One of the ideas in the 4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Bd3 line is 8…h6 9.Ne4 Nd5 10.Be2! Be7 (Qb6 11.O-O Be7 12.d3 O-O 13.Nbd2 f5 14.Nc3 Be6 15.Nb3 Nb7 16.Bf3 +/=) 11.d3! (It is important to free the d2 square for both Knights) O-O 12.O-O f5 13.Ned2 Be6 14.Nb3 (We view this as critical to White’s ability to shut down Black’s counterplay on the Queenside; shielding the b-file is highly desirable) Nb7 15.Nb1d2 Now it is clear that Black is struggling to find compensation for the pawn. At a depth of search of 23/50 Fritz has 15…Qc7 and 15…a5 as +/=. Houdini has the evaluation as = to +/= after 15…a5, Qc7 or Bf6   

We see a victory by White in Laznicka (2617)-Lopez (2555), Pamplona, 2009: 8.Bd3 h6 9.Ne4 Nd5 10.Ng3 g6?! (A known concept but this seems slow. We recommend as best for both sides: 11.Be2 Bg7 12.d3 O-O 13.Nbd2 f5 14.Nc3 & it doesn’t matter whether Black plays, 14…Nb7, Be6, Qc7 or Rb8, 15.Nb3 is +/=) 11.O-O Bg7 12.Re1 O-O 13.Bf1 Qc7 14.a3!? (The start of a far-ranging tactical duel; this proved to be a very sharp middlegame) 14…Rd8 15.b4 Nb7 16.Bb2 a5 17.c4 Nf4 18.Qb3 Be6 19.d3 Rd7 20.Nd2 Rad8 21.Re3 f6 22.Qc2 Bf7 23.Nb3 Ne6 24.Rae1 axb4 25.axb4 Qb6 26.Bc3 h5 27.h4 +/- 1-0:58 

Here is a high level test of 8…h6---Short (2684)-Sokolov (2655), London, 2009

8.Bd3 h6 9.Ne4 Nd5 10.O-O g6 11.Re1 Bg7 12.Bf1 O-O 13.d3 f5 14.Nc5! Qd6 15.Nb3 +/= to +/- We see Short wrap up the win:

15…Nb7 16.g3 Be6 17.c4 Nb6 18.Qc2 Nd7 19.Nc3 a5 20.Na4 g5 21.Bd2 Bf7 22.Bc3 Rfe8 23.d4 e4 24.Rad1 Bg6 25.c5 Qc7 26.d5! cxd5 27.Rxd5 Bxc3 28.Qxc3 Ne5? 29.Red1 f4 30.c6 Qxc6? (Not really a blunder because the position is already +-) 31.Rxe5 1-0