Ian George v Simon Bartlett
County Shield 2013/14
A quiet opening and middle-game lead to an instructive tactical ending.
Or 6...Qb6 7.Nc3 Nxd4 8.Nd5 Nxf3+ (8...Qc5 9.Nxd4 Bxd4 10.Be3 Bxe3 11.fxe3 Kf8 (A 11...Qd6 12.0-0 and white's superior development gives him the advantage. B 11...Kd8 12.Qd4 Qxd4 13.exd4+/-) 12.Qd4 Qxd4 13.exd4+/-) 9.Qxf3 Qd6 (9...Qd8 10.Bf4 Kf8 11.Bc7 Qe8 12.Ba5+/-) 10.Bf4 Be5 (A 10...e5 11.Bd2+- B 10...Qc5 11.Rc1+-) 11.Rc1 (11.Bxe5 Qxe5 and white does not have enough compensation for his pawn.) 11...Bxf4 12.Qxf4 Qxf4 13.Nxf4+/-
The purpose of this move is unclear. 11...Ne8 is better
19...Qxa5 20.Rxa5 Bb7 21.Rxa8 Rxa8 22.Ne5 (22.Bg5 Bxf3 23.Bxf6 exf6 24.gxf3+/-) 22...Ra2 (22...Bxe5 23.dxe5+- Despite the opposite-colour bishops the pair of rooks gives white good prospects of extracting full value from his extra pawn.) 23.Nd3+/-)
The opposite coloured bishops assist white's plan of imprisoning the black king with Bh6 (and e6 in response to ...f6) Taking advantage of the black rook's inability to leave the back rank, he will bring up his king to support the advance of the b-pawn. In executing this plan he must be careful not to lose the e-pawn and he must remember that, if the bishop is on the a8-h1 diagonal, Ra8 is not mate.
To prevent 29...Bc4
This does not put the win at risk and is not an error. However, although it shouldn't have made a difference, in some of the later play I wished I'd played 32.f3
Afterwards Robin Kneebone and Colin Sellwood pointed out the line 43.Ke3 Rb2 44.Ra7 Rxb5 45.Rxe7 This is not objectively better as the move played wins. It is superior from a practical point of view, though, because it reduces the scope for errors that throw away the win, a very important consideration at our level! It is also an example of converting one type of advantage into one rhat is different but clearer, the type of practical play that Fischer was noted for.
This throws away the win. Black's threat is to attack the e-pawn with ...Bd5, so white needs to play 45.Rd7 Bb1 "threatening" 46...Ba2, renewing the attack on the e-pawn. 46.b6 Ba2 47.b7 Rb8 (47...Bxe6 48.Rxe7!+-) 48.Rxe7 Bd5 49.Rg7+ Kh8 50.e7 Re8 (A 50...Ba2 51.e8Q+ Rxe8 52.Rc7 Be6 53.Rc6 Bd7 54.Ra6 Kg8 55.Ra8+- B 50...Bxb7 51.e8Q+ Rxe8 52.Rxb7+-) 51.Kc2 Be6 (51...Bxb7 52.Rf7+-) 52.Kc3+- White's king is now free to support the b-pawn.
This would have three ?s if ChessBase allowed it. As Botvinnik wrote in the notes to one of his World Championship games against Smyslov, "Truly the two players were worthy of each other on this occasion". I can't see anything wrong with 56 Bxa1. I had a few minutes left and Simon's flag was horizontal, so I can't even plead time trouble in migation, not that there would have been any valid excuse for missing this.
I believe Simon was unlucky here. By the time we got to move 61 in the following analysis of the final position, I would have been down to less than two minutes. I can't be confident that I wouldn't have played 61.Ke7 rather than the correct 61.Bh6. 61.Ke7 was the move I suggested afterwards.
(The apparent brilliancy 63.Kxe7?? hxg5 64.h5!! 64...gxh5 (64...gxf4 65.hxg6 f3 66.g7 f2 67.g8Q+ Kh2=) is an imaginative way of throwing away the win. 65.fxg5 h4 66.g6 h3 67.g7 h2 68.g8Q+ Kf2 69.Qh7 Kg2 70.Qg6+ Kf2 71.Qh5 (71.Qh6 Kg2 72.Qg5+ Kf2 73.Qh4+ Kg2=) 71...Kg2
incredibly the position is drawn. White's winning plan consists of approaching the black king with checks along the g-file. The king cannot go to h1 because, with the f-pawn still on the board, Qg3 is not stalemate and therefore black is mated with Qf2 and Qf1. After a check on g3, black must either play ...Kh1, losing as discussed above, or ...Kf1, losing the h-pawn. However, in order to reach g3, white must be able to check on g4. This is prevented by the pawn on f5. White cannot capture this pawn because this would produce the drawn ending of Q v RP on the 7th rank. White can therefore make no further progress.)
Black is one move short of the draw.
Notes by Ian George.