It's amusing that imbalances are such a concern when Nf6/e6 (or more specifically NID/QID users) players have to put up with the London and the near auto-draw it triggers. In comparsion the 2. Bg5 e6 Tromp is dynamic. 

(If anyone's wondering why I called the London an auto-draw, see Prie's annotations to Grachev-Leko, Tal Memorial Blitz, 2008 in CBM 127 - even the suggested improvements are dead equal)

Ametanoitos wrote on 02/18/10 at 14:19:46:

I kind of thought you did  I'm going to have a closer look at this line myself, some local youngsters have started playing the Tromp against me and it would be a bit embarassing to lose to my old favorite weapon!

I wonder if Palliser in his new "Starting Out" book (which I haven't bought yet) mentions these aggressive 2...g6 lines? Did he even realize they were a danger to White?

Ametanoitos wrote on 02/17/10 at 23:38:56:

Interesting. This is a line I never took very seriously when I played the Tromp as White, so that may be a further practical advantage of it. I think it also got a page in Burgess' old book "101 Opening Surprises".

Ametanoitos wrote on 02/17/10 at 23:38:56:

This plan was Davies' recommendation against the Tromp in "The Chess Player's Battle Manual" (1998). There are other move orders for White to reach his standard setup: He can throw in e3, c4, N(b)c3 and N(g)e2 in some order before g3. I assume you have it under control and you have enough waiting moves to simply wait with ...h5 until White plays g3? ...f5, 0-0 and Nd7-f6 are fairly standard moves anyway. By 2005 Davies had changed his mind and in "The Trompowsky, 2nd Ed." he doubted that Black could equalize. His main game there was:

[Event "GRE-chT"]
[Site "Athens"]
[Date "1997.??.??"]
[Round "5"]
[White "Skembris, Spyridon"]
[Black "Nikolaidis, Ioannis"]
[Result "1-0"]
[ECO "A45"]
[WhiteElo "2450"]
[BlackElo "2565"]
[PlyCount "79"]
[Source "ChessBase"]

1. d4 Nf6 2. Bg5 d6 3. Bxf6 exf6 4. e3 f5 5. c4 Nd7 6. Nc3 g6 7. g3 h5 8. h4 Nf6 9. Bg2 Bh6 10. Nge2 Ng4 11. Qd3 c6 12. b4 a6 13. Nf4 O-O 14. a4 Re8 15. Kf1 Qe7 16. Bf3 Ne5 17. Qe2 Nxf3 18. Qxf3 Be6 19. Nxe6 Qxe6 20. d5 cxd5 21. Nxd5 Rac8 22. Nb6 Rc6 23. a5 Bg7 24. Rb1 Bf6 25. Kg2 Bd8 26. b5 axb5 27. cxb5 Rc2 28. Nd5 Bxa5 29. Rhd1 b6 30. Rbc1 Rec8 31. Rxc2 Rxc2 32. Nf4 Qe5 33. Rd5 Qe7 34. Rd4 Rc5 35. Rd5 Rc2 36. Rd4 Kg7 37. Nd5 Qe5 38. Qd1 Rb2 39. Qa1 Rxb5 40. f4 1-0

To me it looks like Skembris was very careful to control all threats to e3 and f2, but I haven't really analyzed the game.

TN wrote on 02/17/10 at 21:35:44:


If Black plays the QGD, he can meet 1.d4 with 1...d5, for crying out loud! 

"2...e6 is just as effective as 2...c5." Just as effective for producing the draw, I'll buy that, but not just as effective for producing the win, imho.  But my main point had to do with repertoire considerations.  Unless Black is a Nimzo player (or a QGD player who unaccountably forgot to play 1...d5), then 2...e6 doesn't make much sense, nor is it very enterprising.  Why allow the Torre when you don't have to?

Markovich wrote on 02/17/10 at 16:04:37:


If Black plays the QGD, he can meet 3.Nf3 with 3...d5, when after White's best move, 4.c4 or 4.e3 Be7 5.c4, we are in a main-line QGD for Black. This does enable White to avert the critical 1.d4 d5 2.Nf3 Nf6 3.Bg5 Ne4!, though. Black can also play 3...c5 4.c3/e3 d5, although then Black has to learn a new variation.

I agree that 2...c5 is a good antidote, but 2...e6 is just as effective. The difference is that both moves lead to positions and pawn structures of a different nature.

White has a few other third moves. I would prefer the flexible 3.Nd2, evt. playing a Stonewall with the bishop outside the pawn chain.