How to Refute the Ruy Lopez by Paul Talsma

[Event "4NCL Online"] [Site "lichess.org"] [Date "2020.04.28"] [Round "4.4"] [White "Li, Henry"] [Black "Talsma, Paul"] [Result "0-1"] [ECO "C80"] [WhiteElo "2315"] [BlackElo "2177"] [Annotator "Martin Walker / Paul Talsma (2020)"] [PlyCount "103"] [EventDate "2020.??.??"] [WhiteTeam "Gonzaga"] [BlackTeam "Anglia Avengers 1"] [TimeControl "2700+15"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Nxe4 6. d4 exd4 $5 {Paul: The opening is called the Riga variation of the Open Ruy. It is more than 100 years old. I published an article about the Riga in New in Chess yearbook 85. It's not well-regarded, but as this game demonstrates much better than its reputation.} 7. Re1 d5 8. Nxd4 Bd6 9. Nxc6 Bxh2+ 10. Kh1 {Paul: Most players go for the endgame I had; but in my opinion it does not offer much for white.} ({Andy: Here, White should have smelt a rat! More prudent was:} 10. Kxh2 { accepting a draw by perpetual check} Qh4+ 11. Kg1 Qxf2+ 12. Kh2 Qh4+ $11) 10... Qh4 11. Rxe4+ dxe4 12. Qd8+ Qxd8 13. Nxd8+ Kxd8 14. Kxh2 Be6 15. Nc3 $6 { Paul: This move is a common mistake: it loses a piece. I've had this in 5 games already. Even strong players fall for it. But it is not as bad as it seems: White gets decent compensation and black has to be very careful about tactics and white's passed pawns.} ({Martin: Now white's KB runs into trouble. Better is either} 15. c3 {or}) (15. Bf4 {followed by c3.}) 15... c5 $1 16. Bg5+ Kc8 17. Nxe4 ({Martin: The Telegraph (9th May 2020) pointed out the best defence is the unlikely:} 17. a3 $1 c4 (17... b5 18. Bb3 c4 19. Ba2 f5 20. a4 b4 21. Ne2 $14 {is good for White: Ba2 cannot be boxed in permanently.}) 18. Nxe4 b5 19. Nd6+ Kc7 20. Nxb5+ axb5 21. Bxb5 $15 {White is an exchange down for a P - as is the game - but retains the 2-Bs, and has a fighting chance of holding the game.}) 17... b5 18. Nxc5 bxa4 19. Nxa4 Kc7 20. Rd1 Kc6 $1 { Martin: Paul clearly knew, and the computer confirms this, that the king finds a safe haven here. Well played Paul!} 21. Rd3 Rac8 {Andy: This is Paul's first "real" move of the game. He played the previous moves at blitz speed, but now (correctly) slowed down to calculate how best to make the most of his promising position.} 22. b3 ({Andy: The point is that:} 22. Rc3+ $2 Kb5 $17 { only helps Black.}) 22... Bf5 23. Rg3 Rhe8 ({Martin: This is good, but why not play} 23... Bxc2 {, winning a pawn for nothing?}) 24. c4 Re4 ({Martin: Black loses his way a little here, though not enough to lose his advantage. The more active} 24... Re2 {is better, although clearly then black must be careful. Given the time constraint, the text is a better percentage move.}) 25. Nc3 Rg4 {Andy: Paul is playing for exchanges. His approach to handling the endgame is that the more pieces get traded off the more dangerous things become for white. } ({Martin: However, now black nearly loses his advantage. He has to backtrack with} 25... Re6) 26. Nd5 Rxg3 27. Ne7+ ({Martin: Now white loses his way! After } 27. Kxg3 Kd6 28. Bf4+ Ke6 29. Nc7+ Kd7 30. Nxa6 $11 {it will be very difficult for black to win.}) 27... Kd7 28. Kxg3 Rc5 29. b4 Rxc4 {Andy: Paul is targetting the R+P v B+N endgame - in which only Black has winning chances: a sound practical choice.} ({Martin: Even better is:} 29... Re5 30. Kf4 Ke6 { when black is just winning.}) 30. Nxf5 Rxb4 31. Nxg7 Ra4 32. Be3 Rxa2 {Martin: The passed a-pawn still guarantees black the advantage - but probably not enough to win.} 33. Kf4 a5 34. Ke5 ({Martin: This surely is the crucial moment of the game. White can probably draw, but needs to play very accurately to do so. He can play} 34. Nf5 a4 35. Nd4 Rb2 (35... h5 36. g3 ({White really doesn't have time for} 36. Kg5 {, when after} Rb2 {, it will be virtually impossible for him to stop the a-pawn. White needs his king to be involved on the queenside. The question is, will he be able to get back to the kingside in time?}) 36... a3 37. Ke5 f6+ 38. Ke4 f5+ 39. Nxf5 Re2 40. Nd4 a2 41. Nb3 Re1 { is probably drawn.}) 36. Ke4 a3 37. Bc1 Rxf2 38. Bxa3 Rxg2 {is tricky but may well be winnable.}) ({Also possible is} 34. Ke4 a4 35. Nf5 (35. Bd4 Rd2 36. Ke3 Rd1 {! wins for black.}) 35... a3 36. Nd4 f5+ {transposes to the variation above, beginning with 35.....h5+, with a probable draw.}) 34... a4 35. Nf5 a3 36. Bd4 Rd2 37. g4 a2 38. Bc3 Rxf2 39. Kf6 {Andy: At this point while watching the game, I thought that White must have decent chances of a draw by eliminating the remaining king-side pawns. Yet Paul wraps up the endgame with no trouble at all!} Rf4 40. g5 Rc4 41. Bb2 Rc6+ $1 42. Kxf7 ({Andy: White might have at least tried:} 42. Ke5 {when Black will have to exercise some technique to win: eg.} Rc2 43. Bd4 Re2+ 44. Kf4 (44. Kf6 Re6+ $19 {etc.}) 44... Ke6 $19) 42... Rg6 43. Ng3 (43. Bf6 $2 Rxf6+ $19) 43... Rxg5 {Paul: With the a-pawn snatched off I was already seriously better. And with passed rook pawns on both sides, white was done for.} 44. Ne4 Rg2 45. Bc3 Kc6 46. Ke6 h5 47. Ke5 h4 48. Kf4 h3 49. Kf3 Rc2 50. Bg7 $2 {Andy: The final inaccuracy. The B is loose on this square!} (50. Be5 Kd5 51. Bf6 h2 52. Ng3 Kc4 53. Ba1 (53. Z0 $140 Rc3+) 53... Rc1 {wins both pieces!}) 50... h2 51. Ng3 Rc5 52. Z0 {White resigned.} (52. Kg2 Rg5 53. Bf6 Rxg3+ $19) 0-1