CHESS#1309

The Leon Masters saw Wesley So win two extremely close matches to take a second consecutive title. So is world number two and rated 2852 (Rapids). His final opponent, Francisco Vallejo Pons is a relatively lowly 2628. His first opponent, R Praggnanandhaa, is rated an absurdly low 1736 (beginner level). Praggnanandhaa is the world’s youngest GM in the world but he hasn’t played rapid tournaments for years. There’s also a wide gulf between Praggnanandhaa’s classical rating of 2529 and So’s 2780.

The 12-year-old won the first game of the four-game match. So equalised in Game 2 and finally won 2.5:1.5. If Praggnanandhaa had converted the chances he created, he could have won 3:1. His time will surely come.

The final went to a tie-breaker. The Spanish GM won Game 1, when So blew up a promising position. So equalised in Game 2. In Game 3 and Game 4, Vallejo couldn’t convert positions from a pawn up each time. In the first tiebreak blitz game, So won and Vallejo couldn’t pull it back in Game 2.

Speaking of positions,

Chessbase.com has published a couple of recent pieces on position recognition. One difference between a strong chessplayer and a casual amateur is that the pro sees positions in patterns of multiple pieces, rather than as individual piece by piece setups.

It’s like reading words without spelling out letters. For example, it’s more difficult to read, and recall “xdfkzy”, than “castle”. Adriaan de Groot, a Dutch IM and psychologist, did some position recall experiments in the 1940s. Fred Friedel of Chessbase reproduced those experiments in the 1980s using sophisticated gear to track eye movements. A young Canadian researcher and chessplayer, Qiyu Zhou, has done a new study on the role of shortterm memory.

Pros familiar with chess “language” can take a brief look at a “normal “position and recall it accurately. For example, a strong player “sees” a fianchetto structure with pawns on f2,g3,h2,Ktf3, Bg2, castled king etc., (seven pieces in 3x3 space) as one unit. But if pieces are jumbled randomly, without “normal” patterns, GMs are not much better at recall than rank beginners.

The Diagram, Black to Play ( White: So Vs Black: Praggnanandhaa, Game 1, 31st Leon GM 2018) is an edgy endgame with both players on increment (10 seconds/ move). Black can win with 54...g3! 55.fxg3 Bxg3 56.Rg8+ Kf4 57.Rf8+ Kg4 58.Rf1 Bf2+! 59.Kd5 Bg1!

Instead he tried 54. --Bb6+? 55.Kd5 Bxf2 56.Rxh2 g3 57.Rh8 Nb5 58.Rg8+ Kf4 59.Rf8+ Ke3 60.e5 g2 61.Rg8 Kf3 62.Kc6?? [Blunders. 62.Rxg2! Kxg2 63.e6 Nc7+ 64.Kd6 Nxa6 65.e7 Bc5+ 66.Kd7 Bxe7 67.Kxe7 draws]. Now 62...Bg3! 63.e6 g1Q 64.e7

Nc7 (0–1).