Best Case Scenario: An early piece sacrifice by GM Alexander Shabalov against GM Ilya Smirin on one of the top boards in the middle phase of the 1998 New York Open had the spectators buzzing, as Shabalov tried to smash open the position to get at Black's uncastled king. In the cold light of hindsight, however, it appears that in every variation, while appealing tactical themes would surface in the notes, White was struggling to get truly sufficient compensation for his missing knight, in his boardwide search for opportunities.

Shabalov was able to execute further combinations, reducing his material deficit to bishop and pawn versus rook. This probably was the best that his original sacrifice had to offer, and earned him the task of trying to defend a demoralizing endgame. 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 a6 The Kan Variation. 5.Nc3 A popular alternative is 5 Bd3 (much less appealing is 5 c4 Nf6 6 Nc3 Bb4), as then 5 ... b5 6 a4 is good for White (6 ... b4 would not gain any time). Meanwhile, 5 ... Nc6 6 Nxc6 renders Black's early ... a7-a6 useless, and White has a comfortable lead in development. Therefore, Black usually responds with 5 ... Nf6 6 O-O d6 7 c4, and White has the "Maroczy Bind" formation against a Scheveningen-like setup. 5...b5 6.g3 The coming opposition of bishops does not cause Black any problems. I would prefer quicker development with 6 Bd3. Amazingly, the quiet move 6 g3 embroils the position with difficult complications almost immediately. 6...Bb7 7.Bg2 Nf6 8.0-0? A sacrifice of at least a pawn. 8 Qe2 was a viable alternative, but Shabalov instinctively selects the sharpest continuation. 8...b4 9.Nd5

The knight sacrifice is definitely thematic in that it arises frequently in this type of position - the e-file is opened, Black will lose the right to castle, and his queenside pieces become bottlenecked. However, White's compensation for the piece is questionable, because he is not able to cause a permanent breach in Black's position. For example, one would like to already have the queen bishop on g5 [to play the fracturing Bg5xf6 at the right moment] to make such a sacrifice. In any event, White would not have sufficient compensation for the pawn after 9 Na4 Bxe4.

9...exd5 10.exd5 Bc5 The only move. Not 10 ... Be7?? 11 d6, 10 ... Bd6 11 Nf5, or 10 ... d6 11 Re1+. 11.Re1+ Kf8 12.Nf5 The question here is whether anything is to be gained from the move order 12 Bf4! d6 (not 12 ... Bxd4? 13 Qxd4 d6 [13 ... a5 14 Bd6+] 14 Qxb4 and White has too much) 13 Nc6 Qb6. This would lead to the same position as the game continuation, had Black played 15 ... Bb7 instead of 15 ... h6. In this way, 12 Bf4 actually cuts down Black's options. 12...d6 13.Bf4 Bc8 14.Nd4 Qb6
15.Nc6

Now after 15 ... Bxf2+ 16 Kh1, Black's defensive task is just harder because d6 is loose: then 16 ... Bc5 17 a3! is rough, while 16 ... Bxe1? is very bad after 17 Bxd6+. A more sensible choice for Black is 15 ... Bb7 (transposing to the position which White could have forced - see the note to White's twelfth with the continuation 12 Bf4 d6 13 Nc6 Qb6). Then White has the diabolical 16 Qd4??! with the idea of making a draw after 16 ... Bxd4 17 Bxd6+ Kg8 18 Ne7+ Kf8 19 Ng6+, etc. And 16 ... Bxc6 fails to 17 Bxd6+,

while 16 ... Nxc6 may be unclear after 17 Bxd6+ Kg8 18 Qxc5 Qxc5 19 Bxc5. However, simply 16 ... Nbd7! Refutes the absurd idea behind 16 Qd4, as Black then threatens to follow up with 17 ... Bxc6. Much more logical after 15 ... Bb7 is 16 Qf3 (not very convincing is 16 Qe2 Nxc6), simply jamming up the long diagonal, with some compensation after 16 ... Nbd7 17 a3. 15...h6! A very fine defensive move. Now 16 Qf3? is very bad due to 16 ... Bg4, while on 16 Qe2 Black can stabilize with 16 ... Nxc6 17 dxc6 Be6. 16.Qd2 Bb7 Now Smirin can focus his energies on dislodging the knight from c6. 17.a3! This consistently pops up as a theme, trying to poke at the bishop on c5 which is holding the precious d6 square. 17...a5 18.axb4 axb4 19.Rxa8 Bxa8
20.Ra1

20...Nxc6! This is much better than 20 ... Bxc6, as then after 21 dxc6, 21 ... Nxc6? is impossible as after 22 Bxd6+ Kg8 23 Ra8+ Kh7 24 Qd3+ g6, White recovers his piece with 25 Ra6, winning. 21.Bxd6+! A unique type of interpolation to get Black to commit to taking on d6. If instead 21 dxc6 Bxc6 22 Bxd6+, Black would just play 22 ... Kg8, while on 21 Rxa8+, Black holds the fort with 21 ... Nd8. 21...Bxd6 22.dxc6 Shabalov is now two pieces down, but he is attacking two bishops 22...Qb8

23.Rxa8 A narrow miss is 23 c7 Qxc7 24 Rxa8+ Ne8 25 Qe2 Qe7 and Black holds on to his extra material. 23...Qxa8 24.Qxd6+ Kg8 25.c7 Qc8

26.Bf1 Ne8 Not 26 ... Kh7? 27 Bd3+ g6 28 Ba6 and White emerges ahead. 27.Qd8 Qxc7 28.Qxe8+ Kh7 29.Qe4+ g6 30.Bc4 Rd8 Black firmly takes over the initiative, as he has all the chances with a rook against a bishop and pawn. 31.Bb3 Shabalov elects to preserve his option to later play c2-c3, liquidating the queenside, but I think it would have been better to play 31 b3, shuttling the bishop between d3 and c4. 31...Rd1+ 32.Kg2 Rd2 33.Qf3 Kg7 34.h4 Rd6 35.Kh2 Rf6 36.Qe2 Qc5 37.Kg1 Qd4 Eliminating any vestige of White counterplay.

38.c3 bxc3 39.bxc3 Qxc3 40.Bd5 Rd6 41.Qa2 Qe5 42.Bf3 Qe1+ 43.Kg2 Rd2 44.Qa7 Qe5 Because the position is lost for White if the queens are exchanged, he has a very difficult defensive task. 45.Qa6 Qd4 46.Qf1 Ra2 47.Qe1 Kf6 48.Qf1 Qd2 49.Kg1 Ke7 50.Kg2 Kf8 51.Bg4 Rc2 52.Bf3 Kg8 53.Be4 Rc1 54.Qb5 Qe1 55.Qe8+ Kg7 56.Qe5+ Kh7 57.Kf3 Qc3+ 58.Qxc3 Rxc3+ 59.Kf4 Kg7 60.g4 Rh3