Wednesday, March 02, 2005

Dragon Games 2: The ten-time state champion

The saga of 14...Nd3+ continues with a game against Arizona's first home-grown established master and the reigning state champion. One of the rewards of chess that has kept me interested over the years has been the life-long friendships I have made through the game. I met Bob Rowley in 1968 when I was 14 and he was 18. We were both 1500 players paired in the opening round of the Summer Rating Tournament at the Phoenix Chess Club, which held its meetings at that time in an old park building in downtown Phoenix. Bob missed a win in a pawn ending (something he soon learned to stop doing) and the game ended in a draw. At first I won a match from him and had a good +2 record in the first 7 games, before he went on a winning streak against me too long to mention. He cleaned out the state championship in 1970, drawing only one game, a feat he repeated against an even stronger field some 12 years later. In the 70s and 80s he and I were frequent guests at the house of the late Armand Bosco, where we played innumerable speed games. Between 1970 and 1991 he won ten state championships, (the Schwarz Memorial was a ten-round double round robin), and fell short only once or twice, as I recall.

I had hoped that one day he would develop into an IM or a GM, but that isn't terribly easy to do unless you develop at a young age or you quit your day job (for him, as a teacher a DeVry Institute of Technology). He has been a regular at the World Open and the National Open, and frequently been able to take half points and some full points out of grandmasters like Helgi Olafsson and Dmitry Gurevich. He got an IM norm at the World Open in, I think, 1992, but he never successfully followed it up.

We met in the fifth round of the 1971 Rocky Mountain Open. He knew that I had studied this line a lot, but he thought he had found a win for white. We followed the Polchinski game until

19. e5

Which has come to be regarded as the main line.

19...dxe5?

A move order error. Rxb3 20. axb3

20. Ne2?

After 20. g5 exd4 21. gf Rxc2 22. Bxc2 Qc3+ Kc1, the White king dances away and the Black one dies, as has occurred in a couple of games.

20...Rxb3

Apparently this move isn't necessary. In the game Bernys-Sinarski corr. 1990 Black played 20....Rac8 21. g5 Rxb3+ 22. axb3 Bf5 and went on to win.

21. axb3 Be6!

An improvement, suggested to me by three-time state high school champion Franklyn Yao, over 21...Bc6 22. g5 Nh5 23. Rxh5 gxh5 24. Ng3 e4 25. f4, which won for White in Grabczewski-Gasiorowski, Polish ch. 1970. I found a couple of games in the database in which Black played 22....Bb5 and did OK.

22. Ng3!

Rowley anticipated my improvement and came up with an improvement of his own. 22. g5 has led to draws in several games in the database. At almost the same time, Mecking played 23 Nc3, with the same idea.

22....Rc8!

At least I got one move right after getting outside of my analysis. That's better than Mecking's victim Joksic, who played 22....e4 23. Nxe4 Rc8 24. Kb1!! Rc6 25. g5 1-0.

23. Ne4!

After Mecking's 22. Nc3 Rc8 White has the option of 23. Rd3, but after Qc7 24. Rh2 Rd8 25. Rxd8 Qxd8 26. Ne4 Qd4+ 27. Kb1 Qg1+ 28 Ka2 a5 drew after 38 moves in Falkowski-Nitsche (great philosophical name!) DDR-cup corr. 1984.

23...Bd5???

The standard move here is 23....Rc6, but according the the Fritz5.32 there is no clear win for White after moves like ...b5 or event ...a6. After 23...Rc6, in the game Kanjo-Poleksic, Yugoslav corr. 1987, play continued 24. g5 Nh5 25. Rxh5 gxh5 26. c4 Ra6 27. Nf6+ exf6 28. gf6 Qa2+ 29. Kc3 Qa5+ 1/2. Nine years after the Rowley-Reppert game, ot one of our nights at Armand's house, Bob uncorked 24. Rd3 Ra6 25. c4 and after something like Qa2+ 26. Kc3 Qa5+ 27. Kc2 Qa2+ 28. Kd1 Qb1+ 29. Ke2 Qc2+ 30 Rd2 White wins, and if 29...Ra2+, then Nd2 wins for White. So for many years I thought that "St. Bob" had slain the 14....Nd3+ variation. But when I started using computers I booted this position up, and Chessmaster 5500 came up with 25...Bc4, and after 26. bc Qa2+ 27. Kc1 Ra1+ 28. Rc3 Qa4+ 29. Kc3 Qb5+ 30. Kd3 Qd7+ 31. Ke2 Qb5+ 32. Ke3 Nd5+ 33. Kf2 Qb6+ 34. Rc5 Rxh1 35. Qxh1 Nc3 36. Kg2 Nxe4 37. Rc8+ Kg7 38. fxe4 Qe6 and Black can probably win.

One strange addendum to all of tis is that after 24. g5 Nh5 some books recommend 25. Rd6 (and some attribute this to Tal) Bxb3 26. Ra1 Qxa1 is recommended, as leading to a draw, but 26...Ba4 pockets a piece. Another book howler.

24. Nxf6+ gxf6 25. Qxh7+ Kf8 26. Qxh8+ Ke7 29. Qxc8 1-0

No comments: