9

I reached this position as white that led to a draw:

[FEN ""]
[White "Bad_Bishop"]
[Black ""]
[StartPly "61"]
[Result "1/2-1/2"]

1. d4 d5 2. c4 dxc4 3. Nf3 Nf6 4. e3 Bg4 5. Bxc4 e6 6. h3 Bh5 7. Nc3 Nbd7 8. O-O Bd6 9. e4 e5 10. g4 Bg6 11. dxe5 Nxe5 12. Nxe5 Bxe5 13. f4 Qd4+ 14. Qxd4 Bxd4+ 15. Kh2 Bxc3 16. bxc3 Bxe4 17. g5 Bd5 18. Re1+ Kf8 19. Bb5 Ne4 20. c4 Nd6 21. cxd5 Nxb5 22. Rb1 a6 23. a4 Nc3 24. Rxb7 Nxd5 25. Re5 c6 26. Bb2 Rd8 27. Rf5 f6 28. gxf6 Nxf6 29. Bxf6 gxf6 30. Rxf6+ Ke8 31. Re6+ Kf8 32. Rxc6 Rg8 33. Rc2 Rg7 34. Rb6 Ra7 35. Rcc6 a5 36. Ra6 Rxa6 37. Rxa6 Rd5 38. Ra7 h5 39. Kg3 Rc5 40. Ra6 Kf7 41. Kh4 Ke7 42. Rh6 Rc4 43. Kg5 Rxa4 44. Rxh5 Ra1 45. Rh7+ Kd6 46. Ra7 a4 47. f5 a3 48. f6 a2 49. f7 Rg1+ 50. Kh6 Rf1 51. Rxa2 Ke6  1/2-1/2

I was up a pawn and had more active rooks. How could white have won this endgame (if there is a win at all)?

Improve this question
1
  • 31. Kg3 to bring the king into the action must also be analysed.
    – Dag Oskar Madsen
    May 19, 2017 at 21:00

5 Answers 5

Reset to default
7

48.Kf6 should be winning. After say 48...a2, white goes h4-h5-h6 and something involving ...Rh1, Rxa2 Rxh6+, Kg5 followed by Re2 will lead to the Lucena position.

Much earlier on, I believe 31.Kg3 is winning, but it's also the best try by elimination: without an active king (which is the most important non-trivial factor in rook endgames), white won't be able to win with just his f- and h- pawns vs. black's h-pawn if he loses his a-pawn, and after Rg8 as happened in the game, white's own king is exposed to a perpetual. Black is doomed to passivity because his rooks are disconnected, and white can slowly capture the rest of black's pawns and/or get his king to e6 and promote the f-pawn.

P/S: 17.Re1 Kf8 18.Ba3+ Kg8 19.Rad1 followed by 20.Re7 looks crushing.

Improve this answer
2
  • 17. Re1 0-0-0 and I see nothing for White.
    – user58697
    May 21, 2017 at 0:02
  • I actually forgot black can castle, but I think after 18.g5 Bd5 19.gxf6 Bxc4 20.fxg7, white may have something.
    – Ken Wei
    May 21, 2017 at 6:09
4

I feel like you should have played 31. Rff7 or 32. Ree7 on the next move, doubling rooks on the seventh rank (even though I don't how much better it is :-) ). Black wouldn't have been able to create any play with his rook. Instead you (naturally) took the pawn freeing both black rooks.

Altogether I liked your game. GG!

Improve this answer
2
  • 1
    "I think the position wasn't a draw until the move 48-49." After 44. Rxh5 you have only 7 men left and can therefore check with a tablebase which will tell you that it is a draw.
    – user1583209
    May 19, 2017 at 20:50
  • 1
    Upps... You're right. I wasn't aware of the fact that one tempo matters so much in that position. In fact if it was white's move again after 44. Rxh5, white would be winning. But "if..." :D (I should probably edit my post).
    – kmartin
    May 19, 2017 at 21:00
3

The original endgame might very well be a draw due to the open position and the large number of weak pawns.

The position after 45 .. Kd6? is won for white...until 48. f6?. You needed this square for the white king in order to escape from rook checks. Could have won with 48. h4.

Improve this answer
0

Starting with 30 Rxf6 + Black to move Houdini 6.02 d = 39/81 takes to a winning seven pieces endgame for White (+ 2.53) . The endgame is even rook and three pawns vs rook. Mate in 30.

Improve this answer
0
0

Stockfish 14.1 NNUE rules this endgame to be a clear win for White. It returns +6.30 after only 3 minutes of analysis. The key move is , after 30…Ke8 31 Ree7!! with a fatal bind for Black.

Improve this answer
3
  • I meant …30 Ke8 31 Re6+ Kf8 32 Ree7!! with a fatal bind for Black.
    – user33339
    Dec 11, 2022 at 13:06
  • 2
    You can edit your answer by the "Edit" button right below your answer.
    – Minot
    Dec 11, 2022 at 14:02
  • 1
    Answer would benefit from an edit to correct the analysis.
    – SecretAgentMan
    Dec 11, 2022 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy