1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3+ 6. bxc3 Ne7 7. Qg4
Better than 8. Nf3 Nbc6 (8... f5 9. exf6 Rxf6 10. Bg5 Nd7!? 11. Bxf6 Nxf6 12. Qf4 Qa5
)
9. Bd3 f5 10. exf6 Rxf6 11. Bg5 e5!
as Kindermann has demonstrated in a number of games.
8... f5!
This move is more forcing than the alternatives and therefore limits White's choices. Other possibilities:
a) 8... Nbc6 9. Qh5!? Ng6! 10. Nf3 Qc7
is well-discussed by John Watson in Play the French and by Neil McDonald in French Winawer.
10. Qh5 h6!? (10... g6 likely
transposes to lines similar to the game) 11. g4!?
Also playable, if less reliable, is 11... h6 ,
about which Charbonneau writes that it is "thought to be dubious (...h6 after
Qh5)." Yet
the move is considered one of the main lines in Kindermann's wonderful book
on the Winawer. Charbonneau's game, which he discusses at his blog,
went: 12. Bxe7! (12. Bh7+!? Kxh7 13. Qxf7)
12... Rxe7 13. Nf3 c4 (13... Qa5! 14. )
14. Bg6! Nc6 15. 
0-1
Charbonneau - Pelletier, France 2006
Also possible is Kindermann's recommendation 12... Qa5 13. Bd2 Nbc6 14. Nf3 Qc7 15. dxc5! e5 16. Ng5 Rf8 17. c4 e4 18. cxd5 exd3 19. d6 Lutz
- Pelletier, Biel 2003, which John Watson analyzes as giving roughly equal
chances.
13. Nf3 Qf8 14.
The idea is to defend the dark squares, but the Bishop is passively placed and has less future here than at c1, where it can reroute to the a3-f8 diagonal after a4 and Ba3, or Bh4-g3 controlling e5 directly. Riordan realizes this later, losing a tempo on the immediate Bc1.
a) 16. Bc1 Bd7 and Black has an easy time gaining control of e5 if he plays correctly: 17. Ne1 (17. a4 Qg7 18. Qd2 Kh7 19. Ba3 g5 20. Rab1 b6 21. Bd6 Ng6!?
1-0 Ponomariov, R-Ivanchuk,V/Linares 2002/[Psakhis] (45)
b) 16. Bh4 helps White keep more control over e5, but the Bishop will be exchanged, weakening the kingside: 16... Nf5 17. Bg3 g5 18. Qd2 Bd7 19. h4 Qe7 20. hxg5 Nxg3 21. fxg3 hxg5 22. Qe3 Raf8
16... Bd7
16... Nf5 17. Qc1 g5 18. Ne5 Nxe5 19. dxe5 Qg7 20. Bh5 Rf8 21. f4 g4 22. Qd1 g3 Kindermann
17. a4
White admits his mistake, since he can make little progress fighting for e5 without getting his dark square Bishop into the game.
If instead 17. Re1 g5 18. Bf1 Ng6 (18... Qg7 19. Ne5!=)
19. Qe2 Qg7
controlling
e5.
17... Kh7 18. Bc1 Qg7 19. Ba3 g5 20. Bd6 Ng6 21. Ne1?!
White is already slightly worse, but this passive plan of defense helps Black. The Knight should keep an eye on e5 and be ready to leap there to force exchanges into a difficult ending.
Stoyko said that one of the problems of this position for Black is that White can just sit and wait, and even sacrifice at e5 if necessary, when it is very hard for Black to win the game. To show this he examined at some length with us the "seemingly ridiculous move" 21. Qe1!? g4 (21... Rg8 22. Ne5)
22. Ne5! Ngxe5 23. dxe5 Rf5 24. f4 gxf3 25. Rxf3 Nxe5 (25... Rg5 26. Qf2 Rg8 27. g3 Nxe5 28. Rf6)
26. Rxf5 exf5 27. Bxe5 Qxe5 28. Bxc4 Qxe1+ 29. Rxe1 Bxa4 30. Bxd5 Bc6! 31. Bxc6 bxc6 32. Re7+ Kg6 33. Re6+ Kg5 34. Rxc6 a5 35. Rb6 and though Black is still better, the win is not so easy--and Black's play in this analysis was computer-assisted!
21... Nf4 22. Bf3 Rg8 23. Bxf4 Rxf4
If 23... gxf4 the g2-square is well fortified, which is probably what White had expected.
25. h3!?
White really could not stop this thematic break once his knight got out of position.
Now Black's pieces are overwhelming.
29... Qh3!? 30. Ne3 (30. fxe5 Rxf1+ 31. Qxf1 Rxf1+ 32. Rxf1 Qe6)
(30. Qd2 Ng4)
30... Rxf4!! 31. gxf4 Qxe3+ 32. Rf2 gxf4 33. Qxd5 Rg8+ 34. Kf1 Ng4
30. Qd4 gxf4 31. gxf4 Qg4 32. Qxa7 Nxf4 33. Qxb7+ R8f7 34. Rxf4 Rxf4 35. Qxd5 R4f5 36. Qd4 Qe2 37. Re1 Rf1+
White resigns due to mate next move.
0-1