8%.) Sub Opt 1 (4.8%.)Exp-Px 0.1 Obs-Px 0.42 Exp IPR 1818
Raw Err= 0.199-19%(11/26), Book Err= 0.091-13.3%(2/15), middle Err= 0.251-28.6%(4/14
White Candidate Moves= Top1- 19% (4/21)-Top2- 52.4% (11/21)-Top3- 66.7% (14/21)-Top4- 66.7% (14/21)

Predojevic, Borki (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.9 Obs-Px 0.58 Exp IPR 2855
Raw Err= 0.057-42.3%(11/26), Book Err= 0.109-20%(3/15)), Middle Err= 0.048-52.63% (10/19), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 42.3% (11/26)-Top2- 61.5% (16/26)-Top3- 76.9% (20/26)-Top4- 84.6% (22/26)

with a performance IPR of only 1818 with a poor top move rate of 19% to me this is overwhelming evidense that this one factor of stopping the broadcast is main effect on outcome of this game.

of all the games in history that i have checked this performace for tornement is highest i have ever found for any player ever. combined with the dramatic change to the performance of Ivanov once the broadcasting of moves was stopped.

As for the consecutive houdini moves made by Ivanov and his opponents:

Ivanov, Borislav consecutive highest consecutive engine top moves was:

5 ,3,13,3,13,8,5,3,10   with average of 7.0 consecutive moves per game.

Ivanov opponent's consecutive highest consecutive engine top moves was:

6,4,5,8,3,2,3,3       with average of 4.333 consecutive moves per game.

the numbers of indicator's of engine assistance to play these games is overwhelming .

and for game by game breackdown for performance is as follows:
____________________________________________________________________

[Event "19th Open A"]
[Date "2012.12.16"]
[Round "1.18"]
[White "Ivanov, Borislav"]
[Black "Schachinger, Mario"]
[Result "1-0"]
[ECO "E92"]
[WhiteElo "2227"]
[BlackElo "2426"]
[PlyCount "69"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Be2 e5 6. Nf3 O-O 7. dxe5 dxe5 8. Qxd8 Rxd8 9. Bg5 c6 10 Nxe5 Re8 11 O-O-O Na6 12 Rd6 Be6 13 f4 Nc5 14 Bf3 h6 15 Bh4 Nfd7 16 Nxd7 Nxd7 17 Kc2 Nf8 18 b3 g5 19 Bg3 gxf4 20 Bxf4 Ng6 21 Be3 Ne5 22 Rhd1 f5 23 Bh5 Nf7 24 R6d2 Re7 25 Bc5 Rc7 26 Rf2 fxe4 27 Nxe4 Kh8 28 Nf6 b6 29 Ba3 c5 30 Re1 Nd8 31 Bb2 Re7 32 Nd5 Bf5+ 33 Rxf5 Rxe1 34 Rf8+ Kh7 35 Nf6+

Ivanov, Borislav (White)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0 (0%.)Exp-Px 0.24 Obs-Px 0.65 Exp IPR 3240
Raw Err= 0.041-66.7%(16/24), Book Err= 0.082-46.7%(7/15), middle Err= 0.041-70.6%(12/17),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 66.7% (16/24)-Top2- 87.5% (21/24)-Top3- 95.8% (23/24)-Top4- 95.8% (23/24)
Most Consectutive Candiate moves made 5 from move 26. Rf2 to move 30. Re1-NQ moves-open:8 Drawn:0 Forced:3 NQval:0

Schachinger, Mario (Black)- Accuracy 81% Blunders 2 (8.3%.) Mistakes 1 (4.2%.) Sub Opt 3(12.5%.) Exp-Px 0.76 Obs-Px 0.35 Exp IPR 1802
Raw Err= 0.547-37.5%(9/24), Book Err= 0.135-42.9%(6/14)), Middle Err= 0.728-35.29% (6/17), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 37.5% (9/24)-Top2- 58.3% (14/24)-Top3- 70.8% (17/24)-Top4- 79.2% (19/24)
Most Consectutive Candiate moves made 2 from move 15.. Nfd7 to move 16.. Nxd7-NQ moves-open:8 Drawn:0 Forced:2 NQval:0

[Event "19th Open A"]
[Date "2012.12.17"]
[Round "2.7"]
[White "Jovanic, Ognjen"]
[Black "Ivanov, Borislav"]
[Result "1-0"]
[ECO "D45"]
[WhiteElo "2583"]
[BlackElo "2227"]
[PlyCount "235"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. e3 e6 5. Nf3 Nbd7 6. Be2 dxc4 7. a4 b6 8. Bxc4 Bb7 9. e4 Be7 10 O-O O-O 11 e5 Nd5 12 Nxd5 cxd5 13 Bd3 Nb8 14 Bd2 Nc6 15 Qb1 h6 16 b4 Rb8 17 Qb3 a5 18 b5 Nb4 19 Be2 Qd7 20 Rfc1 Rbc8 21 h4 Rxc1+ 22 Rxc1 Rc8 23 Rc3 Rxc3 24 Qxc3 Qd8 25 g3 Na2 26 Qb3 Nb4 27 Qc3 Na2 28 Qb2 Nb4 29 Kg2 Qc7 30 Qc1 Qxc1 31 Bxc1 Kf8 32 Bd2 Nc2 33 Bd3 Na3 34 Bc1 Nc4 35 Kf1 Bb4 36 Ke2 Bc3 37 Bc2 h5 38 Kd3 Ba1 39 Bb3 Nb2+ 40 Kc2 Nc4 41 Be3 Na3+ 42 Kd3 Nc4 43 Bg5 Nb2+ 44 Ke2 Nc4 45 Kd3 Nb2+ 46 Kc2 Nc4 47 Be3 Na3+ 48 Kd3 Nc4 49 Ne1 Nxe3 50 fxe3 Bb2 51 Bd1 g6 52 Nc2 Ke7 53 Bf3 f6 54 exf6+ Kxf6 55 Bg2 e5 56 Bh3 e4+ 57 Ke2 Ke7 58 Ne1 Ba3 59 Ng2 Bd6 60 Nf4 Kf6 61 Kf2 Bb8 62 Ne2 Ke7 63 Kg2 Kf6 64 Nf4 Bd6 65 Ne2 Ke7 66 Kf2 Kf6 67 g4 hxg4 68 Bxg4 Ke7 69 Nf4 Kf6 70 Bd7 Ke7 71 Bg4 Kf6 72 Bd7 Ke7 73 Bh3 Kf6 74 Ke2 g5 75 hxg5+ Kxg5 76 Ne6+ Kf6 77 Nf4 Bc7 78 Kf2 Bb8 79 Bd7 Ke7 80 Bg4 Bc7 81 Ne2 Bd8 82 Kg3 Kf7 83 Nf4 Bc7 84 Bf5 Kf6 85 Kg4 Bd6 86 Ne2 Ba3 87 Nf4 Bd6 88 Nh5+ Ke7 89 Kg5 Bh2 90 Nf6 Bd6 91 Ng8+ Kf7 92 Nh6+ Ke7 93 Kg6 Bb4 94 Ng8+ Kd8 95 Kg5 Be1 96 Be6 Bf2 97 Kf4 Kc7 98 Nf6 Kd6 99 Bf5 Bh4 10 Nd7 Bd8 10 Ne5 Bf6 10 Nf7+ Ke7 10 Nh6 Bh4 10 Bh3 Kf6 10 Ng4+ Ke6 10 Ne5+ Kd6 10 Bd7 Be7 10 Bg4 Bf8 10 Kf5 Bc8+ 11 Kg6 Bxg4 11 Nxg4 Ke6 11 Nf2 Bb4 11 Kg5 Be7+ 11 Kg4 Kf6 11 Nh3 Bd6 11 Nf4 Bxf4 11 Kxf4 Ke6 11 Kg5

Jovanic, Ognjen (White)- Accuracy 97% Blunders 0 (0%.) Mistakes 2 (3.2%.) Sub Opt 0 (0%.)Exp-Px 0.89 Obs-Px 0.51 Exp IPR 2564
Raw Err= 0.078-46.8%(29/62), Book Err= 0.102-6.7%(1/15), middle Err= 0.06-70%(7/10),, Endgame Err= 0.088-48.9%(22/45).
White Candidate Moves= Top1- 46.8% (29/62)-Top2- 62.9% (39/62)-Top3- 71% (44/62)-Top4- 74.2% (46/62)
Most Consectutive Candiate moves made 5 from move 16. b4 to move 20. Rfc1-NQ moves-open:8 Drawn:45 Forced:3 NQval:0

Ivanov, Borislav (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.11 Obs-Px 0.49 Exp IPR 3243
Raw Err= 0.021-67.6%(48/71), Book Err= 0.122-28.6%(4/14)), Middle Err= 0.035-58.33% (7/12), Endgame Err= 0.088-73.1%(38/52).
Black Candidate Moves= Top1- 67.6% (48/71)-Top2- 76.1% (54/71)-Top3- 81.7% (58/71)-Top4- 90.1% (64/71)
Most Consectutive Candiate moves made 3 from move 17.. a5 to move 19.. Qd7-NQ moves-open:8 Drawn:35 Forced:3 NQval:0

[Event "19th Open A"]
[Date "2012.12.17"]
[Round "3.12"]
[White "Ivanov, Borislav"]
[Black "Kurajica, Bojan"]
[Result "1-0"]
[ECO "E11"]
[WhiteElo "2227"]
[BlackElo "2565"]
[PlyCount "69"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 Nf6 2. c4 e6 3. g3 Bb4+ 4. Bd2 c5 5. Bxb4 cxb4 6. Bg2 O-O 7. Nf3 d6 8. O-O Qe7 9. a3 bxa3 10 Rxa3 b6 11 Nc3 Bb7 12 d5 e5 13 Nh4 g6 14 Qd2 Nh5 15 Qh6 f5 16 e4 Ng7 17 exf5 gxf5 18 Nb5 Rf6 19 Qg5 Qf7 20 Rxa7 Rxa7 21 Nxa7 f4 22 Ra1 Na6 23 Nc6 Bc8 24 Nf5 Bxf5 25 Rxa6 h6 26 Qh4 Bd3 27 Rxb6 e4 28 Rb7 Qxb7 29 Qxf6 e3 30 fxe3 fxe3 31 Ne7+ Kh7 32 Qf8 h5 33 Qg8+ Kh6 34 Qh8+ Bh7 35 Be4

Ivanov, Borislav (White)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0 (0%.)Exp-Px 0.13 Obs-Px 0.67 Exp IPR 3288
Raw Err= 0.029-65%(13/20), Book Err= 0.042-21.4%(3/14), middle Err= 0.027-84.6%(11/13),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 65% (13/20)-Top2- 85% (17/20)-Top3- 95% (19/20)-Top4- 95% (19/20)
Most Consectutive Candiate moves made 13 from move 23. Nc6 to move 35. Be4-NQ moves-open:8 Drawn:0 Forced:7 NQval:0

Kurajica, Bojan (Black)- Accuracy 86% Blunders 1 (4.5%.) Mistakes 1 (4.5%.) Sub Opt 2(9.1%.) Exp-Px 0.87 Obs-Px 0.33 Exp IPR 1630
Raw Err= 0.623-40.9%(9/22), Book Err= 0.094-14.3%(2/14)), Middle Err= 0.87-53.33% (8/15), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 40.9% (9/22)-Top2- 63.6% (14/22)-Top3- 81.8% (18/22)-Top4- 86.4% (19/22)
Most Consectutive Candiate moves made 5 from move 18.. Rf6 to move 22.. Na6-NQ moves-open:8 Drawn:0 Forced:4 NQval:0

[Event "19th Open A"]
[Date "2012.12.18"]
[Round "4.7"]
[White "Kuljasevic, Davorin"]
[Black "Ivanov, Borislav"]
[Result "1/2-1/2"]
[ECO "D85"]
[WhiteElo "2561"]
[BlackElo "2227"]
[PlyCount "67"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 Nf6 2. c4 g6 3. Nc3 d5 4. cxd5 Nxd5 5. e4 Nxc3 6. bxc3 Bg7 7. Nf3 c5 8. Be3 Qa5 9. Qd2 Nc6 10 Rb1 a6 11 Rc1 cxd4 12 cxd4 Qxd2+ 13 Kxd2 f5 14 e5 h6 15 h4 Be6 16 Bc4 Bxc4 17 Rxc4 O-O 18 Ke2 Rad8 19 Rb1 Rd7 20 Rb6 e6 21 g3 Rfd8 22 Rc1 Re7 23 Rc4 Red7 24 Rc3 Rc7 25 Rc4 Rcd7 26 Nd2 Nxd4+ 27 Bxd4 Rxd4 28 Rxd4 Rxd4 29 Rxe6 Kf7 30 Rb6 Bxe5 31 Rxb7+ Ke6 32 Rh7 Rg4 33 Rxh6 Kf7 34 h5

Kuljasevic, Davorin (White)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0 (0%.)Exp-Px 0.87 Obs-Px 0.5 Exp IPR 2886
Raw Err= 0.048-50%(11/22), Book Err= 0.055-46.7%(7/15), middle Err= 0.021-55.6%(5/9),, Endgame Err= 0.052-66.7%(4/6).
White Candidate Moves= Top1- 50% (11/22)-Top2- 68.2% (15/22)-Top3- 77.3% (17/22)-Top4- 77.3% (17/22)
Most Consectutive Candiate moves made 4 from move 16. Bc4 to move 19. Rb1-NQ moves-open:8 Drawn:3 Forced:1 NQval:0

Ivanov, Borislav (Black)- Accuracy 98% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 1(4.3%.) Exp-Px 0.13 Obs-Px 0.5 Exp IPR 2849
Raw Err= 0.051-47.8%(11/23), Book Err= 0.035-53.3%(8/15)), Middle Err= 0.038-36.36% (4/11), Endgame Err= 0.052-40%(2/5).
Black Candidate Moves= Top1- 47.8% (11/23)-Top2- 56.5% (13/23)-Top3- 69.6% (16/23)-Top4- 73.9% (17/23)
Most Consectutive Candiate moves made 3 from move 10.. a6 to move 12.. Qxd2+-NQ moves-open:8 Drawn:0 Forced:2 NQval:0

[Event "19th Open A"]
[Date "2012.12.19"]
[Round "5.7"]
[White "Ivanov, Borislav"]
[Black "Zelcic, Robert"]
[Result "1-0"]
[ECO "A45"]
[WhiteElo "2227"]
[BlackElo "2560"]
[PlyCount "63"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 Nf6 2. Bg5 Ne4 3. Bf4 c5 4. f3 Qa5+ 5. c3 Nf6 6. d5 Qb6 7. b3 e6 8. e4 exd5 9. exd5 d6 10 Bc4 Be7 11 Ne2 Nh5 12 Be3 O-O 13 O-O f5 14 Nd2 Nd7 15 b4 Qc7 16 Re1 cxb4 17 cxb4 Ne5 18 Rc1 Bh4 19 g3 f4 20 Nxf4 Nxf4 21 Bxf4 Rxf4 22 gxf4 Nxc4 23 Re8+ Kf7 24 Qe2 b5 25 Nxc4 bxc4 26 Rxc4 Qb6+ 27 Kh1 Bd7 28 Rxa8 Qb7 29 Rh8 h6 30 Qe4 Bf6 31 Rc1 Qa6 32 Rg1

Ivanov, Borislav (White)- Accuracy 94% Blunders 0 (0%.) Mistakes 1 (5.6%.) Sub Opt 0 (0%.)Exp-Px 0.13 Obs-Px 0.64 Exp IPR 3004
Raw Err= 0.132-66.7%(12/18), Book Err= 0.047-53.3%(8/15), middle Err= 0.176-81.8%(9/11),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 66.7% (12/18)-Top2- 77.8% (14/18)-Top3- 88.9% (16/18)-Top4- 88.9% (16/18)
Most Consectutive Candiate moves made 13 from move 17. cxb4 to move 29. Rh8-NQ moves-open:8 Drawn:0 Forced:6 NQval:0

Zelcic, Robert (Black)- Accuracy 80% Blunders 1 (5%.) Mistakes 2 (10%.) Sub Opt 2(10%.) Exp-Px 0.87 Obs-Px 0.36 Exp IPR 1510
Raw Err= 0.436-45%(9/20), Book Err= 0.065-53.3%(8/15)), Middle Err= 0.633-38.46% (5/13), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 45% (9/20)-Top2- 70% (14/20)-Top3- 80% (16/20)-Top4- 90% (18/20)
Most Consectutive Candiate moves made 3 from move 10.. Be7 to move 12.. O-O-NQ moves-open:8 Drawn:0 Forced:3 NQval:0

[Event "19th Open A"]
[Date "2012.12.19"]
[Round "6.2"]
[White "Kozul, Zdenko"]
[Black "Ivanov, Borislav"]
[Result "0-1"]
[ECO "A37"]
[WhiteElo "2638"]
[BlackElo "2227"]
[PlyCount "68"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. c4 c5 2. Nf3 Nc6 3. Nc3 g6 4. g3 Bg7 5. Bg2 e6 6. d3 Nge7 7. h4 h6 8. Bd2 d5 9. Qc1 b6 10 O-O Bb7 11 Rb1 d4 12 Na4 Qc7 13 a3 a5 14 Qc2 O-O 15 b4 axb4 16 axb4 Nxb4 17 Bxb4 cxb4 18 Rxb4 Nd5 19 Rbb1 Nc3 20 Nxc3 dxc3 21 e3 Ra3 22 Rb3 Rfa8 23 d4 Qxc4 24 Nd2 Qc7 25 Bxb7 Rxb3 26 Nxb3 Qxb7 27 Qxc3 Qd5 28 Qc2 Ra3 29 Rb1 Bf8 30 Nc1 b5 31 Nd3 Bd6 32 Nc5 Bxg3 33 fxg3 Rxe3 34 Kh2 Qf3

Kozul, Zdenko (White)- Accuracy 88% Blunders 1 (3.8%.) Mistakes 1 (3.8%.) Sub Opt 2 (7.7%.)Exp-Px 0.91 Obs-Px 0.45 Exp IPR 1697
Raw Err= 0.275-38.5%(15/23), Book Err= 0.117-13.3%(2/15), middle Err= 0.319-52.6%(10/19),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 38.5% (10/26)-Top2- 53.8% (14/26)-Top3- 61.5% (16/26)-Top4- 65.4% (17/26)
Most Consectutive Candiate moves made 6 from move 22. Rb3 to move 27. Qxc3-NQ moves-open:8 Drawn:0 Forced:0 NQval:0

Ivanov, Borislav (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.09 Obs-Px 0.55 Exp IPR 3189
Raw Err= 0.028-65.2%(15/23), Book Err= 0.054-33.3%(5/15)), Middle Err= 0.022-75% (12/16), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 65.2% (15/23)-Top2- 87% (20/23)-Top3- 91.3% (21/23)-Top4- 91.3% (21/23)
Most Consectutive Candiate moves made 8 from move 15.. axb4 to move 22.. Rfa8-NQ moves-open:8 Drawn:0 Forced:3 NQval:0

[Event "19th Open A"]
[Date "2012.12.20"]
[Round "7.1"]
[White "Sumets, Andrey"]
[Black "Ivanov, Borislav"]
[Result "1/2-1/2"]
[ECO "A61"]
[WhiteElo "2638"]
[BlackElo "2227"]
[PlyCount "83"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 Nf6 2. c4 c5 3. d5 e6 4. Nc3 exd5 5. cxd5 d6 6. Nf3 g6 7. Nd2 Bg7 8. Nc4 O-O 9. Bf4 Ne8 10 g3 Qe7 11 Bg2 Nd7 12 O-O Ne5 13 Bxe5 Bxe5 14 Nxe5 Qxe5 15 Qd2 Bd7 16 a4 Nf6 17 e4 Rfe8 18 Rad1 Qh5 19 f3 Kg7 20 b3 a6 21 a5 Rac8 22 Rfe1 h6 23 Qb2 Rb8 24 f4 b5 25 Nb1 Bg4 26 Rc1 Bf3 27 Nd2 Bxg2 28 Kxg2 g5 29 Rf1 g4 30 b4 c4 31 Qd4 Qg6 32 Rce1 Kg8 33 Rf2 h5 34 e5 Qd3 35 Qxd3 cxd3 36 Ne4 Nxe4 37 Rxe4 dxe5 38 Rxe5 Rxe5 39 fxe5 Re8 40 Rd2 Rxe5 41 Rxd3 Kf8 42 d6

Sumets, Andrey (White)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0 (0%.)Exp-Px 0.91 Obs-Px 0.5 Exp IPR 2641
Raw Err= 0.066-28.1%(9/32), Book Err= 0.076-40%(6/15), middle Err= 0.07-22.7%(5/22),, Endgame Err= 0-66.7%(2/3).
White Candidate Moves= Top1- 28.1% (9/32)-Top2- 59.4% (19/32)-Top3- 71.9% (23/32)-Top4- 78.1% (25/32)
Most Consectutive Candiate moves made 3 from move 14. Nxe5 to move 16. a4-NQ moves-open:8 Drawn:0 Forced:2 NQval:0

Ivanov, Borislav (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.09 Obs-Px 0.5 Exp IPR 3233
Raw Err= 0.024-62.5%(20/32), Book Err= 0.044-73.3%(11/15), Middle Err= 0.032-50% (11/22), Endgame Err= 0-100%(3/3).
Black Candidate Moves= Top1- 62.5% (20/32)-Top2- 81.2% (26/32)-Top3- 81.2% (26/32)-Top4- 84.4% (27/32)
Most Consectutive Candiate moves made 5 from move 10.. Qe7 to move 14.. Qxe5-NQ moves-open:8 Drawn:0 Forced:1 NQval:0

(Game below is when broadcasting of live moves was stopped and dramatic change to Ivanov performance figures? )

[Event "19th Open A"]
[Date "2012.12.21"]
[Round "8.1"]
[White "Ivanov, Borislav"]
[Black "Predojevic, Borki"]
[Result "0-1"]
[ECO "D23"]
[WhiteElo "2227"]
[BlackElo "2600"]
[PlyCount "68"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. Qc2 dxc4 5. Qxc4 Bg4 6. Ne5 Be6 7. Qd3 Nbd7 8. Nf3 Bg4 9. Nc3 e6 10 g3 Be7 11 Bg2 O-O 12 O-O Rc8 13 Rd1 Qa5 14 Bd2 Rfd8 15 h3 Bh5 16 Qc4 b5 17 Qb3 Qb6 18 g4 Bg6 19 Bf4 a5 20 a3 a4 21 Qa2 Bc2 22 Rd2 Bb3 23 Qb1 c5 24 e3 b4 25 axb4 Qxb4 26 Ne5 Nxe5 27 Bxe5 Nd7 28 Bg3 Nb6 29 Qe4 Nc4 30 Re2 cxd4 31 exd4 Nxb2 32 Rxb2 Qxc3 33 Qe2 Qxd4 34 Be5 Qd3

Ivanov, Borislav (White)- Accuracy 93% Blunders 0 (0%.) Mistakes 1 (4.8%.) Sub Opt 1 (4.8%.)Exp-Px 0.1 Obs-Px 0.42 Exp IPR 1818
Raw Err= 0.199-19%(11/26), Book Err= 0.091-13.3%(2/15), middle Err= 0.251-28.6%(4/14),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 19% (4/21)-Top2- 52.4% (11/21)-Top3- 66.7% (14/21)-Top4- 66.7% (14/21)
Most Consectutive Candiate moves made 3 from move 23. Qb1 to move 25. axb4-NQ moves-open:8 Drawn:0 Forced:5 NQval:0

Predojevic, Borki (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.9 Obs-Px 0.58 Exp IPR 2855
Raw Err= 0.057-42.3%(11/26), Book Err= 0.109-20%(3/15)), Middle Err= 0.048-52.63% (10/19), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 42.3% (11/26)-Top2- 61.5% (16/26)-Top3- 76.9% (20/26)-Top4- 84.6% (22/26)
Most Consectutive Candiate moves made 8 from move 25.. Qxb4 to move 32.. Qxc3-NQ moves-open:8 Drawn:0 Forced:0 NQval:0

[Event "19th Open A"]
[Date "2012.12.22"]
[Round "9.3"]
[White "Saric, Ivan"]
[Black "Ivanov, Borislav"]
[Result "0-1"]
[ECO "B80"]
[WhiteElo "2626"]
[BlackElo "2227"]
[PlyCount "80"]
[EventDate "2012.12.16"]
[EventRounds "9"]
[EventCountry "CRO"]
[Source "crochess.com"]
[SourceDate "2012.12.22"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 e6 7. f3 b5 8. Qd2 Bb7 9. g4 Nc6 10 O-O-O Be7 11 h4 Nd7 12 g5 Nde5 13 Qf2 O-O 14 Nb3 Qc7 15 Bb6 Qc8 16 f4 Nd7 17 Be3 Qc7 18 Kb1 b4 19 Na4 Na5 20 Nb6 Nxb3 21 Nxa8 Bxa8 22 axb3 Bxe4 23 Rh2 a5 24 Bd3 Bxd3 25 cxd3 Qb7 26 Qg2 Qa6 27 d4 a4 28 Qc2 axb3 29 Qxb3 Ra8 30 Rc1 d5 31 Rc7 Qa1+ 32 Kc2 Nb6 33 Rxe7 Rc8+ 34 Kd2 Qf1 35 Qxb4 Nc4+ 36 Kc3 Nxe3+ 37 Kb3 Qd3+ 38 Ka4 Ra8+ 39 Qa5 Qc4+ 40 b4 Rxa5+

Saric, Ivan (White)- Accuracy 86% Blunders 2 (8%.) Mistakes 1 (4%.) Sub Opt 1 (4%.)Exp-Px 0.91 Obs-Px 0.46 Exp IPR 1203
Raw Err= 0.339-28%(17/26), Book Err= 0.12-20%(3/15), middle Err= 0.399-33.3%(6/18),, Endgame Err= - NQ moves.
White Candidate Moves= Top1- 28% (7/25)-Top2- 48% (12/25)-Top3- 60% (15/25)-Top4- 80% (20/25)
Most Consectutive Candiate moves made 3 from move 28. Qc2 to move 30. Rc1-NQ moves-open:8 Drawn:0 Forced:7 NQval:0

Ivanov, Borislav (Black)- Accuracy 100% Blunders 0 (0%.) Mistakes 0 (0%.) Sub Opt 0(0%.) Exp-Px 0.09 Obs-Px 0.54 Exp IPR 2699
Raw Err= 0.062-65.4%(17/26), Book Err= 0.123-13.3%(2/15)), Middle Err= 0.018-84.21% (16/19), Endgame Err= - NQ moves.
Black Candidate Moves= Top1- 65.4% (17/26)-Top2- 73.1% (19/26)-Top3- 76.9% (20/26)-Top4- 80.8% (21/26)
Most Consectutive Candiate moves made 10 from move 31.. Qa1+ to move 40.. Rxa5+-NQ moves-open:8 Drawn:0 Forced:6 NQval:0

• 3 years ago

sorry about the confusion kris. my notes contain only the first letter of the last name and the result and they are:

136, s-i, 0-1: 117, i-p, 0-1: 100, s-i, 1/2: 87, k-i, 0-1: 75, i-z, 1-0: 58, k-i, 1/2

i think that your suggestion of doing all 9 games is an excellent one and i will try to find the time to do that work. i see that you live in the bay area.

• 3 years ago

The PGN downloaded from FIDE has the following players at the numbers you listed (with the result appended as W-white wins, D-draw, B-black wins):

58  [White "Kuljasevic, Davorin"] [Black "Ivanov, Borislav"]  D

75  [White "Ivanov, Borislav"] [Black "Zelcic, Robert"]       W

87  [White "Kozul, Zdenko"] [Black "Ivanov, Borislav"]        B

100 [White "Petrusic, Toni"] [Black "Videnova, Iva"]          W

117 [White "Biliskov, Vjekoslav"] [Black "Paljusaj, Edmond"]  D

136 [White "Veleski, Robert"] [Black "Paljusaj, Edmond"]      W

The above does not correspond to what you described in your post and for a while got me royally confused as to what you were doing. But I guess that you meant games 103, 120 and 139 in the FIDE PGN (which happens to be the same as the one on the tournament site).

103 [White "Sumets, Andrey"] [Black "Ivanov, Borislav"]       D

120 [White "Ivanov, Borislav"] [Black "Predojevic, Borki"]    B

139 [White "Saric, Ivan"] [Black "Ivanov, Borislav"]          B

Overall, good work, and another independent confirmation of quite suspicious play on the part of "NGM"; analyzing all 9 of his games would be more convincing, though.

• 3 years ago

http://ratings.fide.com/view_pgn.phtml?code=73508

• 3 years ago

news from chessbase.com " The latest incident involves the former mayor of an Italian town who has been banned for allegedly using a hidden micro camera and earpiece to receive electronic assistance in tournament games. "

• 3 years ago

Please clarify who were the players in "games 58, 75, 87, 100, 117, and 136 in which the NGM played GM" (or perhaps provide the URL to the PGN you downloaded from chessbase.com).

Please clarify how you selected the GM-vs-non-GM games for analysis.

• 3 years ago

the longest streak of houdini moves by NGM in my sample was 14 moves long. curiously, the longest streak of houdini moves in the sample taken from the 2012 world championship was also 14 moves long (by anand).

• 3 years ago

Keep in mind "changes at different depths" must take into account that Ivanov did not have forever on each turn. Clearly in the marathon games where he had to move faster, he was getting faster moves from the engine. I'm quite certain at no point did he take more than 10 minutes on any given move where he was cheating. As such, we can't simply wait unil the engine changes its mind and say "Ah see! He didn't match the engine!". That's a false premise to argue from. When you review the games, It's just a little more than pure coincidence that a quad-core PC with Houdini 3 likes almost all of Ivanov's moves within 5 minutes given per turn (most of which instantly match).

• 3 years ago

I think the key to this sort of argument is doing what Jamalov has done. You have to select a given depth, and then compare the results with different players. Its not enough to say someone played the 'top move', because that changes at different depths. It would be interesting for someone to run all the world championship matches through such a proceess, creating a fairly clear baseline.

• 3 years ago

"There was only one game there where he didn't play like Houdini at all (obviously wasn't cheating in that game), and got crushed "

------------------------------------

yes, i do recall that game. of course your analysis has more depth because you are a better chess player and also you have h3pro. i was just trying to move the data from selective anecdotes to random samples to see if we come to the same conclusions and i think that we do.

• 3 years ago

I do have access to Houdini 3 pro, and my results when going over the games were quite blatant.

In the first game where he beat a 2400+, virtually every single move he played beyond the first 6 opening book moves were a perfect match for Houdini. All but one move matched the #1 choice, and the remaining move matched #2.

In the 117-move game against a GM where he blundered at the very end and lost in time trouble, 87 out of the 100 non book moves were deemed to be flawless by Houdini. At one point in the game at around move 64 to 71, he lost contact with the computer and started playing bad moves to the point of technically being lost. Then on move 72, he started cheating again and pulled back to equal, only to later lose at the very end on move 115 with a silly blunder.

These are just two examples of the incredible Houdini-like play by Ivanov at Zadar. There was only one game there where he didn't play like Houdini at all (obviously wasn't cheating in that game), and got crushed like a clueless patzer in it.

• 3 years ago

Very informative.

• 3 years ago

The Turk beats Napolean

• 3 years ago

The Turk

• 3 years ago

"The Turk" was a machine that played chess during Napolean's time and even beat Napolean but the machine cheated by secretly acquiring moves from a human and then making human moves.

• 3 years ago

Houdini is NOT the best engine? ตกใจ