Ultimate Challenge: A Sharp King's Indian Reversed

The following game is an exciting encounter in the King's Indian with reversed colors. Thomas_A_Anderson as White attacked on the kingside, while Zor as Black tried to get through on the queenside, sacrificing a piece in return for two strong pawns. The game became razor sharp and eventually a just draw was agreed.
[Event "Long 90' + 30'', Unrated, Infinity Ches"] [Site "?"] [Date "2017.04.26"] [Round "3"] [White "Thomas_A_Anderson (ct)"] [Black "Zor"] [Result "1/2-1/2"] [ECO "A08"] [WhiteElo "2646"] [BlackElo "2563"] [Annotator "GM Lutz"] [PlyCount "105"] [EventDate "2017.??.??"] [TimeControl "5400+30"] 1. Nf3 {[%emt 0:00:26]} ({To get an understanding of the mutual ideas that occurred in the game, here's a sample from a 'human' game in the King's Indian with the 'right colors', i.e. Black fianchettoes the king's bishop and White gets an advantage in space.} 1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. O-O Nc6 8. d5 Ne7 9. Ne1 Ne8 10. Nd3 f5 11. f3 f4 12. Bd2 g5 13. c5 Ng6 14. cxd6 cxd6 15. Nf2 h5 16. a4 Rf7 17. h3 Bf8 18. Ra3 Rg7 19. a5 Nh8 20. Kh2 Nf6 21. Rh1 Nf7 22. Kg1 Nh6 23. Kf1 Rg6 24. Nb5 a6 25. Nc3 g4 26. hxg4 hxg4 27. Na4 gxf3 28. gxf3 Bd7 29. Nb6 Rb8 30. Nxd7 Qxd7 31. Ke1 Qg7 32. Bf1 Be7 33. Ke2 Ne8 34. Qe1 Nc7 {and a draw was agreed, Bischoff,K-Rapport,R (Deizisau 2012)}) 1... d5 {[%emt 0:00:00] [%B]} 2. g3 {[%emt 0:00:44]} c5 { [%emt 0:00:00] [%B]} 3. Bg2 {[%emt 0:00:14]} Nf6 {[%emt 0:00:00] [%B]} 4. O-O { [%emt 0:00:19]} e6 {[%emt 0:00:01] [%B]} 5. d3 {[%emt 0:02:47]} Nc6 {[%emt 0: 00:01] [%B]} 6. Nc3 {[%emt 0:02:00]} Be7 {[%emt 0:06:12]} 7. e4 {[%emt 0:02:51] } d4 {[%emt 0:04:43]} 8. Ne2 {[%emt 0:00:06]} e5 {[%emt 0:00:15]} 9. h3 { [%emt 0:16:49]} O-O {[%emt 0:01:40]} 10. Ne1 {[%emt 0:04:09]} Ne8 {[%emt 0:03: 11] By comparison to the game Bischoff-Rapport after Black's 9th move (see above) White is two tempi up (one tempo for being White, the other since Black played ...e7-e6 und ...e6-e5). But one of White's extra moves is the not so useful h2-h3, so White is only one real tempo up. One can discuss by what degree this improves White's position. I personally nevertheless would like to play Black's position.} 11. f4 {[%emt 0:00:12]} f6 {[%emt 0:00:45]} 12. f5 { [%emt 0:04:05] White goes for the typical 'running game', putting all his hopes on the attack on the kingside with g3-g4-g5...} b5 {[%emt 0:06:17] ... but Black is not lazy either, preparing ...c5-c4. Who comes first on 'his' wing, White or Black? One should note that if Blacks get through on the queenside it means a material gain - whereas White breaking through on the kingside means mate.} 13. g4 {[%emt 0:01:39]} Nd6 {[%emt 0:11:59]} 14. h4 { [%emt 0:03:45]} Bd7 {[%emt 0:04:10]} 15. Nf3 {[%emt 0:03:51]} Rc8 {[%emt 0:03: 27]} 16. Rf2 {[%emt 0:07:17] The key point in this structure is to keep an eye on both wings. The rook on f2 might move to g2, supporting the thrust g4-g5. On the other other side, the rook helps defending the c2-pawn and prepares moving the bishop to f1, another defensive resource.} c4 {[%emt 0:04:32]} 17. Ng3 {[%emt 0:07:55]} Nf7 $6 {[%emt 0:01:53] Black also plays a double purpose move: Moving the knight to f7 stops g4-g5 and frees the e7-bishop for an attack on the queenside. But it is more accurate to take on d3 first.} (17... cxd3 $1 18. cxd3 Nf7 $15 {would have been Black's most accurate continuation, along the lines of the game Bischoff-Rapport above. Compared to the game, when Black closed the queenside with ...c4-c3, Black has a direct avenue to White's camp along the c-file. On the other hand, White has more 'breathing space' for his pieces and can develop his c1-bishop and his a1-rook. I prefer Black here.} ) 18. Bf1 {[%emt 0:02:20]} c3 $6 {[%emt 0:00:59] This is a critical junction. By moving the pawn to c3, Black corners in the a1-rook and the c1-bishop. But on the other hand Black's attacking potential on the queenside is sharply reduced, as it is not so easy anymore to open lines.Therefore Black's choice is dubious and taking on d3 has to be preferred.} (18... cxd3 {was still the most natural continuation. But here White has the additional option of taking back with the bishop:} 19. Bxd3 Nb4 {with an unclear position. Eventually Black will have to exchange on d3, relieving White from the bishop that is usually buried behind the pawn. So if Black had intended to take on d3, it should have happened one move earlier.}) 19. b3 {[%emt 0:01:14] Of course White tries to avoid opening the queenside.} Kh8 {[%emt 0:01:34]} (19... b4 { was possible, so that Black can open the a-file with ...a7-a5-a4. But then Black doesn't have the b4-square for his pieces. So if Black follows up with .. .a5-a4xb3, White can take back with cxb3, keeping the position as closed as possible.}) 20. a3 {[%emt 0:03:15] Now White can avoid a Black pawn lever on the queenside. It seems that Black has no more attacking ideas at the queenside, but he still has a piece sacrifice at his disposal.} (20. Rg2 a5 { intending ...a5-a4 to open the a-file. Black still has the b4-square for his knight or bishop.}) 20... a5 {[%emt 0:04:49]} 21. Bh3 {[%emt 0:01:57] Presumably White considered that - due to Black's 18...c3 - he has plenty of time to prepare his attack. But in fact Black has an idea (the sacrifice of a piece), so White shouldn't dally around. The next few moves by White don't really help in supporting g4-g5, White's main attacking idea. Putting the bishop on h3 only blocks the h-file.} (21. Rg2 {goes straight for the preparation of g4-g5. After e.g.} Be8 22. Nh1 $5 {White is ready for g4-g5.} Nd6 {Black should play some moves aimed at defending the kingside. That way Black is not in danger losing, but on the other hand he cannot make any improvements on th queenside. The d6-knight blocks the e7-bishop, so the idea of playing ...a5-a4 and ...Nxb4 no longer works.} ({Becoming active on the queenside is now out of the question:} 22... a4 23. b4 Nxb4 $2 24. axb4 Bxb4 25. g5 {and White is already very fast in his attack. One should notice that with the queen on d1, threats like g5-g6 followed by an eventual Qh5 are in the air.}) (22... h6 {doesn't stop g4-g5, as White can still play e.g. Nf2-h3.} ) 23. Be2 (23. g5 $6 g6 {opens up the position in Black's favour.}) 23... Rg8 24. Nf2 $11 {and the position is about balanced. If White plays g4-g5, Black is ready to counter with ...g7-g6. On the other hand, with his pieces aimed at defending the kingside, Black lacks attacking ressources on the queenside.}) 21... Rc7 {[%emt 0:02:53]} 22. Rh2 {[%emt 0:02:55]} Rg8 {[%emt 0:00:00]} 23. Qe2 {[%emt 0:03:28]} Ra7 {[%emt 0:00:00] Apparently Black didn't appreciate yet the idea of sacrificing the knight against two pawns. A few moves later the rook moves to b7, so the text move looks like a loss of time.} 24. Kh1 { [%emt 0:01:33] White moves the king before Black opens the g-file with ... g7-g6, but with his next move Black makes clear that Black has other intentions.} h6 {[%emt 0:04:33] Black now has two pawns and three pieces aiming at the critical g4-g5 square. White has to do some regrouping until he can renew his threat. One should notice that moving the bishop to h3 and the king to h1 has hindered the manoeuvre Ng3-h1-f2-h3.} 25. Bf1 {[%emt 0:05:04]} Be8 {[%emt 0:01:24]} 26. Qg2 {[%emt 0:02:41]} Rb7 {[%emt 0:00:00]} 27. Be2 { [%emt 0:00:12] Now White only has to move the knight to f1 and then he is ready for g4-g5, as the h6-pawn is pinned. So Black has to come up with something.} (27. Rb1 $5 {is a precaution aginst the ensuing sacrifice on b4, but Black might play} Bd7 {intending ...Qf8, e.g.} 28. Be2 Qf8 29. Ra1 { and now again} a4 30. b4 Nxb4 31. axb4 Bxb4 {is strong, e.g.} 32. g5 Be7 33. gxh6 gxh6 34. h5 Rg7 35. Nh4 Qe8 36. Ng6+ Kg8 {and White cannot come through on the kingside while Black's pawns on the queenside are rolling.}) 27... a4 { [%emt 0:00:50]} 28. b4 {[%emt 0:00:09]} Nxb4 {[%emt 0:03:28] There it is, Black sacrifices a piece. He gets two dangerous pawns in return, but Black has to consider that White can immediately play g4-g5. It's difficult to say what caused Black to go for this sacrifice: Did the engine see the positive effects or did it just consider that doing nothing has negative consequences? Eiter way, now White has to play accurately to keep the balance.} 29. axb4 {[%emt 0: 00:20]} Bxb4 {[%emt 0:00:00]} 30. g5 {[%emt 0:00:10]} Bf8 {[%emt 0:01:24]} 31. g6 {[%emt 0:00:45] White finds an elegant way to bring his knight to e6, but also closes his attacking lines against Black's king. But alternatives are already too slow.} (31. Qg1 b4 32. Rg2 {followed Nh5 and opening the g-file would be strong if Black had not counterplay. But unfortunately for White after } b3 $19 {Black's pawns are already close to queening.}) 31... Nd6 {[%emt 0:01: 15]} 32. Ng5 {[%emt 0:07:11] Bringing the knight to e6.} b4 {[%emt 0:01:28] Taking the g5-knight would be desastrous in view of the open h-file against Black's king, but Black can simply advance his pawns.} (32... fxg5 $2 33. hxg5 $18) 33. Ne6 {[%emt 0:01:36] White intends Nh5 and then taking or h6 or f6 is in the air. But Black has two strong passed pawns...} Qb6 {[%emt 0:00:02]} 34. Bd1 {[%emt 0:01:43] White clears the second rank so that he can return his extra piece against one of Black's passed pawns. White is struggling to stay in the game.} (34. Nh5 b3 35. cxb3 axb3 $17 {is good for Black.}) 34... b3 { [%emt 0:00:36]} 35. cxb3 {[%emt 0:00:04]} axb3 {[%emt 0:00:17]} 36. Qf3 { [%emt 0:01:09]} b2 {[%emt 0:00:00]} 37. Bxb2 {[%emt 0:00:08]} cxb2 {[%emt 0:02: 48]} 38. Rb1 {[%emt 0:00:12]} Be7 {[%emt 0:02:39] If Black looks for improvements over the game he must start looking here. But even then it is not clear whether Black could have done better.} (38... Qb4 $5 {aims at e1} 39. Rhxb2 (39. Re2 Qa3 $17 {and Black's queen hinders White's attack on the kingside.}) 39... Qxb2 40. Rxb2 Rxb2 {with some material advantage to Black, but White can stay in the game with} 41. Nh5 Nb7 $5 {intending ...Nc5} (41... Bd7 42. Nd8 Be7 43. Nf7+ Nxf7 44. gxf7 Rf8 45. Nxg7 Rxf7 46. Ne6 {also gives White counterplay}) (41... Be7 42. Nhxg7 Rxg7 43. Qh5 Rh7 {and White at least has} 44. Qf3 Rg7 45. Qh5 $11 {with a repetition of moves}) 42. Ng5 $5 {and the game is still far from clear} (42. Qg4 Nc5)) (38... Bd7 $5 {is another option}) 39. Rc2 {[%emt 0:00:49] White now should not (and does not have to) take the b2-pawn.} (39. Rhxb2 Qxb2 40. Rxb2 Rxb2 41. Nh5 Bd7 {and Black has a crucial extra tempo compared to lines after 38...Qb4 (see above).} 42. Nhxg7 Rxg7 43. Qh5 Bxe6 44. Qxh6+ Kg8 45. fxe6 Rb1 $19) (39. Nh5 Bd7) 39... Bc6 {[%emt 0:00: 01]} (39... Bd7 $6 {makes White's life easy after} 40. Nc5 {followed by taking on d7 and b2.}) 40. Nh5 {[%emt 0:08:08] Taking on g7 followed by Qh5 is a threat now.} Nxf5 {[%emt 0:00:00] Black initiates a series of mutual sacrifices. It look strange, but the game is still balanced.} (40... Qa6 { allows White an attack on the kingside:} 41. Nhxg7 (41. Rcxb2 $2 {is inaccurate:} Rxb2 42. Rxb2 Qa1 $17 {(threatening ...Nxf5)} 43. Nhxg7 $2 Qxb2 44. Qh5 Qc1 $19 {and White's attack is stopped}) 41... Rxg7 42. Rbxb2 Rxb2 43. Rxb2 Qa1 44. Qh5 Kg8 (44... Qc1 45. Rc2 $18) 45. Rb8+ Ne8 46. Nc7 Kf8 47. Ne6+ Kg8 48. Nc7 $11 {with a repetition of moves}) 41. Qxf5 {[%emt 0:00:06]} Qa6 { [%emt 0:00:44]} 42. Rcxb2 {[%emt 0:00:10]} (42. Nhxg7 $2 Qxd3 $19) 42... Rxb2 { [%emt 0:01:58]} 43. Rxb2 {[%emt 0:00:00]} Qa1 {[%emt 0:00:14]} 44. Nhxg7 { [%emt 0:00:54] Now it is White's turn to shed some more material. As the great Michail Tal once said: "It doesn't matter how many of your pieces are hanging, your opponent can take only one at a time." Indeed, Black has three possible captures, but no way to an advantage.} (44. Rd2 Qc1 45. Qf2 Bb4 $17) 44... Qxb2 {[%emt 0:01:17]} (44... Qxd1+ 45. Kh2 Bd7 46. Qf2 {only favours White.}) (44... Rxg7 $2 {is even worse:} 45. Rb8+ Rg8 46. g7#) 45. Qh5 {[%emt 0:00:29]} Qd2 { [%emt 0:00:13]} 46. Nf5 {[%emt 0:00:08]} Rxg6 {[%emt 0:01:03] ... and now it's Black's turn.} (46... Be8 47. Nxh6 ({or} 47. Nxe7 Bd7 $11) 47... Qe1+ 48. Kh2 Qf2+ 49. Kh3 Qf1+ 50. Kh2 {is another way to the draw}) 47. Qxg6 {[%emt 0:00: 15]} Qxd1+ {[%emt 0:00:03]} 48. Kh2 {[%emt 0:00:12]} Qe2+ {[%emt 0:00:54]} 49. Kh3 {[%emt 0:00:11]} Qf1+ {[%emt 0:00:10] Technically this is Black's only move to prevent a loss, but of course it is sufficient.} (49... Qxd3+ $2 50. Ng3 $18 {would be inappropriately greedy.}) 50. Kg4 {[%emt 0:00:25]} Qe2+ { [%emt 0:01:31]} 51. Kh3 {[%emt 0:00:11]} Qf1+ {[%emt 0:00:30]} 52. Kg4 { [%emt 0:00:41]} Qe2+ {[%emt 0:02:10]} 53. Kh3 {[%emt 0:00:08] And the players agreed to a draw. Even though I am somewhat critical of 18...c3, there are not so many improvements for either side in the subsequent play. A game that shows that even draws can be exciting!} 1/2-1/2