The best move

does not exist. Usually, there is more than one move that leaves the theoretical result of a position unchanged (whereas mistakes alter it). The difference between theoretically correct moves lies in the difficulty of the problems they pose the opponent. Which depends on who, or what, the opponent is.

Imagine a computer that had solved chess. It would always draw against itself, unless the theoretical result of the game is a win for White or Black, which seems unlikely. But put that computer in a round-robin against other, less capable machines. It might not win the tournament. And in order to win, it might have to take some risks; play some theoretically losing moves. Theoretically sound chess might be too quiet to win enough games.

At the human level, there is a kind of ethic of correctness: some risks one is ashamed to take. One feels guilty playing an objectively bad move on the assumption that the opponent will not see the refutation. But the level of risk one takes probably varies according to the opponent: both what one knows beforehand, and how one senses their play at the board. I doubt if absolute neutrality is psychologically possible; you are, much of the time, riding a wave.

In this game from the Central London League my 23rd and 24th moves constitute a pawn sacrifice which has a positional rationale but incurs some risk. I have no idea whether it loses against best play; and I don’t know whether I would have been more cautious against a different player. It gave my opponent something to think about, anyway.

You need to activate javascript to enhance chess game and diagram visualization.
1. d4 {I am playing Black, with 75 minutes for the game plus 15 second increment. I didn&#8217;t know my opponent&#8217;s grading but knew the opposing team&#8217;s lower boards were likely to be ranked rather below ours.} 1... Nf6 2. c4 c5 3. e3 {I had nothing prepared against this move and wanted to keep the game strategically complex.} 3... e6 4. Nf3 b6 5. Nc3 cxd4 6. exd4 d6 {Black&#8217;s setup is like a Hedgehog, but against the c4-d4 centre rather than c4 and e4.} 7. Be2 (7. Bd3 {seems more natural and has better results though in a small sample of games.}) 7... Be7 8. O-O O-O 9. Qc2 Bb7 10. Be3 Nbd7 11. b3 Rc8 12. b4 Qc7 13. Nb5 Qb8 14. a3 Rfe8 (14... Ba6 15. Qa4) 15. Nd2 {My opponent stakes out space with his pawns but plays a little cautiously with his pieces; a slight preference for statics over dynamics ?} Qa8 16. f3 Nd5 17. Qb3 ({or} 17. Bf2 Nf4) 17... Nxe3 {Hoping the two bishops and White&#8217;s dark square weakness will tell.} ({I considered} 17... a6 18. Nxd6 ({not} 18. Nc3 Nxe3) (18. cxd5 Bxd5 19. Qb2 axb5 {is possible.}) 18... Bxd6 (18... Nxe3 {fails to} 19. Nxc8) 19. cxd5 Bxd5) 18. Qxe3 h6 19. Rac1 e5 {Feeling I had to start doing something, and hoping my majority would be more mobile than his.} 20. d5 a6 21. Nc3 f5 22. Bd3 Bg5 23. Qe2 { <span class="PgnWidget-anchor-diagram">[]</span> Now I transform the position} e4 24. fxe4 f4 { <span class="PgnWidget-anchor-diagram">[]</span> Black gets a dark-squared blockade. I liked this Petrosian-inspired idea, but it&#8217;s a big commitment and I had fallen quarter an hour behind on the clock over the last few moves. Moreover, White has a strong pawn centre and Black&#8217;s queenside pieces are not so easy to mobilise. The computer doesn&#8217;t like the pawn sacrifice but I am not sure how grave a risk I was running.} 25. Nf3 {I think I&#8217;d missed this but fortunately it doesn&#8217;t matter too much.} 25... Bf6 26. Qf2 b5 27. Ne2 bxc4 28. Rxc4 Rxc4 29. Bxc4 g5 { <span class="PgnWidget-anchor-diagram">[]</span> } ({Reluctantly deterred from} 29... Rxe4 {by} 30. Nd2) 30. h3 {?} (30. Nd2 {is better and the computer still prefers White.}) 30... Rxe4 {The disappearance of the pawn revives Black&#8217;s queen&#8217;s bishop.} 31. Nd2 Re3 {Possibly too loose.} (31... Re5 {may be better}) 32. Kh1 {? After playing fast and keeping ahead on the clock my opponent now started thinking a lot and had overtaken me by move 37. But this move is again too passive, and from now on Black&#8217;s activity overwhelms White. I miss some better options but keep a winning position.} ({After} 32. Nxf4 {!} gxf4 33. Qxf4 {I think I was planning} 33... Qe8 {but the machine thinks White is doing OK. I don&#8217;t know if my opponent considered the knight sacrifice.}) 32... Bxd5 33. Rc1 Ne5 {In such a sharp position, the trick is to combine a bit of analysis with recollection of the basic positional imperatives. Here it&#8217;s important to activate all the pieces.} 34. Rc2 Nd3 35. Bxd3 Rxh3+ (35... Rxd3 36. Nxf4 {is not good for White but short of time it seemed worth preventing.}) 36. Kg1 Rxd3 37. Qf1 ({Now if} 37. Nxf4 {then} 37... Bd4) 37... Qe8 38. Nc4 Bxc4 (38... f3 {is crushing according to the machine. But I was happy to simplify a bit while keeping control.}) 39. Rxc4 Qe3+ (39... Rd2) 40. Kh1 Rd2 41. Qb1 {At last an active move, but I had it covered.} (41. Ng1) 41... Qd3 (41... Qxe2 {??} 42. Qg6+ {would be sad}) 42. Rc8+ Kg7 43. Qxd3 Rxd3 { <span class="PgnWidget-anchor-diagram">[]</span> The ending is clearly winning for Black. I had four minutes on the clock and was playing mainly on the fifteen-second increment, but the position is not too difficult.} 44. Rc6 Be5 45. Rc7+ (45. Rxa6 {loses to} 45... Rd1+ 46. Ng1 Bd4 47. Rxd6 Rxg1+ 48. Kh2 Be3) 45... Kg6 46. Rc1 Rxa3 47. Kg1 Ra1 48. Rxa1 Bxa1 49. Kf2 Kf5 50. g3 Ke4 {Still trying to restrict his pieces.} 51. gxf4 gxf4 52. Nc1 Bd4+ 53. Ke2 f3+ 54. Kf1 Bc3 {Here White resigned. After 54 Na2 I was planning Bd2 when both knight and king are tied down and simply advancing the h pawn will win.} *
() -
Commented by