I think white just pushes 48. h6 and if Ra1 49. Rxa1 or else 49. h7 Rh8 50. a8=Q Rxa8 51. Ra1+ K-wherevuh 52. Rxh8 with ample time to reach the back rank and stop the black pawns.

That's the case with today's Tuesday (48. ?) puzzle, as 48. h6! +- (+11.16 @ 28 depth, Stockfish 8) offers Black an enprise Rook. Despite the loss of the Rook, White will win the race to promote a pawn with decisive advantage.

25…Nd7?! In fact he had two good possibilities: the natural and sharp 25…Rc2!? and the simple 25…Bg6. Let’s look at 25…Rc2 first. White is now obliged to play 26 Bb4! (why didn’t I make this move earlier?) since 26 Bb2? would fail to 26…Nxd5 27 Nxg7 Rxb2 28 Nh5 Rc2! 29 b4 Qb6 30 Nf6+ Kh8 31 Nxh7+ Qxd4 32 Nxd4 Rc4 33 Nxf8 Rxd4. And after 26 Bb4 Black’s best response is 26…Nc8, as variations such as 26…Rxa2 (if 26…Bg6 27 Nf4 Rxa2 28 Nxg6 fxg6 29 Re6 g5 30 Bc3 with compensation) 27 Bc3 f6 (27…Rc2 28 Nxg7 Rxc3 29 Re8 Qc7 30 Nh5 Rc1+ 31 Kh2 f5 32 Rxf8+! Kxf8 33 Qh8+ Bg8 34 Qxh6+ Ke7 35 Qg7+ Bf7 36 Ng5 wins) 28 Qg4 (28 Nf4?! Nd7 29 Ne6 Qc8) 28…Nd7 29 Re6 Kh8 30 Nxg7! look pretty dangerous.

However, after 26…Nc8 it’s White who has to be very careful. Probably he is not worse after 27 Qg4 (if 27 a4 Ne7) 27…Bg6 28 Nf4 Ne7 29 Nxg6 (29 Nh4 Qb6 30 Nhxg6 Qxf2+ 31 Kh2 Nxg6 32 Nxg6 fxg6 is good for Black) 29…Nxg6 30 Qf5 (30 a3 or 30 a4 also lead to unclear play) 30…Rxa2 31 Qb1 Rxf2 32 Kxf2 Qb6+ 33 Re3 Qxb4 34 Qe4, but I doubt he would have serious winning chances.

But sometimes one doesn’t feel like calculating anything, in which case Black could just try 25…Bg6!? 26 Nf4 Bh7 when, objectively speaking, White should probably repeat the position with 27 Nh5.

Once again I look at the game, played many years ago, with a strange feeling. I had a good position in the opening, then played logical moves but got nothing out of the whole thing. And even worse: for six years I couldn’t understand why. However, today it’s clear to me that both 16 Ba3 and, later on, 21 Ne4 were wrong and therefore I finally put the mark ‘?!’ on those moves. Instead I could have played prophylactically with 16 a4 or 21 Bb4, which would be more in Tigran Petrosian’s style. Black’s pieces, especially the knight on b6 and bishop on d7, would be far less active than in the game and White would have control over the action. Trying to attack the enemy king is good of course, but preventing your opponent’s play is sometimes better!

Well, back to the game… Thanks to Short’s desire to defend an inferior ending, I was able to demonstrate reasonably good technique and still feel satisfied at the end of all.