Eliot Hearst and John Knott blog about blindfold chess
Saturday, December 18, 2010

Simultaneous Blindfold: FM Lang Plays 35, Can He Beat Najdorf’s World Record of 45 in 2011?

Over the course of the last year and a half we have written two blogs on this website about the German FIDE master Marc Lang’s successive and dramatic increases in the number of blindfold games he has played all at once. He moved from playing 15 simultaneously in June 2009 to becoming the holder of the German record of 23 in November of the same year. Just recently (Nov.27-28, 2010) he successfully took on 35 opponents in Sontheim, Germany, which eclipsed the 34-board performance of George Koltanowski in Edinburgh, Scotland in 1937. Koltanowski’s accomplishment became the world record for number of simultaneous blindfold games played up to that time, but a decade later Miguel Najdorf played 45 at once in São Paulo, Brazil and this currently stands as the generally accepted world record.

So, with 35 games, Marc Lang now holds both the German and European records. Only Najdorf’s achievement stands between Lang’s and the world record. He expects to play 46 late next year to establish a new world record and it seems likely that he will reach this goal. Psychologists would consider all the displays mentioned above, as well as others described in our book, to be among the greatest memory feats that humans have accomplished.

Lang’s recent display received exceptional coverage in the German television and general print media, perhaps as much as or more than has been devoted to regular world chess championship reports. Maybe that is because no German GM has been a solid world championship contender for many years! If you know the German language you can read a detailed report of Lang’s 35-board display here, which also includes a listing of all the individual board results, photos and videos of the exhibition, as well as other historical and relevant material. It even shows Lang playing chess at home with his two young children and observing wife, or riding his bike to maintain his general physical health and to keep in shape for his strenuous displays. We reproduce here the U-tube video from German TV, which will give our readers a view of the playing arrangement and computer-controlled setup, as well as many other features of the exhibition. Unfortunately, the audio part of the video is in German, but the visual part is easy for most of us to follow, even without a knowledge of that language.

Rather than sending our readers to German language websites, I think they would like to find out some details and sidelights of Lang’s recent display written in English. He has corresponded extensively with me and much of what follows is derived from his emails. His final score was 19 wins, 13 draws, and 3 losses, a very good winning percentage of 72.9%. The whole event took 23 hours from about 9 AM on Saturday, November 27 to about 8 AM on Sunday, with a total of four breaks (half-hour each). Lang sat in the center of the exhibition room, facing all his opponents whose chess positions were concealed from him by cardboard barriers in front of each game. Lang has used this arrangement before and he could chat and joke with each opponent if he wanted to. Also, seeing the people he faced probably enabled him to build up stronger associations with the moves that had occurred in each game. Lang allowed opponents to be replaced by another person if they got too tired and did not want to stay until the finish.

Lang, a computer professional. sat in front of a laptop and entered the moves called out by his opponent and then typed in his answer. His computer was connected to a video projector, which could be viewed by everyone, that indicated his opponent’s last move briefly and his reply briefly so that his and his opponent’s moves could be checked for accurate transmission and a complete score of the game would be available at the game’s conclusion.

To be even clearer, only a single move appeared on the projector at a time and disappeared after each was input. Thus Lang had the added task of typing moves into a computer, which apparently did not distract him from his major task of playing 35 games simultaneously! Of course he was playing “blindfolded” because he never saw anything more than his opponent’s last move and then his own reply, which is intrinsically the same as all previous blindfold displays where the exhibitor heard his opponent’s last move and then called out his reply. The use of a computer-projector arrangement like this was a novel way of conducting a blindfold display and it probably helped prevent incorrect transmission of moves that have occurred in past displays because of simple errors made in calling out moves.

Another novel and useful feature of this exhibition was that all but two of his opponents had standard numerical ratings based on the German DWZ system. Transformed into USCF ratings, the strength of his opponents ranged from around 1000 to around 2100, with an average rating of around 1600. One of the problems comparing all previous blindfold displays over the past three centuries is that the strength of opponents could never be adequately evaluated. Thus it was almost impossible to decide whether a performance involving 20 opponents might be superior to one of 25 or 30, because the opposition might have been by far the strongest in the 20-board display. That is why the world blindfold record has always been decided in terms of the number of opponents one has faced. Now actual ratings, rather than subjective judgments, can be used as an additional factor in comparing winning percentages in displays involving different numbers (or even the same number) of opponents.

Of course, you will not find masters and strong experts participating as opponents in most record-setting displays (except in exhibitions given in the 1920s and 1930s by Alexander Alekhine, who was to become or actually was the regular world champion at about that time and who is considered by almost all knowledgeable blindfold experts as the best blindfold player ever). As GM Erich Eliskases wrote me with respect to Najdorf’s 1947 display, “no one can play such a lot of (blindfold) games against players of first-class strength”. Except maybe Alekhine.

For example, my impression is that the average and range of strength of Lang’s 35 opponents was greater than Koltanowski’s 34. Readers can play over the games from Koltanowski’s display in our book and determine that many opponents were very weak and lost very quickly. Furthermore, he offered or accepted quite a few draws within the first 15 moves, in order to rapidly decrease the number of games he still had to play and recall. Koltanowski finished 47.1% of his games in fewer than 16 moves, whereas Lang finished only 8.6% of his games in fewer than 16 moves and took no quick draws, as he had vowed to do before the exhibition. Lang’s percentage is remarkably close to Najdorf’s on this measure, since the latter had only 8.9% of his 45 games ending in fewer than 16 moves. However, my impression is that Najdorf’s opposition in 1947 was somewhat stronger than in Lang’s recent display. All of Najdorf’‘s games in this display are included in our book and we welcome opinions from readers about the strength of his opposition compared to, say, Koltanowski’s. Unfortunately, no numerical ratings were being calculated as far back as 1947 and so such judgments have to be subjective. (The German website cited above gives several of Koltanowski’s very short games.) Incidentally, I calculated the average number of moves in Koltanowski’s, Najdorf’s, and Lang’s exhibitions to be, respectively, 18, 26, and 24 moves.

To handle so many games at once, blindfold champions must develop a system beforehand for making memory of and distinctiveness between the various games far easier than playing without a well-practiced system. We discuss many such mnemonic systems, often quite different for each individual exhibitor, in our book on blindfold chess. Lang decided to group the games into 7 sets of 5 games, with the first four games of each set being based on a certain theme or opening pattern, with him having the White pieces, and the fifth game “anchoring” this set by his playing Black. For example, in the first five games he employed what he calls a “motto” (“knight to the left”, Nc3 with White or Nc6 with Black very early in each game). He actually started Games 1 and 4 with 1.Nc3 but played 1.d4 in Game 2 and 1.e4 in Game 3 followed on his next move with 2.Nc3. With Black in Game 5 he answered 1.e4 with 1…Nc6. For the second set of five games his “motto” was “c-pawn” and he played 1.c4 in Games 6 and 9 and c4 on his third move in Game 7; with Black in Game 10 he answered 1.d4 with 1…c5. In Game 8 he decided to play c3 on his third move because c4 was an inappropriate move (some flexibility is usually necessary depending on his opponent’s play).

Interestingly, Lang reported that whenever he had Black he couldn’t turn the positions upside down in his memory. He played them all with a view from the White side!

Lang also included one computer as an opponent, a Fritz 11 program set to play at a USCF rating of approximately 1800. Lang outplayed the computer and should have won, but (as many of us have unfortunately discovered in our own games against computers), the machine took advantage of some weak moves by Lang and eventually scored one of the three victories against him (the other two were due to terrible blunders by Lang; forgetting the exact position, he placed his queen where it could be taken for nothing in both games). The game against the computer is given below, along with two wins by Lang in this exhibition.

Unlike many previous blindfold champions, Lang was not haunted by persisting images of the games for days or months afterward. Furthermore, he was able to fall asleep soon after the display and did not suffer the insomnia that so many other blindfold champions have reported.

All the above evidence indicates that Lang will be successful in his attempt to play 46 games in 2011, but we will have to wait about a year to find out.

M. Lang – H-P. Muck
Blindfold Simul, Sontheim, Germany, November 27-28, 2010
Board 28 (of 35)
Irregular King’s Pawn Opening B00

Muck was Lang’s strongest opponent, with an approximate USCF rating of 2100. Here he apparently tried to confuse Lang by playing an eccentric opening, but it backfired completely. The diagram shows the final position, in which Muck resigned.

1.e4 g5 2.d4 h6 3.h4 g4 4.Qxg4 d5 5.Qf4 dxe4 6.Nc3 Bg7 7.Bc4 Nf6 8.Nxe4 Nd5 9.Qg3 Kf8 10.Nf3 c6 11.Bd2 Bf5 12.Nc3 Nb4 13.0–0–0 Bxc2 14.Qf4 Nd3+ 15.Bxd3 Bxd3 16.Ne5 Bxe5 17.Qxe5 f6 18.Qh5 Kg7 19.Rh3 Bg6 20.Rg3 Qe8 21.Re1. 1-0

M. Lang - F. Jarchov
Blindfold Simul, Sontheim, Germany, November 27-28, 2010
Board 18 (of 35)
Caro-Kann Defense B12

1.e4 c6 2.d4 d5 3.e5 Bf5 4.Nf3 e6 5.Be2 Bb4+ 6.c3 Ba5 7.0–0 Ne7 8.Nh4 Bg6 9.Nd2 Nf5 10.Nxg6 hxg6 11.Nf3 Nd7 12.Bg5 f6 13.exf6 gxf6 14.Bf4 g5 15.Bg3 Nf8 16.Bd3 Nxg3 17.fxg3 g4 18.Nh4 f5 19.Qe2 Rxh4 20.gxh4 Kd7 21.g3 Qf6 22.Qxg4 Qh6 23.Qe2 Nh7 24.Rae1 Qf6 25.b4 Bc7 26.Qh5 Rh8 (Diagram) 27.Bxf5 exf5 28.Rxf5 Qd8 29.Rf7+ 1–0

Fritz 11 Computer – M. Lang [D78]
Blindfold Simul, Sontheim, Germany, November 27-28, 2010
Board 35 (of 35)
Grünfeld Defense (Irregular) D78

1.c4 g6 2.d4 Bg7 3.g3 Nf6 4.Bg2 0–0 5.Nf3 d5 6.0–0 c6 7.Nbd2 Bf5 8.Qb3 Qb6 9.Qe3 Nbd7 10.c5 Qc7 11.b4 Rfe8 12.Bb2 e5 13.Qg5 h6 14.Qe3 exd4 15.Qb3 Rxe2 16.Bxd4 Rae8 17.Qc3 Nh5 18.Bxg7 Nxg7 19.Nd4 R2e7 20.Bf3 Ne5 21.Rfe1 Nxf3+ 22.N2xf3 Rxe1+ 23.Rxe1 Rxe1+ 24.Qxe1 Kf8 25.Qe3 g5 26.h4 f6 27.Nxf5 Nxf5 28.Qe6 g4 29.Qxf5 gxf3 30.Qxf6+ Qf7 31.Qxh6+ Ke8 32.g4 Qe7 33.Kh2 Qe2 34.Qg6+ Kd8 35.Qg8+ Kd7 36.Qg7+ Kc8 37.Qf8+ Kd7 38.Qf7+ Kc8 39.Kg3 1-0 (Diagram of final position; with all those White passed pawns on the kingside, Black’s position is hopeless, especially against a computer.)

Permalink  |  Posted by Eliot Hearst at 01:53 PM


Home  |  Excerpt  |  Blog  |  Reviews & Press  |  About the Authors  |  Contact the Authors  |  Blog RSS (Atom)