Commentary: Nakamura-Li, World Team Championship Round 5

This next commentary game also features American super-GM Hikaru Nakamura as White.  His play in this game is something to be emulated, as he expertly calculates, evaluates and makes winning decisions all along the way.  His play is dynamic and very instructive in the way he sacrifices a pawn in the opening, then immediately takes over the initiative with active piece play, tying Black in knots using repeated threats and then regaining his material while maintaining his positional advantage. Some highlights:

• Black's decision to take the pawn on d4 may not be the worst move in the position, but it certainly leads to strategic problems for him. One of the variations included shows how Black could retain the material, albeit with major difficulties as White has more than sufficient compensation. In the game continuation, Black ends up with less than nothing to show for his troubles and his dark-square bishop is also traded off, giving White a major strategic and tactical advantage.
• White's b-pawn is tactically protected or "poisoned" for the entire game in a remarkable fashion, due to a variety of different tactics.
• White makes the practical decision to exchange down to a winning endgame, rather than go in for additional middlegame complications. The point is that if you can calculate to a point where you evaluate you can win the game, it doesn't matter whether computer analysis afterwards would give you additional points for a different continuation.

[Event "World Teams 2013"] [Site "Antalya TUR"] [Date "2013.11.30"] [Round "5.5"] [White "Nakamura, Hikaru"] [Black "Li, Chao b"] [Result "1-0"] [ECO "D90"] [WhiteElo "2786"] [BlackElo "2679"] [Annotator "ChessAdmin/Houdini"] [PlyCount "77"] [EventDate "2013.11.26"] [WhiteTeam "USA"] [BlackTeam "China"] [WhiteTeamCountry "USA"] [BlackTeamCountry "CHN"] 1. Nf3 Nf6 2. c4 g6 3. Nc3 d5 {we now have a Gruenfeld-English setup on the board} 4. cxd5 Nxd5 5. Qb3 {White has a number of different options here. The text move scores well at 62 percent in the database.} Nb6 {the overwhelming choice is to retreat, rather than exchange off the knight. Interestingly, Nakamura has also played this line as Black (see move 7 game variation).} 6. d4 Bg7 7. e4 {developing the dark-square bishop here is more popular, as in this earlier top-level game featuring Nakamura as Black:} (7. Bf4 Be6 8. Qa3 Nc6 9. e3 O-O 10. Be2 a5 11. O-O Nb4 12. Rfc1 c6 13. Be5 Bh6 14. Ne4 Nd7 15. Nc5 Nxc5 16. Rxc5 Nd5 17. Bc4 Qb6 18. e4 Nf6 19. Bxe6 fxe6 20. Rc2 Qb4 21. Qd3 Rac8 22. Ne1 Nd7 23. Bg3 c5 24. d5 Nf6 25. d6 Qxe4 26. dxe7 Rfe8 27. Qb5 Qb4 28. Qf1 Rxe7 29. Nd3 Qb6 30. Ne5 Nd5 31. Qe2 Nb4 32. Rc4 Bg7 33. Bh4 Ree8 34. Rd1 Nc6 35. Nd7 Qxb2 36. Qxb2 Bxb2 37. Rxc5 Rc7 38. Bg3 Rcc8 39. Rb5 Bg7 40. Rxb7 Nb4 41. a4 Nd5 42. Nb6 {1/2-1/2 (42) Kramnik,V (2800)-Nakamura,H (2758) Moscow 2011 }) 7... Bg4 {Black indirectly pressures the White center with this move. Strategically, we have a situation with a classic pawn center versus an Indian-style development by Black designed to put pressure on it.} 8. Bb5+ { White knows well that Black will block with the c-pawn. The point of the move is to cause Black to take away the c6 square from his own knight, forcing it to a lesser square that does not influence d4.} c6 9. Ng5 {we are still in theory here, but this is nonetheless an instructive in-between move. Rather than simply retreat the bishop, White looks to improve his position first. But what about the d-pawn?} e6 $146 {all previous games had featured castling, as in the following high-level contest:} (9... O-O 10. Be2 Bxe2 11. Nxe2 e5 12. dxe5 h6 13. Nh3 N8d7 14. f4 Nc5 15. Qc2 Nd3+ 16. Kf1 f5 17. exf6 Rxf6 18. Be3 Rf7 19. Rd1 Rd7 20. Nf2 Nxb2 21. Rxd7 Qxd7 22. Bc1 N6c4 23. h4 b5 24. Rh3 Rd8 25. Rg3 Nd1 26. e5 Qf5 27. Qxf5 gxf5 28. Rg6 Nce3+ 29. Kg1 c5 30. Ng3 Kf7 31. Ra6 Bf8 32. Rxa7+ Be7 33. Nxd1 Rxd1+ 34. Kf2 Ng4+ 35. Ke2 Rxc1 36. Nxf5 Ke6 37. Nxe7 h5 38. Nc8 c4 39. Nd6 Rc2+ 40. Kf3 Rc3+ 41. Ke4 Re3+ 42. Kd4 Rd3+ 43. Kc5 c3 44. Rc7 c2 45. Kxb5 Ne3 46. f5+ Kxe5 47. Nc4+ Nxc4 48. Rxc4 Rd5+ 49. Kc6 Rd6+ 50. Kc7 Rd2 51. a4 Rd4 52. Rc5+ Kf6 53. a5 Rd5 54. Rc6+ Kxf5 55. a6 Ra5 56. Kb6 Ra2 57. a7 Rb2+ 58. Kc7 Ra2 59. Kb7 Rb2+ 60. Kc7 {1/2-1/2 (60) Topalov, V (2769)-Caruana,F (2786) Bucharest 2012}) 10. Be2 Bxe2 11. Nxe2 Bxd4 {in the analagous position in the variation where Black castles first, only one person in the database dared take the central pawn, and lost.} 12. Nxd4 Qxd4 {What is going on here, then? Black is currently a pawn up and White apparently has no immediate threats that could recoup the material. For Class players, most of us would therefore hate to be White. One clue, however, is that Houdini rates this position as a half-pawn advantage to White. Let's see how this is translated on the board.} 13. Bd2 {threatening the skewer on c3} Qc5 (13... O-O 14. Qh3 $16) 14. Rc1 Qe7 15. a4 {the idea is to kick the Nb6 and expose the half-open b-file.} N8d7 $6 {this allows White to implement his plan without a fight.} (15... h6 $5 16. Nf3 Na6 17. a5 Nc8 $14 {would leave Black's pieces awkwardly placed and not set up well for castling, but at least he would retain the pawn for some compensation.}) 16. a5 Nc8 17. Qxb7 Rb8 18. Qxc6 { now White is a pawn up and still has a much better position. The b2 pawn is tactically protected from the Rb8, given the lack of other defenders of the Nc8.} O-O 19. Nf3 Rd8 (19... Rxb2 {is still not possible:} 20. Qc3 Rb8 21. Bh6 {and Black either loses material or gets mated on g7.}) (19... Nd6 $16 { is Houdini's suggestion, at least getting the knight back in the fight.}) 20. O-O e5 (20... Rxb2 $2 {although the Black rook is no longer on f8 to be threatened, taking the b2 pawn is *still* not possible, due to yet another tactical comeback involving Bh6:} 21. Qxc8 Rxc8 22. Rxc8+ Nf8 23. Bh6 $18) 21. Rfd1 $18 {White brings his remaining piece into the game and Black is amazingly helpless in this position. As noted in the previous variation, he actually has a back rank problem, despite both his rooks being there, due to the holes around his king and the unopposed White dark square bishop.} f6 22. Bh6 Kf7 {getting off the back rank, but White has other threats as well. Note how Black's pieces, especially the knights and rooks, are uncoordinated and get in each others' way.} 23. Qc4+ {Nakamura no doubt calculated the next sequence through to the endgame.} (23. Qc7 {played immediately, according to Houdini, would give White a greater advantage in the middlegame, for example} Ke8 24. Qc4 Nf8 25. Rxd8+ Qxd8 26. Qg8 Qd6 27. Nd2 {with Nc4 coming}) 23... Qe6 24. Qc7 {White exploits Black's weak rooks and overloaded Nd7.} Ke8 25. Rxd7 Qxd7 26. Qxb8 Nb6 {a clever way to get the knight into the game, but White returns the favor with his own counterattack.} 27. Nxe5 Rxb8 28. Nxd7 Nxd7 { White has engineered a transition to what should be a winning endgame.} 29. h4 {preventing Black from attempting to shut the bishop in with ...g5} Rb5 (29... Rxb2 {still does not work for Black:} 30. Rc8+ Ke7 31. Ra8) 30. Rc8+ Ke7 31. Rh8 Rxb2 32. Rxh7+ {White decides to munch on Black's kingside pawns rather than on the queenside one.} Ke6 33. Be3 a6 34. Rg7 Ne5 35. Bc5 f5 36. Bd4 Rb1+ 37. Kh2 Rb5 38. Bxe5 Kxe5 39. Rxg6 {Black now acknowledges that he is lost.} 1-0