Opposite Exceptions (part three)

Black has just captured on e4 in Spanton (163) - T Bean (173), Sutton (Surrey) 1995. How big is White's advantage?

*****

*****

*****

*****

*****

*****

*****

*****

*****

*****

Stockfish12 gives White a tiny edge (it actually shows the = sign, but with White up .25 of a pawn), while Komodo11.01 reckons White has the upper hand (+.75 of a pawn).

40.Bd2!?

This is the move I relied on when I exchanged queens on move 39. How should Black respond?

*****

*****

*****

*****

*****

40...f3

Very natural, but the engines want Black to play 40...e3!? After 41.fxe3 fxe3 42.Bxe3 White is two pawns up. Moreover the extra pawns are passers and are in different sectors of the board. Nevertheless it is far from clear White is winning. Komodo11.01 still gives White 'just' the upper hand, while Stockfish12 only concedes White a slight edge. A sample continuation runs: 42...Ke6 43.Kf2 Kf5 44.Bh6 Ke4 45.Bg7 b5 46.Ke2 a5 47.g3 Bd7 48.Kd2 Be6 49.Kc2 Bf7 50.a3 Bd5 51.Kc3 Be6 52.Bf6 Kg3 53.Be5 Ke4 54.Bc7 a4 55.bxa4 bxa4 56.Be5 Kd5 57.Kd3 Bf5+ 58.Ke3 Bd7 59.Kf4 Ke6 60.g4 Bb5 61.g5 Bd3 62.Ke3 Bf5 63.Kd2 Bg6 64.Kc3 Be8 65.Kc4 Bf7 66.Kb4 Bb8+ 67.Kc5 Bg6 68.Bf6 Be4. Both engines have long had White winning, and by quite a margin (Komodo11.01 reckons White's advantage is worth almost seven pawns). But as far as I can see there is no way for White to make progress. This line is far from conclusive as there are many alternatives for both sides at almost every move, but I do think it is indicative of the problems White faces.

41.g4

Black has a much harder task now. True, Black is only one pawn down, but as well as having two passers to cope with, Black has a weak e4 pawn and, by extension, a weak f3 one.

41...Ke6 42.Kh2 b6?!

Black wants to restrict the white bishop by putting queenside pawns on dark squares, but this is probably the wrong plan. Another way to restrict the white bishop is to play 42...Kd5, obliging White to reply 43.Be3 (43.Bc3? e3! 44.fxe3 Ke3 is fine for Black). The drawback with 42...Kd5 is it takes the king away from the passed g pawn, but it is not clear White is well-placed to take advantage of this.

43.Kg3 a5 44.Kf4 Bd5 45.g5 Bb7

Not 45...a4? 46.bxa4 Bxa2 47.Kxe4.

46.a3!?

This is Komodo11.01's choice, but Stockfish12 is not keen. The idea is to open lines on the queenside with b4.

46...Bd5 47.b4 axb4!?

The engines prefer this to 47...a4 48.Kg4.

48.Bxb4 Bb7

When I first analysed this game, I thought 48...e3!? draws. The engines disagree, but after, say, 49.Kxe3 Kf5 50.Be7 b5 51.Bf6 Be4 52.Kd2 Ke6 53.Kc3 Kd5 54.Kb4 Kc6 it is hard to see how White makes progress.

49.Bd2 Ba8 50.Kg4 Bc6 51.Kh5 Kf7 52.Kh6 Bd5 53.Kh7 Kf8 54.Kg6 Bg8 55.Kf5 Bd5 56.Ke5 Bb7 57.d5

Black's position is hopeless.

The game finished:

57...Ke7 58.Bb4+ Kf7 59.Kxe4 Kg6 60.Bd2 Kf7 61.Ke5 Bc8 62.Kd6 bb7 63.Be3 b5 64.Kc5 Ke7 65.Bf4 1-0