1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Bc5 5.c3 0-0 6.d4 Bb6 7.Bg5 h6 8.Bh4 d6 9.Bxc6 bxc6

Analysis postition after 9...bxc6

10.dxe5

10.Nbd2 Re8 11.Re1 (11.dxe5 dxe5 transposes into the text below and avoids 11...Qd6) 11...exd4! (11...Bg4?! 12.dxe5 dxe5 13.Qa4 Bxf3 14.Nxf3 Qd6 15.Rad1 Qe6 16.Bg3 Nd7 17.Qb3 and Black loses a pawn) 12.Nxd4 g5 (I believe that 12...Bd7 is also adequate, for example 13.Qf3 g5 14.Bg3 Bg4 15.Qd3 c5 and now 16.Nb5, which was a strong idea in the analogous position after 9.a4 a5, is pointless because of 16...a6) 13.Bg3 and here my idea is 13...Bg4

Analysis postion after 13...Bg4

What follows is my analysis. (The alternative 13...Nd7 14.Qa4 strongly favors White.)

A) 14.f3 Be6

A1) 15.Qa4 Qd7 16.e5 c5 17.Qxd7 Bxd7 18.Nc2 Nh5 and it looks even;

A2) 15.Kh1 Qd7 16.Nxe6 Qxe6

A2a) 17.Qa4 Qd7 18.Nb3 Rab8 19.Nd4 (19.Rad1 Nh5) 19...Bxd4 20.cxd4 c5! with equality;

A2b) 17.Nb3 Nh5 18.Nd4 Nxg3+ 19.hxg3 Qg6 20.Qa4 g4 21.f4 Rxe4 22.Qxc6 Rxe1+ 23.Rxe1 Rb8 and Black has sufficient counterplay;

A3) 15.e5 Nh5 16.Bf2 Bd7 17.Nc4 c5 18.Nb3 Be6 with a reasonable game for Black;

B) 14.Qa4 c5 (14...Bd7?! 15.Nxc6 Qc8 16.e5 Qb7 17.exf6 Bxc6 18.Qg4 Bxg2 19.Bf4 Qd5 20.Qxg2 Qxg2+ 21.Kxg2 gxf4 22.Re7 favors White) 15.Nc6 (15.Nf5 Bxf5 16.exf5 Qd7 17.Qxd7 Nxd7 18.Rad1 Nf6 with approximate equality) 15...Bd7! (worse is 15...Qd7 16.Nf1) 16.e5 Nh5 and now:

B1) 17.Qe4 Bxc6 18.Qxc6 Nxg3 19.hxg3 Re6! with an even game (not so good is 19...Rxe5 20.Rxe5 dxe5 21.Nc4) ;

B2) 17.Nf3 Qc8 18.exd6 (18.Qe4 Bf5 19.Qd5 Be6) 18...Nxg3 19.hxg3 cxd6 20.Rxe8+ Qxe8 21.Re1 Be6 is equal;

C) Unambitious is 14.Qc2 Qd7 15.h3 (15.Nc4 Nh5) 15...c5 16.Nf5 (16.N4b3 Be6 17.a4 a5) 16...Bxf5 17.exf5 Rxe1+ 18.Rxe1 Re8 with a perfectly adequate game for Black;

10.a4 a5 transposes into the Variation 9. a4.

10...dxe5 11.Nbd2

A major alternative is 11.Qxd8 Rxd8

Analysis position after 11...Rxd8

It looks as if Black must simply lose a pawn, but appearances can be deceiving.

A) Of course, White doesn't have to take: 12.Re1 Ba6 13.Na3 (13.Nxe5 g5 14.Bg3 Nxe4 15.Nxc6 Re8 16.Na3 f5 17.Ne5 Bb7 is good for Black) 13...g5 14.Bg3 Nd7 and now:

A1) 15.b4 f6 16.c4 (16.Nd2 Nf8 17.Ndc4 Rd7 18.Rad1 Rad8 was played in DeFirmian-Nogly, Wichern Open, 1997. The game is even. Black went on to draw, not a bad result for a 2220 against a GM.) 16...h5 17.Rac1 (17.c5 Nxc5 18.bxc5 Bxc5 19.Nc2 h4 20.Nxh4 gxh4 21.Bxh4 Kf7 looks good for Black) 17...h4 18.Bxh4 gxh4 19.c5 Nxc5 20.bxc5 Ba5 though Black is effectively a pawn down, the two bishops offer ample compensation;

A2) 15.Nxe5 Nxe5 16.Bxe5 Rd2 with a good game for Black;

B) So why not take the pawn? 12.Nxe5 g5 13.Bg3 (13.Nxc6 Re8 14.Bg3 Nxe4 transposes) 13...Nxe4 14.Nxc6

Analysis position after 14. Nxc6

Does Black have compensation?

B1) I analyzed the prosaic 14...Re8 which nevertheless turns out to be enough:

B1a) 15.Re1? Bb7

B1a1) 16.Nd4 Nxg3 17.Rxe8+ Rxe8 18.fxg3 (18.hxg3 Bxd4 19.cxd4 Re1+ 20.Kh2 Be4) 18...Re2;

B1a2) 16.Ne5 f5 17.Nd7 f4 18.Nxb6 fxg3;

B1b) 15.Nd4! c5

B1b1) 16.Nf3 f5 17.h3 (worse is 17.Re1 f4 18.Nbd2 Nd6) 17...Nxg3 18.fxg3 Bc7 is slightly better for Black;

B1b2) 16.Nb5! Nxg3 17.fxg3 (17.hxg3 Ba6 18.N1a3 Rad8! with good compensation for the pawn) 17...Re2 18.Rf2 (worse for White is 18.N1a3 Bb7) 18...Re1+ 19.Rf1 Re2 and Black has plenty of compensation;

B1c) 15.Na3 is slightly favorable to Black: 15...f5 16.Nc4 (16.Nd4 f4 17.f3 Ba6 18.Rfe1 Nxg3 19.hxg3 fxg3) 16...f4 17.Nxb6 axb6 18.f3 Bb7 (18...Nxg3 19.hxg3 fxg3 20.Rfe1 Bd7) 19.fxe4 (worse is 19.Nd4 Nd2 20.Rfd1 Nc4) 19...Bxc6 20.Bf2 Bxe4;

B2) I later discovered some games with 14...Ba6! 15.Nxd8 Rxd8 16.Rc1 (16.Re1 f5 17.Na3 Rd2 18.Red1 Rxd1+ 19.Rxd1 f4 20.Bxf4 gxf4 21.c4 Nxf2 with an excellent game for Black in Hernandez-Majigsuren, Yerevan 1996.; 16.c4 f5 17.c5 Bxc5 18.Bxc7 Rd7 19.Rc1 Rxc7 20.Nc3 with an easy win for Black in Gass-Gabriel, Bundesliga 1990.) 16...f5 17.Kh1 f4 18.f3 Nxg3+ 19.hxg3 Be3 20.Re1 Bf2 21.Rc1 Be3 22.Re1 Bf2 with a draw;

11.Qa4 Qd6 12.Na3 Nd7 (12...Qe6 13.Bxf6 Qxf6 14.Rad1 Bg4 was even in Rosenthal-Jukic, Bundesliga 1993.) 13.Nc4 Qe6 14.Nfd2 ECO-3 claims a slight advantage for White here, but I disagree. (14.Nxb6 cxb6 was equal in Rachels-Torre, Manila 1990.) 14...a5 (14...Nc5 15.Qc2 Ba6 16.b3 Nb7! 17.Rfe1 Nd6 18.Ne3 Bxe3 19.Rxe3 f6 and the game is even.) 15.b4 (15.Rad1 Ba6 16.b3 Bc5! and Black, threatening ...Bb5, has a big advantage) 15...Ba6 16.bxa5 Nc5 17.Qb4 Nd3 18.Qa4 Nc5 19.Qb4 Ba7 and White is in some difficulty.

11...Re8

11...Qd6

Analysis position after 11...Qd6

This appears to be a viable alternative, but players of the Black pieces should be aware that White can avoid this line by playing 10. Nbd2 Re8 11. dxe5 dxe5 12. Qc2, which transposes into the text below. Perhaps someone can discover a good way to avoid this transposition.

A) I analyzed 12.Nc4 Qxd1

A1) 13.Raxd1 g5 14.Bg3 Nxe4 15.Nfxe5 Be6 16.Rfe1 Nxg3 17.hxg3 Rfe8 18.Nd2 (18.b3 Bd5 is satisfactory for Black) 18...a5 19.Nxc6 Bxa2 and now:

A1a) 20.b3 a4 21.Rxe8+ Rxe8 22.bxa4 Bd5

A1a1) 23.Nd4 Ra8 24.Ra1 Ba5 25.Ra3 (25.c4 Bxd2 26.cxd5 Bc3 27.Rd1 Rxa4 28.Nb5 Be5 and the game is even; 25.Rc1 Bb6 26.Ra1 Ba5 is also equal) 25...Re8 Black has enough for his pawn;

A1a2) 23.Nb4 Be6 Black has sufficient compensation for his pawn.;

A1b) 20.Rxe8+ Rxe8 21.b3 Re6 22.Nd4 Bxd4 23.cxd4 Rb6 with an even game;

A2) 13.Rfxd1 g5 14.Bg3 Nxe4

A2a) 15.Nfxe5 Be6 16.a4 (16.Nd2 Nxg3 17.hxg3 c5; 16.Nxb6 axb6!) 16...Bd5 with equality;

A2b) 15.Ncxe5 f5 16.Re1 Bb7 17.Rad1 c5 18.Ng6 Rfe8 19.Rd7 (19.Be5? Re6µ) 19...f4 20.Ne7+ Kf8 21.Ng6+ Kg8 22.Ne7+ with an even game;

B) 12.Qa4 Nd7 13.Nc4 Qe6 14.Nfd2 Nc5 15.Qc2 Ba6 16.b3 f6 17.a4 Qg4 18.Bg3 was played in Rodriguez-Sariego, Bayamo 1991.

B1) I don't see why Black didn't play 18...Bxc4 19.bxc4 (19.Nxc4?! Qxe4 is hard to justify) 19...a5 with what appears to be a perfectly even game;

B2) Instead, Black went in for 18...Qe2 and was worse after 19.Rfe1;

C) 12.Bg3 Nd7 13.Nc4 (13.Qa4 Qe6) 13...Qxd1 14.Rfxd1 f6 is even;

D) 12.Bxf6 Qxf6 13.Nc4 Re8 14.Qa4 Bg4 is even;

An idea that may be worth considering is 11...Qe7 12.Qc2 (12.Qa4 Qe6 Uhlmann-Joerg, Bavaria 1994) 12...Qe6 13.c4 Nd7 Soylu-Calderin, Elista 1998.

12.Qc2

12.Qa4? g5 13.Bg3 g4 Hellers-Finegold, Reykjavik 1990.;

12.Re1 Qd6 13.Nc4 Qxd1 14.Raxd1 Nd7 followed by ...f6, is even.;

12.Bg3 Nd7 13.b4 Qe7 14.Nc4 f6 15.Nfd2 Qe6 was equal in Hellers-Van der Weil, Haninge 1990.

12...Bg4!?

Almasi's recommendation.

Worth considering is 12...Ba6.

Almasi-Gulko, Pamplona 1997 went 12...g5 13.Bg3 Nh5 14.c4 Nxg3 15.hxg3 and here Gulko should have played 15...Bc5 (15...Qe7 16.Nb3 Bc5 17.Qc3 was played instead, with a substantial advantage for White) 16.Nb3 Bf8 with a manageable disadvantage, according to Almasi.

13.h3

13.Qa4 Re6 14.Rad1 Qe8 15.Nc4 Bxf3 16.gxf3 Nd7 and Black, continuing ...g5 and ...f6, can defend satisfactorily.;

13.Bg3 Nd7 14.Rad1 Qe7

A) 15.c4 Bc5 16.Qa4 (16.Nb3 Bxf3 17.gxf3 Bd6 18.Na5 c5 19.Nc6 Qe6 20.Qd3 g5 and White's bishop is doomed to a life of inactivity.) 16...Qe6 17.b4 (17.Qa5 Bb6 18.Qa3 a5) 17...Bf8 18.Qa5 c5 19.a3 (19.bxc5 c6) 19...Bd6 and it's difficult to see much advantage for White.;

B) 15.h3 Bh5

B1) 16.c4 Bc5 17.Qa4 (17.Nb3 Bxf3 18.gxf3 Bd6 19.Na5 Qe6 20.Kg2 c5 is good for Black, who will continue ...g5 and ..f6, imprisoning White's bishop) 17...Qe6 18.b4 (18.Qa5 Bb6 19.Qa3 a5) 18...Bf8 19.Qa5 c5 20.a3 Bd6 appears satisfactory for Black;

B2) 16.Qa4 Qe6 17.Qc4 f6 18.Qxe6+ Rxe6 19.Nc4 Re7 (19...Rd8 20.b4 Ree8 21.a4 is awkward for Black) 20.b3 Bf7 21.Nh4 (21.Nfd2 a5 22.Rfe1 Nc5) 21...Bxc4 22.Nf5 Rf7 23.bxc4 Nc5 and Black is certainly no worse.

13...Bh5 14.g4 Bg6 15.Rfe1 Qd6

Analysis position after 15...Qd6

This position arose in Van der Weil-Winants, Lyon 1990, but by a different move order.

16.Nc4

This certainly seems to be the critical move. However, there is no practical experience of which I am aware. All that follows is my own analysis. Van der Weil continued instead 16.Rad1 Qe6 17.Qb3 Rad8 18.Qxe6 Rxe6 19.Nc4 Rxd1 20.Rxd1 Bxe4 21.Rd8+ Re8 22.Bxf6 Rxd8 23.Bxd8 Bxf3 24.Nxe5 Be4 with equality.

16...Qe6 17.Ncxe5

Less promising is 17.Nfxe5 Bxe4 18.Qe2 (18.Rxe4 Nxe4 19.Qxe4 f6 20.Bg3 fxe5 21.Nxe5 Rad8 looks good for Black) 18...g5 19.Bg3 Rad8 with an even game.

17...Bxe4

Worse is 17...Nxe4 18.Nxg6 (18.Qa4 Nc5 19.Qxc6 Qxc6 20.Nxc6 Be4 gives Black good compensation for his pawn) 18...Qxg6 transposing into a note to move 20 below (20...Qg6), but reaching the position two moves sooner.

18.Qa4

18.Rxe4 Nxe4 19.Qxe4 f6 20.Bxf6 (20.Bg3 fxe5 21.Nxe5 Rad8 and Black's activity is worth more than White's pawn) 20...Qxf6 with sufficent compensation for the pawn.

18...Bxf3

18...Bd5 19.Rad1 looks somewhat better for White, but it may be worth trying, for example 19...Bc5.

19.Nxf3 Ne4

19...Qd5 20.Kg2 Rxe1 21.Rxe1 Re8 22.c4! Qc5 23.Rxe8+ Nxe8 24.Bg3 (or simply 24.Qc2 Nd6 25.b3) 24...Nd6 25.Ne5 Qd4 26.Qc2 and White not only has the better pawns, but also the more active pieces.

20.Qc2!

20.Nd2 Nc5 21.Rxe6 Nxa4 22.Rxe8+ (22.Rxc6 Nxb2 is dead even) 22...Rxe8 23.Rb1 Re2 24.Nc4 Re4 25.Nd2 (25.b3 Nxc3 26.Rb2 f5 is better for Black) 25...Re2 is a draw;

20.Bg3 Qd5

A) 21.Kg2 Ng5 22.Qf4 (22.Qd1 Rxe1 23.Qxd5 cxd5 24.Rxe1 Ne4 and the game is roughly even) 22...Rxe1 23.Rxe1 Qxa2 24.Re7 (24.Nxg5 hxg5 25.Qd2 Qd5+ 26.Qxd5 cxd5 27.Re5 d4 looks slightly better for Black) 24...Qd5 with equality;

B) 21.Rad1 Nxc3 with balanced chances;

20.Rad1 Qg6 21.Nd4 h5 22.Nf5 (22.f3 hxg4 23.hxg4 Qh6 and White's king is badly exposed) 22...hxg4 23.Rxe4 Qxf5

A) 24.Rxg4 Re2 25.Qf4 (25.Qxc6 Bxf2+ 26.Kh1 Rae8 and Black's attack is very strong) 25...Qxf4 26.Rxf4 Rxb2 with advantage to Black;

B) 24.Rf4 Qb5 25.Rxg4 Qxa4 26.Rxa4 Re2 and Black is no worse.

20...f5

Analysis position after 20...f5

20...Qc4 21.Rac1! threatening b3, is good for White; for example 21...Ba5 22.Rcd1;

20...Qg6

Analysis position after 20...Qg6

Black's move gives rise to a position that could have been reached two moves sooner if Black had gone in for 17...Nxe4 18. Nxg6 Qxg6. It is my conclusion that this position is favorable for White, but because of its importance in both branches, I will treat it in some detail. Maybe someone can discover an improvement for Black in the critical line D below.

A) 21.Rad1? Nxf2;

B) 21.Re2 f5 22.Nd2 h5 23.Nxe4 fxe4 24.g5 Qf5 25.Rae1 Qxh3 26.Rxe4 Rf8 with counterplay for Black;

C) 21.Kg2 (White consolidates his kingside) 21...h5 (21...f5 22.Nh2 Qf7 23.f3 Nd6 24.b3 is no good for Black)

C1) 22.Nh2 hxg4 23.hxg4

Analysis position after 23. hxg4

23...Bc5! (threatens ...Bd6) 24.Bg3

C1a) 24...Bd6 appears satisfactory, for example 25.Bxd6 cxd6 26.f3 Nf6 (or 26...Nc5 ) 27.Qxg6 fxg6 28.f4 (28.Rad1 g5!) 28...Nd5 29.Kg3 Rf8 30.Re4 Rab8 31.c4 Nf6 32.Re2 d5;

C1b) More enterprising is 24...f5 25.Qe2 (25.b4 Bb6 26.a4 a5) 25...Qf7 (25...Nxg3? 26.Qc4+) 26.Qf3 Nxg3 27.Kxg3 Bd6+ 28.Kg2 Rxe1 29.Rxe1 Qxa2 30.Qxc6 Rf8 with balanced chances;

C2) 22.Ng5 Bxf2 23.Bxf2 Nxg5 24.Qxg6 fxg6

C2a) 25.Rad1 Rxe1 26.Rxe1 (26.Bxe1 Re8 27.Bh4 Nf7 is roughly even) 26...hxg4 27.hxg4 Kf7 28.Bc5 (28.Be3 Ne6 29.Kf3 a5 is similar) 28...Ne6 (a very good post for the knight) 29.Rf1+ Ke8 30.Be3 a5

Analysis position after 30...a5

In spite of Black's weak pawns, I believe this ending is roughly even. Black has chances of counterplay on the b-file.;

C2b) 25.Rxe8+ Rxe8 26.h4 (similar is 26.gxh5 gxh5 27.h4 Nf7! 28.Bxa7 Re2+) 26...Nf7! (keeping the e-file free for the rook) 27.Bxa7 Re2+ 28.Kg3 (28.Kf3 Rxb2 29.Bd4 g5) 28...Rxb2 29.Bd4 (29.Bc5 Rc2; 29.a4 Rb3) 29...Nh6 and Black has counterplay in spite of White's passed a-pawn;

D) 21.Nd4! (White hastens f2-f3) 21...Bxd4 22.cxd4 Ng5 23.Rxe8+! Rxe8 24.Qxg6 Nf3+ 25.Kf1 fxg6 26.Bg3 and the ending is quite favorable to White, for example 26...Nxd4 27.Rd1 c5 28.Bxc7 Re6.

21.Nd2

21.gxf5 Qxf5 allows Black good counterplay;

21.Rad1 fxg4 22.Nd2 Qg6 23.Nxe4 gxh3+ 24.Bg3 Rxe4 25.Qxe4 Qxg3+ 26.Kh1 Bxf2 offers Black good compensation for the exchange.

21...Qg6 22.Nxe4 fxe4 23.Rad1 Re6

Analysis position after 23...Re6

White is somewhat better, but Black, intending ...Rf8 and ...h5, has counterchances on the kingside light squares.