Zwischenzugzwang wrote on 07/31/11 at 15:23:21:
After 1.e4 e5 2.Nf3 Nc6 3.Bb5 f5 4.Nc3 fxe4 5.Nxe4 d5 6.Nxe5 dxe4 7.Nxc6 Qd5 8.c4 Qd6 9.Nxa7+ Bd7 10.Bxd7+ Qxd7 11.Qh5+ Kd8! 12.Nb5 Nf6 13.Qe2 Bc5 14.0-0 Rf8, I wonder if 15.d4, as played in the games Vogt - Senarega and Fedorov - Morgado, is so strong; my computer doesn't like it at all. What about 15.Nc3 (threatening the third black pawn and, more importantly, freeing b1 for the rook) and after 15...Qf5 the sacrifice 16.b4 Bxb4 17.Rb1 Bxc3 18.dxc3?
Indeed, these 14...Rf8 corr. games are not convincing. 15.Nc3! Qf5 16.b4! (Zwischenzugzwang) looks strong. If 16...Bxb4 17.Rb1 Bd6, White has 18.Rb5 Qg6 19.h4!, and if 17...Bc5 18.Rxb7 Bb6, the sacrifice 19.Rxb6 cxb6 20.Nb5 gives White a +1.20 position with fine winning chances.
Ivan Sokolov's 14...Re8 may be slightly more precise, after 15.Nc3 (16.b4 Bxb4 17.Rb1 Rb6 +0.60) 15...Ra6 Black's Re8 is more useful than the Rf8. I had already mentioned the idea 15.a4 intending 15...Ng4? 17.b4!. Now I think that even the more logical 15...Ra6 runs into problems after 16.b4! Bxb4 17.Ba3 +0.88, when Black's strongest piece Bb4 is exchanged or pushed aside to the square a5.
So I agree that 13.Qe2 should be the main line (+/-), and 13.Qh3 += is almost as strong and a simple alternative.