I played on Monday, here is the game. My opponent was an old guy rated 1674, he had a break for a quite a few years, played just in one tournament afterwards, but unexpectedly didn’t look rusty, quite the opposite. He played very fast, even on d5 he didn’t spend much time. I didn’t feel comfortable until he played 18… bxc6.

I felt that bxc6 wasn’t the best move and after 19. Bb6 thought  that I am better. This is a typical move in this line and this bishop is really strong. He started to think more, especially after he understood my intention to occupy “d” vertical. His 28… Kh7 was an already losing move, then Kg6 was a final mistake. I saw that I am better in the bishop endgame and then I realized that b5 is winning.

All bishop endings are drawn ???  Every joke is half the truth,  it’s exactly related to this rephrasing of Tarrash’s famous  “All rook endings are drawn”.   What if we have opposite- colored bishops? Wikipedia says: 

 ” Positions when one side has an extra pawn are usually drawn and even two extra pawns (and occasionally more) may not be enough to win too”. 

I got a little example of that in one of my thematic correspondence games.  My Marshall attack with Black didn’t go very well and I found myself practically two pawns down in this position:


Crafty recommends 34. Bh6+ Kh7 35. Qh4 Re1+ 36. Kh2 Kg8 37. Rg2 Re2 38. Qg3 b4 39. cxb4  with estimate 2.12. My opponent decided to exchange rooks and queens,  so we came to the opposite colored  bishops endgame. Computer estimate dropped to 0.73, we will see later why. 34. Rxe2 Qxe2 35. Qf2 Qxf2+ 36. Kxf2


So, the plan for Black is to block advancing of the white pawns.  36. … Bc2 37. d5 Kf8 38. Ke3 Bb3 39. Kd4 Ke7


40. Kc5 Kd7 41. Bh6 Bc4 42. f4 Kc7 43. Bg7 f5 44. Be5+ Kd7 45. d6 


45. … Bd3 46. b3 Be2 47. c4 bxc4 48. bxc4 Bf3


49. Bf6 Bg2 50. Be7 Bf3 51. Kb6 Be4 52. c5 Bd5


53. Bg5 Be4 54. Bf6 Bd5 55. Be5 Be4 56. Bf6   1/2-1/2