Tuesday, January 12, 2021

Evans-Steiner Thriller at the 1948 US Championship

Herman Steiner
     The 1948 US Championship was big...20 players. Originally there was supposed to be 22 players, but the USCF decided to eliminate two spots. It was a tournament with the best conditions and one of the best prize funds ever even though the USCF was on the verge of bankruptcy. 
     It was played 100 miles away from New York City in South Fallsburg, New York which at the time was a major resort area in the the Catskill Mountains. 
     It was also one of the weakest US Championships ever. Fine didn't play because he was working towards his degree in psychology. Reshevsky's absence was a little more complicated. He was willing to play if he received his fee, but the organizers couldn't afford it because they had already cut costs by moving the tournament out of the big city plus they had put all the available money into the prize fund. Besides that, they balked at paying Reshevsky just to show up when they were already offering a generous prize fund that he could win. 
     After negotiations broke down Reshevsky not only wouldn't play, he demanded that his name be removed form all the advance publicity for the tournament. Olaf Ulvestad of Seattle was selected as his last minute replacement. 
     The tournament started on August 23rd and on the 9th Reshevsky agreed to play without his fee, but it was too late. They weren't going to cancel Ulvestad's invitation and besides that, there would be a problem booking another room in the resort town. Another problem was that it would lengthen the tournament which meant more problems with the hotel bookings. Plus, they would have to convince the other players to revise their schedules to accommodate Reshevsky. Consequently, Reshevsky was told to go pound salt. 
     The tournament was further weakened by the absence of Arthur Dake, Arnold Denker and I.A. Horowitz. That meant Isaac Kashdan had an excellent chance to finally win the Championship...something he had been trying to do since 1932. But, it wasn't to be. 
     Kashdan started out well with 8 points in the first 9 rounds with Steiner trailing at 7-2. That in itself was somewhat surprising because Steiner had had difficulty qualifying from the regional event and, honestly, while he was regarded as steady player, nobody believed he was good enough to win the tournament, or even rank among the best US players of the day. After all, earlier in the year Reuben Fine had absolutely crushed him 5-1 in a match. 
Larry Evans

     When Kashdan was upset by Sol Rubinow in the 10th round things got interesting and it was only in the last round that Steiner pulled out the victory.
     Going into the last round Steiner and Kashdan were tied at 14-4. Steiner was playing Franklin S. Howard who stood at 5.5-12.5 while Kashdan was playing George Kramer who had 12.5-5.5. Steiner got into trouble and appeared on the verge of defeat, but a nervous Howard couldn't find the win, frittered away his advantage and resigned on his 65th move. Meanwhile, Kashdan allowed Kramer to draw by a perpetual check and Steiner was the champion. 

     The US Women's Championship was held concurrently and first place was shared by Gisela Gresser and Mona May Karff, both of whom easily outdistanced the field. 

     In the following game Larry Evans got an advantage in the opening, but wasn't able to parlay it into anything. Things soon got pretty complicated and both players played some colorful chess.

Larry Evans - Herman Steiner

Result: 0-1

Site: South Fallsburg, New York

Date: 1948.08.17

King's Indian

[...] 1.d4 ♘f6 2.c4 g6 3.♘c3 ♗g7 4.e4 d6 5.g3 e5 6.d5 It was the opinion of Hans Kmoch that in the K-Indian white should keep his P on d4 as long as possible in order to "retain full freedom of action. " He did admit though that the text move is also good and even safer because it precludes early complications in the center. Also, in Kmoch's opinion the advance of the d-Pawn deprives white of the chance of opening a file for the Rs and so it's "somewhat dull." 6...a5 7.♗g2 ♘a6 8.♘ge2 ♘c5 9.O-O O-O 10.h3 ♘e8 11.♗e3 f5 Black is more or less forced to make this move without careful preparation as otherwise white may get time to carry out his own breakthrough, after due preparation, c4-c5. The move ...f5 is not without risk as the more open character of the position can end up favoring white who controls a more space. Denker stated that Steiner was always willing to ignore positional considerations if it lead to sharp play.
11...♗d7 This solid move didn't lead to anything for black is the following game. 12.♕d2 b6 13.♖ae1 ♕e7 14.f4 exf4 15.♗xf4 ♖d8 16.♗g5 f6 17.♗f4 ♗c8 18.♘d4 ♗b7 19.♔h2 ♖a8 20.g4 ♖f7 21.b3 ♗h8 22.♗f3 ♘d7 23.♘e6 ♘e5 24.♗g2 ♘g7 25.♘xg7 ♗xg7 1/2-1/2 Bolbochan,J-Gentil,L/Mar del Plata 1952
12.exf5 gxf5 13.f4 With the threat 14.fxe5 forcing 14...Bxf5 leaving black with an isolated f-Pawn and at the same time allowing white to take control of d4 and f4. 13...b6
13...e4 It's interesting that Kmoch considered this a bad move because it yields white the square d4. However, 13...e4 is Komodo 12's recommended move. 14.♘d4 ♘f6 15.♖f2 ♗d7 16.♗f1 ♕e7 17.♖g2 h5 18.♗e2 ♕f7 19.♕d2 with equality. Kempinski,R (2606)-Leniart,A (2458)/Gorzow POL 2012
14.fxe5 This is best, but Kmoch thought a good alternative is 14.Qc2 with the intention of quietly strengthening the position by Rad1. 14...dxe5 15.d6 ♖b8 16.dxc7 ♘xc7
16...♕xc7 is inferior. 17.♘d5 ♕d8 18.♘xb6 ♕xb6 19.♕d5+ ♗e6 20.♗xc5 is to white's advantage. Blom, K-Niemela,I/Tel Aviv 1964
17.♘d5 Trade Qs or not? According to Kmoch, Denker and Soltis the answer is yes. Kmoch observed that with Qs on white is handicapped by the somewhat precarious position of his K as well as the inability to take any action on lhe K-side. Additionally, he faces the danger of black launching a K-side attack on the g-file. What to make of all this when three strong masters disagreed with the stronger Komdo and Stockfish engines? I would tend to agree with the masters because in the absence of any immediate tactics, long range positional factors come into play and humans are still better judges of such things. Besides, humans are playing this game, not engines.
17.♕xd8 According to Hans Kmoch white has a slight edge in the ending after the trade of Qs. Andrew Soltis and Arnold Denker agreed that 17.Qxd8 and 18.Rad1 followed by Nd5 results in a favorable ending for black. Komodo 12 disagrees. 17...♖xd8 18.♖ad1 ♖xd1 19.♖xd1 ♗e6 20.♘d5 ♘xd5 21.♗xd5 ♔f7 and white does not appear to have any advantage. In 5 Shootouts with each engine neither Komodo or Stockfish could demonstrate any superiority and all 10 games were drawn.
17...♘xd5 White could still trade Qs, but the circumstances have changed. 18.♗xd5+
18.♕xd5+ ♗e6 19.♕xd8 ♖bxd8 Black's pieces are better placed which gives him a slight advantage.
18...♔h8 19.♔h2 ♕c7 20.♕d2
20.♕c2 The Q is well placed here, but Evans wanted to avoid complications arising from... 20...f4 21.gxf4 ♗f5 22.♕d2 ♘d3 Here though white has an interesting idea that would keep both players on the edge of their seat. 23.♖g1 ♖be8 24.♖xg7 ♕xg7 25.♖g1 ♕f6 26.♖g3 exf4 27.♘xf4 ♘xf4 28.♗xf4 ♗e6 with unclear complications.
20...♗b7 21.♘c3 ♖bd8
21...f4 A most interesting move, but the complications could easily get away from either side. 22.gxf4 exf4 23.♗d4
23.♖xf4 is also rather complicated. 23...♘d3 24.♕xd3 ♖xf4 25.♗xf4 ♕xf4+
23...f3+
22.♗h6 Evans seeks to trade Bs and so eliminate the dangers that come from a potential ...f5. 22...♖d6 White has the choice between facelng a dangerous attack after 23.Bxg7 or playing a difficult ending. He chooses the latter. 23.♘b5 ♗xh6 24.♕xh6
24.♘xc7 going for material but overlooking... 24...♗xd2 25.♗xb7 ♘xb7
24...♖xh6 25.♘xc7 Although technically the ending is about equal, practically speaking black has the better chances. 25...♗c8 Threatening the advance of his f-Pawn attacking h3, but at least with the Qs off the messy consequences are lessened. 26.♗g2 ♘d3 27.b3 All the annotators missed the fact that this natural move is a serious mistake.
27.♘d5 saves the day. 27...♘xb2 28.♖ab1 ♘xc4 29.♖fc1 ♘d2 30.♖xb6
30.♖b2 This is preferred by the engines. 30...♘e4 31.♘xb6 ♗e6 32.♖c6 with a technically equal position, but one that looks difficult to play.
30...♖xb6 31.♘xb6 ♗a6 and black's advantage may not be enough to win.
27...f4 Finally! 28.♘d5 One move too late. This is not as good now as it was a move ago/
28.gxf4 is not much better. 28...exf4 29.♖ad1 ♘e5 30.♖de1 ♖h5 31.♘b5 f3 32.♘d4 ♗b7 33.♗h1 ♖e8 34.♘xf3 ♘xf3+ 35.♗xf3 ♖xe1 36.♖xe1 ♗xf3 with a favorable ending.
28...♗g4 In annotating this game Kmoch said black has concentrated his forces very effectively and threatens ...Be7. The first part of his statement is correct, but as will be seen, the second part is not. 29.gxf4 ♗e2 This move is a mistake that was missed by all the annotators! It gives white a way out of his predicament.
29...exf4 is correct. Then after 30.♘xf4 (30.♘xb6 f3 31.♗h1 ♖xh3+ 32.♔g1 ♖g8 white can't avoid mate.) 30...♘xf4 31.♖xf4 ♖xf4 32.♔g3 ♗xh3 33.♔xf4 ♗xg2 with a won game.
30.♖g1 Kmoch wrote that this move loses decisively because trying to hold his own by passive measures will fail. Instead, he should give up the R for the B and a strong P which would stop black's attack and get a sound position with hardly any danger of losing after 30.c5. If you don't have a powerful engine to consult this may be a correct statement, but if you do, things aren't so clear cut!
30.c5 Kmoch's move resulted in black winning 5 Shootouts. Here is how Komodo 12 managed it at 23 plies. 30...♗xf1 31.♖xf1 ♘xc5 32.fxe5 ♖xf1 33.♗xf1 ♘d7 34.♔g3 ♘xe5 35.a3 ♖d6 36.♗g2 ♖g6+ 37.♔h2 ♘d7 38.♗f3 ♔g7 39.♘f4 ♖d6 40.♔g3 ♖d2 41.♗d5 ♖c2 42.♔h4 ♖c3 43.a4 ♔h6 44.♘e6 ♖d3 45.♗c4 ♖e3 46.♘f4 ♘e5 47.♗d5 ♖c3 48.♗e6 ♖f3 49.♘g2 ♔g7 50.♗d5 ♘g6+ 51.♔g4 ♖d3 52.♗e4 ♖xb3 53.♗c2 ♖c3 54.♗d1 ♔f6 55.h4 ♘e7 56.♔f4 ♘f5 57.h5 ♖c1 58.♗b3 ♖b1 59.♗g8 ♖b4+ 60.♔f3 ♔g7 61.♗d5 ♖xa4 62.♘e1 ♔f6 and black wins.
30.♖fe1!! is the engine way. 30...♘xe1 31.♖xe1 ♗g4 32.fxe5 ♗xh3 33.♗xh3 ♖f3 34.♔g2 ♖hxh3 35.e6 and black must take the perpetual check because of the e-Pawn.
30...exf4 This mighty passed P, backed by the superior activity of black's forces, secures victory - Kmoch. True, black's advantage is significant and it should be enough to win, but white is not dead yet. 31.♗f1 ♖e8 32.a3 After this white is lost for sure.
32.♖g5 This offers white drawing chances. 32...f3 33.♘c3 ♖g6 34.♖xg6 hxg6 35.♔g3 and black faces some annoying problems trying to find the win. In Shootouts Komodo white scored +0 -3 =2. Here is a draw at 23 plies. 35...♖e5 36.h4 g5 37.hxg5 ♖xg5+ 38.♔h4 ♖e5 39.a3 ♗xf1 40.♖xf1 f2 41.♔g3 ♖e3+ 42.♔h2 ♘e5 43.♘d5 ♖xb3 44.♔g2 ♘xc4 45.a4 ♖d3 46.♘f4 ♖a3 47.♘d5 ♖xa4 48.♖c1 b5 49.♔xf2 ♖a3 50.♔e2 ♖b3 51.♘c7 ♘d6 52.♖d1 ♖b2+ 53.♔d3 b4 54.♖a1 ♖b3+ 55.♔d4 ♖c3 56.♘e6 ♘b5+ 57.♔e5 ♖a3 58.♖h1+ ♔g8 59.♖g1+ ♔f7 60.♖g7+ ♔e8 61.♘c5 ♖h3 62.♖b7 ♖h5+ 63.♔e6 ♘d4+ 64.♔d6 ♔f8 65.♖b8+ ♔f7 66.♖a8 ♖h6+ 67.♔d5 ♘c6 68.♔c4 ♖h4+ 69.♔b5 ♘d4+ 70.♔a4 ♖h2 71.♖a7+ ♔f6 72.♖a6+ ♔e5 73.♔xa5 ♔d5 74.♔xb4 ♘c6+ 75.♖xc6 ♔xc6 with a theoretical draw.
32...f3
32...♗xf1 was better. 33.♖axf1 ♖e2+ 34.♖g2 f3 35.♖xf3 ♖xg2+ 36.♔xg2 ♘e1+ 37.♔g3 ♘xf3 38.♔xf3 ♖xh3+ with a winning position.
33.b4
33.♖g3 was better. Then after 33...♖g8 34.♖xg8+ ♔xg8 35.b4 The win is not guaranteed.
33.♖g3 ♗xf1 is not so good. 34.♖xf1 ♘e5 35.♖f2 and there does not seem to be any way that black can force a win.
33...♗xf1 34.♖axf1 ♖e2+ 35.♔g3 f2 36.♖g2 ♖e4 This little slip makes the win more difficult. 37.♔h2 Tricky devil that Evans! If black snatches the c-Pawn he loses the exchange. 37...♖he6
37...♖xc4 38.♘e7 and now the natural move yields white the advantage! 38...♖xh3+
38...♖g6 39.♘xg6+ hxg6 40.bxa5 bxa5 41.♖xg6 Theoretically at least white is winning.
39.♔xh3 ♘f4+ 40.♔h2 ♘xg2 41.♔xg2 axb4 42.axb4 ♖xb4 43.♖xf2 with a theoretical draw.
38.bxa5 bxa5 39.♖b1 (39.♘c7 isn't any better. 39...♖6e5 40.♘d5 ♖xc4 41.♘f6 ♖c8) 39...♖e1 40.♘f6 Threatening mate, but black has a winning riposte. (40.♖b8+ would lose after 40...♖e8 41.♖xe8+ ♖xe8 42.♖xf2 ♘xf2) 40...f1=♘+ A rare under promotion. (40...f1=♕ 41.♖g8#) 41.♔g1 ♘g3+ 42.♖xe1 ♖xe1+ 43.♔h2 ♘f1+ 44.♔h1 ♘e3+ 45.♖g1 ♖xg1+ 46.♔xg1 ♘xc4 Evans resigned. A great game that's was a credit to both players.
Powered by Aquarium

No comments:

Post a Comment