A castled king may be secure behind pawns until the opponent breaks down the door.
Mikhail Tal White
David Bronstein Black
Tbilisi 1982
Caro-Kann Defense B16
1 e4 c6 2 d4 d5 3 ♘c3 dxe4 4 ♘xe4 ♘f6 5 ♘xf6+ gxf6 6 c3 ♗f5 7 ♘f3 ♕c7 8 g3
The pawn at g3 supports ♗f4 and blunts the impact of ...♖g8.
8...♘d7 9 ♗g2 0-0-0 10 0-0 e6 11 ♕e2 ♗d6? 12 c4! c5
Black naturally didn’t want to allow 13 c5 ♗e7 14 ♗f4.
13 b4!
But he missed how this opens a path to his king (13...b6 14 bxc5).
White would be winning after 14…a6 15 ♘xf5 exf5 16 c5 ♗e7 17 ♖b1, threatening 18 c6.
13...cxb4
After 13...cxd4 14 ♘xd4 the main threat is 15 ♘b5! ♕b8 16 ♗e3.
14 c5! ♗e7
Now 15 ♗f4 is good (15...♕a5 16 ♘h4 ♗g6 17 ♖fc1). White had another fine idea.
15 ♘h4 ♗g6 16 ♘xg6 hxg6 17 ♖b1
So that 17...♕a5 18 ♗d2 and a capture on b4 wins. Or 18...♕xa2 19 ♖a1 ♕b3 20 ♕b5. And the queenside is ripped open by 17...a5 18 a3!.
17...♘b8 18 ♗f4 ♕d7
Question 370: Was 18...♕a5 19 ♗xb8 ♔xb8 better?
19 ♗xb8 ♔xb8
Black resigned before 20 ♖xb4 (20...b6 21 ♕a6 and 22 ♖xb6+!).
Maia Chiburdandize White
Daniel Borner Black
Biel 1991
1 e4 c6 2 d4 d5 3 ♘d2 dxe4 4 ♘xe4 ♘f6 5 ♘xf6+ gxf6 6 c3 ♗f5 7 ♘f3 e6 8 g3 ♕c7 9 ♗g2 ♘d7 10 ♘h4 ♗g6 11 0-0 0-0-0 12 ♕e2 ♗d6 13 ♗e3 f5 14 ♗g5 ♖de8 15 c4 c5 16 b4! cxd4 17 c5 ♗f8 18 ♖ac1 ♘b8 19 b5 ♗g7 20 b6 resigns.