SeanWinshand vs sokidze: game review

4…Nd6! The only good move here: every alternative is clearly worse.

• Moves the attacked knight to safety.
• The next best move, 4…f5, would drop Black's winning chances from 51% to 22% (+3.45).
• 4…Nd6 gets the knight out of the way of fxe4.
• White was threatening fxe4, winning a knight for a pawn. The knight steps away.
• White answers 5.Nc3, attacking the d5 pawn, which is undefended.
• Engine line: 5.Nc3 e6 6.h4 Nf5 7.Bf4 Bd6 (−0.15).

12…bxc5! The only good move here: every alternative is clearly worse.

• Takes the pawn on c5.
• Protects the knight on d6, which was hanging.
• The next best move, 12…Nf5, would drop Black's winning chances from 54% to 38% (+1.36).
• 12…bxc5 stops cxd6.
• White was threatening cxd6, winning a knight for a pawn. bxc5 takes care of it.
• Trades pawns.

17…Nd6! The only good move here: every alternative is clearly worse.

• Pins the bishop on f3 to the rook.
• Moving away uncovers the bishop on b7, which now hits the bishop on f3.
• The next best move, 17…f5, would drop Black's winning chances from 51% to 34% (+1.85).
• 17…Nd6 gets the knight out of the way of Nxe4.
• White was threatening Nxe4, winning a knight. The knight steps away.
• White answers 18.Bxb7, taking the bishop on b7, leaving Black with no bishop pair.

31…Ra5 is a mistake: Black's winning chances go from 47% to 30%.

• White answers 32.c4, which pushes the passed pawn to c4.
• Better was 31…Ra1+ (+0.34): check, and White has only one legal reply.

Better: 31…Ra1+ gives check, leaving one reply. Engine line: 32.Kb2 Rh1 33.Nd7+ Ke7 34.Nb8+ Kd6 (+0.34).

32.Nc4 misses the chance 31…Ra5 gave: c4 would have punished it. White's winning chances go from 70% to 55%.

• Black answers 32…Rf5: the rook takes the open f-file.
• Better was 32.c4 (+2.25), which pushes the passed pawn to c4.

Better: 32.c4 pushes the passed pawn to c4. Engine line: 32…Ra6 33.Nd7+ Ke8 34.Ne5 Ra5 35.Nd3 (+2.25).

• Both sides followed opening theory up to 2.g4: 3 moves in all, 2 by White and 1 by Black.
• Theory runs out after 2.g4: too few games between strong players reached this position to go further.