Speelman's Agony #54

[Event "?"] [Site "?"] [Date "2017.04.27"] [Round "?"] [White "Mahbub Shahalam"] [Black "Jim Guill"] [Result "1-0"] [ECO "D45"] [WhiteElo "2219"] [BlackElo "1957"] [Annotator "Guill,Jim/Speelman,Jonathan"] [PlyCount "201"] [EventDate "2017.??.??"] [SourceDate "2015.07.13"] 1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. e3 Nbd7 6. Qc2 Bd6 {This was my first attempt at a Semi-Slav.} 7. b3 O-O {[#]} 8. Be2 ({JS: The move order is quite complicated around here. In principle, White would like to play Bb2 and probably Bd3 rather than Be2 followed by 0-0, but there are lots of moments at which Black can muddy the waters with ...e5, and these can be very different depending on the exact position.} 8. Bb2 e5 9. cxd5 cxd5 10. dxe5 (10. Nb5 Bb4+ 11. Bc3 Bxc3+ 12. Nxc3 e4) 10... Nxe5) (8. Bd3 {would be more desirable if possible but here e5 is that much better than with the bishop on e2.} e5) 8... b6 ({JS:} 8... e5 9. cxd5 cxd5 10. Nb5 Bb4+ 11. Bd2 Bxd2+ 12. Nxd2 a6 13. dxe5 Nxe5 14. Nd4 Bg4 {is a fairly typical line in which Black has an isolated pawn but is very active.}) 9. O-O Bb7 10. Bb2 Rc8 {Here White generally sticks rooks in the centre with Rad1 most common.} 11. e4 Nxe4 12. Nxe4 dxe4 13. Qxe4 Nf6 14. Qh4 h6 (14... c5 15. Rad1 Ne4 {is possible}) 15. Rfd1 (15. Rad1 { would prepare the retreat Ba1 if Black played Ba3.}) 15... Qe7 16. Ne5 Ba3 ( 16... c5 {would activate the b7 bishop and was much more natural.}) 17. Bc3 Bb4 18. Rac1 Bxc3 19. Rxc3 Nd5 20. Qxe7 Nxe7 {Okay, exchanged off the worrisome bishop and queen, but now he gets an annoying pin.} 21. Bf3 Rfd8 22. Rcd3 f6 23. Ng4 Nf5 ({[#] JS: Since White can now offer an annoying exhange it was probably better to play some other improving move such as} 23... Rc7) 24. Ne3 Nxe3 {I was reluctant to take on e3 because I thought that "promoting" the f-pawn strengthened his position. On the other hand, retreating seemed passive. Decisions, decisions. I settled on the exchange, after which he set his center in motion. JS: Nxe3 does look wrong. White now gets a pleasant edge.} 25. fxe3 Rc7 26. e4 {JS: On e3 the pawn protects what was previously a small weakness. If you can get in e4-5 then it makes some sense to advance it, though you need to be able to deal with ...f5 and ...g5. But first Kf2 was normal.} ({Or very possibly} 26. c5) 26... g6 27. Kf2 (27. e5 f5 28. g4 {JS: was also interesting and hard to assess.}) 27... Kf7 28. Ke3 Ke7 29. h4 e5 30. d5 ({JS:} 30. dxe5 Rxd3+ 31. Rxd3 fxe5 32. Bg4 {is presumably dead equal.}) 30... cxd5 31. exd5 Kd6 32. g4 a5 {Missing an opportunity to gain space and perhaps connected passers with f5.} 33. Be4 Bc8 34. Rf1 Rf8 35. Bxg6 Bxg4 36. Rd2 ({JS:} 36. Bf5 {looks better to prevent f5, though Black should be fine anyway.} Rg7 37. Rd2 Bxf5 38. Rxf5 Rg3+ 39. Ke4 Rg4+) 36... f5 {Guess I get my passers anyway.} 37. h5 e4 {Did I just do that? Really? Why on earth did I gift my opponent all those great squares in the center? I blame my hand – it was my hand's fault! JS: Jim is quite right of course: contol of squares with the opposite colour to your bishop is crucial in bishop endings.} 38. Rd4 (38. Kd4 $1 {was correct. The king is wonderful here}) 38... Bf3 39. Kf4 Bg4 40. Ke3 Bf3 41. Kf4 Bg4 42. Re1 Re7 43. Re3 {The central squares that "my hand" gave away on move 37 are now occupied. From here I'll fast forward over the rest of a messy, mistake-riddled middlegame to the endgame, where the Agony lies.} Bf3 44. a3 b5 {JS: Black should wait, since when he weakens his black square control he encourages c5 ideas.} 45. Rd2 bxc4 46. bxc4 Rb7 $2 {[#] Allowing a breakthough. } 47. c5+ $1 Kxc5 48. d6 Rd7 49. Rc3+ Kb5 50. Rb2+ ({JS:} 50. Bxf5 $1 {puts White in full control.}) 50... Ka4 51. Rb6 Rfd8 52. Rc4+ Kxa3 53. Rc1 Ka2 {[#]} 54. Bxf5 ({JS:} 54. Rc7 $1 {was winning because if} Rxd6 55. Bf7+ Ka3 (55... Ka1 56. Rc1#) 56. Rc3+ Ka4 57. Bb3+ Ka3 58. Bc2+ Ka2 59. Bb1+ Ka1 60. Ra3#) 54... Rxd6 55. Be6+ Rxe6 56. Rxe6 Rb8 57. Rc5 a4 58. Ra5 a3 59. Rxh6 Kb2 60. Rha6 Be2 61. Rxa3 Bxh5 {[#] JS: As you'd imagine, this should be drawn, and I verified this with the Lomonosov tablebases at Moscow State University. Tablebases are wonderful at pointing out winning lines or difficult defences, but in positions like this they need human guidance and the first choice here - which of course doesn't actually affect the evaluation - is the very splendid Rc6.} 62. Kxe4 Bf7 63. Rg3 Bb3 64. Ra7 Bc2+ 65. Kd4 Rd8+ 66. Kc5 Rc8+ 67. Kb5 Rb8+ 68. Ka5 Rb3 69. Rxb3+ Bxb3 {Whew. What a relief! Weathered the storm. At several points I thought I was just dead lost. But now my king is in the right corner, so all I have to do is avoid a couple of traps and tough it out. I've studied B v R before, so no problem, right? What could possibly go wrong? Well, I was down to one minute on the clock with a 30 second increment.} 70. Kb4 Bc2 71. Rd7 Ka1 72. Rd2 Bg6 73. Kc3 Bf5 74. Rf2 Bb1 75. Rf1 Ka2 76. Rf2+ Ka1 77. Kb3 Be4 78. Re2 Bf5 79. Ra2+ Kb1 80. Re2 Ka1 81. Ra2+ Kb1 82. Ra5 Be6+ 83. Kc3 Ba2 84. Ra7 Ka1 85. Rg7 Bb1 86. Kb3 Ba2+ 87. Kc2 Bb1+ 88. Kc1 Ba2 89. Rg2 Be6 90. Rg7 Ba2 91. Rg2 Be6 92. Kc2 Bf5+ 93. Kb3 Be6+ 94. Ka3 Bf5 95. Ra2+ Kb1 96. Rg2 Ka1 97. Ra2+ Kb1 98. Rf2 {[#]} Bd3 {After I punched the clock I realized with horror what I'd just done. I stared in disbelief at my hand, bloody traitor that it was. Et tu, Brute?} 99. Kb3 Ka1 $6 {JS: This loses fairly simply.} (99... Kc1 100. Kc3 Bb5 101. Rf5 Be2 102. Rg5) (99... Bb5 { JS: would make White work harder} 100. Rb2+ Kc1 101. Kc3 Bc6 102. Rb6 Bd5 103. Rb5 Bc6 104. Rc5 Bd7 105. Rc7 {and the bishop has run out of places to hide.}) (99... Ba6 100. Rd2 $1 Kc1 101. Rd4 Bb5 102. Kc3 {and eventually the bishop will run out of squares as in the previous line.}) 100. Ra2+ Kb1 101. Rd2 1-0

[Event "DCCL"] [Site "?"] [Date "2017.03.31"] [Round "?"] [White "Andrew Samuelson"] [Black "Jim Guill"] [Result "0-1"] [ECO "C24"] [WhiteElo "2377"] [BlackElo "1957"] [Annotator "Guill,Jim/Speelman,Jonathan"] [PlyCount "104"] [EventDate "2017.??.??"] [SourceDate "2015.07.13"] 1. e4 e5 2. Bc4 Nf6 3. d3 c6 4. Nf3 d5 5. Bb3 a5 ({JS: If} 5... Bd6 6. Nc3) ( 5... Bb4+ 6. c3 Bd6 {is therefore now often played, claiming that taking the square away from the knight is more important than the tempo loss.}) 6. Nc3 ({ JS:} 6. a3) ({or} 6. a4 {are more common.}) 6... Bb4 7. a3 Bxc3+ 8. bxc3 a4 9. Ba2 Qc7 10. O-O O-O 11. Qe2 Nbd7 12. exd5 cxd5 13. c4 $6 {[#]} d4 {I went for a policy of grabbing space and holding on for dear life. JS: Since this displeases the a2 bishop, it makes a lot of sense.} 14. Bd2 b6 15. Bb4 Re8 16. Ng5 Nc5 17. f3 Bb7 18. Ne4 Bxe4 {The knight has to go. In addition, I've now saddled him with the two bishops The rest of the middlegame is about me trying to keep things bottled up and him trying to bust out.} ({JS:} 18... Nh5 { was a good altternative.}) 19. fxe4 Nfd7 {With the idea of trading a pair of rooks to neutralize his sure-to-come heavy piece buildup on the f-file and kingside.} 20. Rf5 Re6 21. Raf1 Rf6 {[#]} 22. c3 (22. g4 {was quite a good idea to recapture on f5 with the g-pawn. Since c3 doesn't free the bishop and creates weaknesses it was a bad idea.}) 22... Rxf5 {Just like George W Bush in Iraq, Mission Accomplished! Oh, wait. You mean the war, er, game goes on?} 23. Rxf5 Qd6 24. Bb1 f6 {Dark square blockade coming.} 25. Rf1 dxc3 26. Bxc3 Ne6 27. Bb4 Ndc5 28. g3 Qc6 29. Qe3 Nd4 30. Rf2 Rd8 {[#] Made the time control with a few minutes to spare.} 31. Rb2 Qc7 {Marking time and keeping the entry squares covered in the event of a b-file rook infiltration.} 32. Bc3 Ncb3 { Ok, I'm solid in the center and I'm past the time control, but how do I make forward progress from here? What's my plan? A stable center implies wing play, yes? But I felt uncomfortable advancing on or committing to either wing. My eye was on the a3-pawn, but I was afraid that I might inadvertently "release the Kraken" (the bishop pair) in attempting to get at it. So instead I thought I'd wait a bit. Maybe a better idea would come. First do no harm.} 33. Kg2 Qc6 {[#] My queen does the two-finger "I'm watching you" gesture to his king. The idea is possible stuff like f5 or Nf5 should an opportunity present itself.} 34. Bc2 $2 Nxc2 $1 35. Rxc2 Qd6 {My opponent is a talented tactician, but something went wrong; although, at the time, I wondered what kind of trap I was walking into. JS: The bishop was so awful that it must have been fairly easy not to notice that Black could surrender one of his magnificient steeds for it in order to win a big pawn.} 36. Be1 Qxd3 {Okay, my planning woes are over and I can now continue in straightforward fashion.} 37. Qxd3 Rxd3 38. Bf2 Rd2 {If he exchanges on d2 then e4 soon falls and Black has a winning endgame.} 39. Rc3 Nc5 40. Re3 Nd3 41. Kf3 Rxf2+ 42. Kg4 Rd2 {[#] Are you ready for it? Here it comes: "And the rest is just a matter of technique." There. Done. Always wanted to say that – now I can cross it off my bucket list. You'll notice that the computer has no respect for my technique and rudely points out that I miss faster ways to end the game.} 43. h4 Nc5 44. h5 Rd4 45. Rc3 Nxe4 46. Rc2 Kf7 47. Kh3 Rd3 48. Rb2 Rxg3+ 49. Kh4 g5+ 50. hxg6+ hxg6 51. c5 g5+ 52. Kh5 f5 {JS: A very nice positional win by Jim in which he made White's bishops, and in particular the casualty on a2 look ridiculous. By not rushing he gave his opponent the chance to play in an inferior position and this greatly simplified matters when White duly blundered.} 0-1

[Event "DCCL"] [Site "?"] [Date "2016.12.09"] [Round "?"] [White "Chris Sherwin"] [Black "Jim Guill"] [Result "0-1"] [ECO "C45"] [WhiteElo "2041"] [BlackElo "1905"] [Annotator "Guill,Jim/Speelman,Jonathan"] [PlyCount "102"] [EventDate "2016.??.??"] [SourceDate "2015.07.13"] 1. e4 e5 2. Nf3 Nc6 3. d4 exd4 4. Nxd4 Nf6 5. Nxc6 bxc6 6. Bd3 d5 7. Qe2 dxe4 8. Bxe4 {JS: I really don't like this much, though it has been played a number of times. Admittedly, Black ends up with a damaged queenside, but the two bishops are a serious force in a fairly open position.} Nxe4 9. Qxe4+ Qe7 10. Nc3 Bf5 11. Qxe7+ Bxe7 12. Bf4 O-O-O 13. Rc1 Bf6 {The c1 rook is tied down, at least temporarily.} 14. O-O Rhe8 15. Be3 {[#]} Bxc3 {I thought the exchange would accentuate the differences in piece activity. I didn't mind the pawn structure symmetry or giving up the two bishops, because the black rooks are ready to rock while the white rooks are not, and the black king is a step closer to the center. In symmetrical pawn structures better piece activity is an advantage, yes?} (15... a5) 16. bxc3 a6 {The computer (IPad Pro + Stockfish) recommends Kb7, but I wanted to keep the king closer to the action.} 17. Rfe1 Rd5 18. Bd4 Rxe1+ 19. Rxe1 f6 {I thought the threats to his bishop and king (weak back rank) meant that he couldn't afford to check on e8 with his rook. But I failed to see far enough. The computer points out that it's a playable option because White is able to stir up enough complications to maintain the balance.} 20. f3 $6 {JS: Now White ends up completely passive but} (20. Re8+ Kd7 (20... Kb7 21. f3 Ra5 22. Re7 c5 23. Bf2 g5 24. h4) 21. Rg8 c5 22. Rxg7+ Kc6 23. g4 cxd4 24. gxf5 dxc3 {gives Black a clearly better rook ending.}) 20... Kd7 {Sometime around now I was remembering the game Meulders-Karpov from Marin's "Learn from the Legends". I wanted to get my king to c4 to belly up to the c3-pawn, where the self-interference of bishop and c3-pawn might give me some chances. One of my teammates later pointed out that in the upcoming endgame structure White would probably prefer to have his pawn on f2 so his bishop could defend it.} 21. Kf2 Bxc2 22. Re2 Bd3 23. Rd2 Bc4 24. Rb2 ({JS:} 24. a3 {was less bad}) 24... Ra5 25. Rd2 {[#]} Rxa2 26. Rxa2 (26. Rxa2 { JS: This is a very big decision. Of course if the pure opposite coloured bishop endgame is winning then it's completely correct. But it's very hard to be sure.}) 26... Bxa2 {It seems natural for White to aim for the bopp's endgame and seek drawing chances there, although Karpov's opponent avoided (for good reasons) rook trades that would lead to pure bopp's. But now I have an a-pawn runner and the self-interference issue becomes a real problem for White. It's difficult to use the bishop to hold up the a-pawn or cover the queening square, so the king takes on that role.} 27. Bc5 ({JS: I wondered whether White could defend with} 27. c4 Bxc4 28. Ke3 Ke6 29. g3 Kd5 30. h4 { If Black had f5, g6 then with a long diagonal to use I think White should be able to maintian a blockade. But here there may be a problem because the diagonal is too short. For instance if Black could get wKd2,Bc3,Pf4,g3,h4 / bKb3,Bf5,Pa3,c5,c7,f6,g7, h5 then White to move is in zugzwang since if Ba1, Ka2-b1}) 27... Ke6 28. Ke3 Kd5 29. Bf8 g6 30. Kd3 Bb1+ 31. Kd2 Kc4 {I could barely contain my excitement when my king settled onto c4.} 32. Kc1 Bd3 33. Kb2 Be2 {Why not Bf1? Because I wanted to tie down his king or build a cage for his king in the white queenside quadrant of the board and get a K v B mismatch on the other side.} 34. Bg7 f5 35. Be5 a5 36. Bxc7 a4 37. Be5 Bd1 38. h4 Bb3 { Mistake. Pointless move. The bishop should have kept the heat on the kingside pawns. White could now have gotten in g4 and exchanged some pawns.} 39. Bg7 ( 39. g4 Kd3 40. gxf5 gxf5 41. f4 Ke4 {JS: This would presumably be drawn with the bishop on g5, since White waits and if Black plays ...Kh5 to prepare h6 then Bf6 h6 Bg6 defends (I don't think Black can combine queenside and kingside play). But with Be5 h5 followed by Kf3-g4 is a threat.} 42. h5 { is therefore forced, but after} h6 {Black will still get a passed pawn on the kingside.}) 39... Kd3 {It starts to become apparent why the f-pawn would "prefer" to be on f2. How would Black make progress if the structure of the white kingside happened to be, as my teammate suggested, pawns on f2, g3, h4 and a bishop on e3 holding it together? In Marin's book that would correspond to Averbakh Scenario 1, where the defending king is in front of the passer and the bishop defends the remaining pawns. But now, due to f3, the bishop is unable to hold the kingside.} 40. f4 Ke3 41. Bh6 Kf2 42. Ka3 Kxg2 43. Kb2 Kg3 44. Bg5 Kg4 45. Ka3 Kh5 46. Kb2 (46. Bf6 h6 47. Bg7 g5 $1) 46... h6 47. Bf6 Kg4 48. Bg7 Kxf4 49. Bxh6+ Ke4 {I moved toward the center because, thematically, I wanted to keep his king boxed in.} 50. Bf8 f4 51. Kc1 Kd3 {Sorry. No exit! JS: Another excellent positional game by Jim.} 0-1