My opponent in this game was Konstantin Dolgitser of New York City.

1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Qc2 c5 5.dxc5 Na6 6.a3 Bxc3+ 7.Qxc3 Nxc5

When I saw that two of my opponents in this section, Dolgitser and Pete Dyson, were playing this way, I spent a lot of time looking for improvements on the book ideas following from this position.

8.b4 Nce4 9.Qb2

Another challenging continuation is 9.Qd4 d5 10.cxd5!

Here my recommendation to Nimzo players is

A) 10...0-0! after which:

A1) 11.f3 is GM Ivan Sokolov's recommendation, and with no further continuation, he claims a slight advantage for White. But I remain impressed by Black's piece activity, for example: 11...Nd6 12.Bg5 Nf5 13.Qb2 exd5 14.g4 (14.e3 Qe7 and I like Black ; 14.Bxf6 gxf6 is also good for Black) 14...Nd6 15.e3 Re8 16.Rd1 Bd7 with good play for Black;

A2) 11.Nf3 a5 12.b5 Qxd5 13.Qxd5 Nxd5 14.e3 Bd7 with a good game for Black in Bonsch-Tischbierek, Hanover 1981;

A3) 11.Bb2 e5! 12.Qe3 Bf5 13.Nh3 Nxd5 14.Qd3 Bg6 with a superb game for Black in Huebner-Christiansen, Koeln 1990;

B) 10...exd5 11.Bb2 0-0 12.e3 Nd6 This was played in some high-level games, but maybe somebody can explain to me why it is so important to withdraw this knight right now. It seems to me that simple development would be a better plan. 13.Nf3 Bf5 14.Be2 and you can check the books for the latest theory, which shows that White is slightly better.

9...d5 10.c5 d4

This is recommended by Sokolov in his 1995 Cadogan book, Nimzoindian Defense, Classical Variation.

10...h6 11.f3 Ng5 12.Qe5 (12.Bf4 also looks good ) 12...Ngh7 13.Bb2 is advantageous for White.

11.e3 e5

Sokolov's most recent recommendation is 11...Ng4 12.Nh3 dxe3 (12...Nc3? 13.Bd2 0-0 14.Bxc3 dxc3 15.Qxc3 is a curious echo of the game continuation and was played in Maksimenko-Lipka, Lubniewice 1994. Comparing with 13...O-O! in the note to Black's 13th move, it is clear that Black is much worse off here.) 13.Qxg7 Ke7

Analysis position after 13... Ke7

Looking at the diagram, it's apparent that the computeresque 14. Qxg4 lets Black's queen in via d4, and also permits the survival of the dangerous e-pawn, with consequences that are difficult to forsee. So before following that demanding path to whatever analytical conclusion it might have, let's try a little chess by smell. White has the two bishops, a big advantage on this open board, and since it's his move, he can win the e3 pawn. And it's not just any pawn, but as seen in the diagram, it's a virtual piece. Winning the pawn also has the nice effect of depriving Black's queen of d4. Still further, after 14. fxe3, Black's g4 knight remains "in take" and his king can hardly expect much comfort on e7. True, Black's knights are very active, but I don't see any immediate threats after 14. fxe3. Is the queen trapped? I don't see how. She has a nice exit down the long diagonal and if Black blocks that by retreating a knight to f6, White can play Ng5, which appears to force exchanges. The verdict of chess by smell is clear. 14. Qxg4 smells funny. 14. fxe3 smells good.

A) 14.fxe3! Rg8 (14...Nef6 15.Ng5; 14...h5 15.Ng5; 14...Ngf6 15.Ng5) 15.Qd4 Ngf6 16.Nf4 in all cases with a big advantage for White;

B) Sokolov gives only the dubious 14.Qxg4 and then demonstrates that White's play is punished by 14...Qd4 15.Ra2 Qc3+ (15...e5? 16.Qh4+ f6 17.Bxe3 Qc3+ 18.Bd2 Nxd2 19.Rxd2 Qc1+ 20.Ke2 Be6 21.Rd6! and White is winning, according to Sokolov) 16.Ke2 Nf6 17.Qf4 Qxc1 18.fxe3 Rd8 and Black has excellent compensation. But I doubt that Sokolov or many other GMs would play 14. Qxg4 at the board.

This was my novelty.

12.Bb5+ Bd7 13.Bxd7+ Qxd7 14.Nf3 Ng4 15.0-0 0-0-0 16.h3 h5! 17.exd4 exd4 18.Bf4 Now here we have a funny position from Arlandi-Grosar, Debrecen 1992, concerning which Sokolov seems to have changed his mind. In his 1997 Informant monograph, he stops here and says White is slightly better. In his 1995 Cadogan book, he quotes more of the game: 18...Qf5 19.Qc1 Rd5 20.Qc4 Ngf6 21.Bd6 Nxd6 22.cxd6+ Kb8 23.Nxd4 Qd7 and here Sokolov says Black is fine. I'd be curious to know why Sokolov changed his mind about this evaluation.

12...Nc3 13.Bd2 Nfe4

13...Na4 14.Qb3 Bd7 (14...Qd7 15.Rc1 and White, threatening 16. c6, has a significant advantage) 15.c6! Bxc6 16.b5 Nc5 17.Qc4 and Black's development will be insufficient to atone for the missing piece.;

It seems that Black can equalize with 13...0-0! a move to which I did not give serious attention during the game:

A) 14.exd4 This smells bad because it opens the e-file. 14...exd4 15.Bxc3 dxc3 16.Qxc3 Qe7+ 17.Kf2 Nd5 18.Qd4 Rd8 19.Bc4!

A1) Tricky but insufficient is 19...Nxb4 20.Bxf7+ Kxf7 (20...Kf8 21.Qxb4 Kxf7 22.Re1 is even more in White's favor) 21.Qf4+ Kg8 22.axb4 Re8 23.h4 and White, planning such moves as Nh3 and Kg3, has good winning prospects;

A2) 19...Be6 20.Bxd5 (20.Bb3 a5 is good for Black) 20...Rxd5 21.Qe4 Rad8 and it looks to me as if Black has plenty to compensate for his pawn;

B) 14.Bxc3! smells better and now 14...dxc3 15.Qxc3 Nd5!

Analysis position after 15... Nd5!

B1) 16.Qxe5 Re8 17.Qd4 Nxe3 18.Qxd8 Rxd8 White has succeeded in exchanging queens, but Black's lead in development gives good compensation for the pawn;

B2) 16.Qb3 Be6 17.Rd1 (17.Bc4? Nxb4! is a nice point) 17...Qh4+ 18.g3 Qh6 19.Rxd5 Bxd5 20.Qxd5 Qxe3+ and though White has two minor pieces for a rook, Black's superior piece activity makes up for it;

B3) The most solid is 16.Qd2! but Black still seems to have enough for his lost pawn: 16...Be6 17.e4 (17.Rd1 Qg5 with compensation ; 17.Ne2 Qg5 18.f4 Qf6 with a good game for Black) 17...Nc7! (but not 17...Nf4 18.g3 Qxd2+ 19.Kxd2 Rfd8+ 20.Ke3 which leaves Black with insufficient compensation) a possible continuation being 18.Rd1 Bb3 19.Qxd8 Rfxd8 20.Rxd8+ Rxd8 21.Be2 a5!;

14.fxe4 Qh4+ 15.g3 Qxe4 16.Bxc3 Qxh1

Game position after 16...Qxh1

17.Qf2!

This resource is a key point of 12. f3.

17...Qe4

17...dxc3? 18.Bg2 Qxh2 19.Bc6+ and White wins.

18.Bg2 Qd3 19.Bb2 0-0!

Black courageously gives up a pawn in return for play on the central files.

Worse is 19...dxe3 20.Qe2 and Black will be required to exchange queens, since 20...Qb3 runs into 21.c6!

20.exd4 exd4 21.Qxd4 Re8+ 22.Kf2 Qc2+ 23.Kf1 f6 24.Rc1 Qf5+ 25.Bf3

This is a little too poetic, perhaps. I liked the irony of White's knight defending everything from its original square. But as I look at the game now, I think 25. Qf4 would have been more efficient.

25...Be6

Black's counterplay is reaching a crescendo.

Game position after 25...Be6

Here I sniffed pretty hard and came up with:

26.Re1!

The point is to try to get some major pieces off the board.

26...Rad8 27.Qe4 Qxe4 28.Rxe4 Rd2 29.Re2 Bc4 30.Ke1

Forseeing this, I was able to play 26. Re1.

30...Bxe2

30...Rexe2+ 31.Bxe2 Rxb2 32.Bxc4+ is worse, because it removes White's less active bishop instead of his more active one.

31.Kxd2 Bxf3 32.Nxf3

Game position after 32.Nxf3

The win from here is a technical exercise. White's menacing queenside majority and his material advantage point clearly to the eventual result.

32...Kf7 33.Nd4 a6 34.Nf5 Rd8+ 35.Nd6+ Ke6 36.Ke3 Rd7 37.Kd4 b5!

Black tries to set up a blockade.

38.h4

To help establish the bishop on f4.

38...h5

38...h6?! 39.h5; 38...g5 39.Ke4 and the Black kingside pawns are exposed, for example 39...gxh4 40.gxh4 Rg7 41.Bxf6.

39.Bc1 g6 40.Bf4 Rd8 41.Ke4 Rd7 42.Ne8 Rd1 43.Nc7+ Kd7 44.Nxa6 Re1+ 45.Kd4 Rd1+ 46.Kc3 Ra1 47.Nb8+ Kc8 48.Kb2 Rg1 49.Nc6 g5 50.Na7+ 1-0

The losing move was 13...Nfe4.