sloughter wrote on 11/19/12 at 23:51:41:
While practical play can provide guidance all too often it deviates quickly from best play. Here White makes a strategic blunder and Pinski in The Two Knights Defence, doesn’t even question it. The critical variation is reached after:
8.Qf3 h6?! 9.Ne4 Nd5 10.Ba4! In the game Van der Wiel-Spassky, Reggio Emilia, 1986/87 White made the strategic blunder 10.Nbc3?
This is wrong for four reasons: 1)It gives up the Bishop pair on an open board, 2)It repairs Black’s pawn structure, and 3)It frees the c6 square for the Knight, 4)It opens up the long diagonal for the Queen Bishop. No wonder Black is doing well!
With the superior 10.Ba4 here is analysis from Fritz 12 and Houdini: After this move both engines have White slightly better. Here are a few additional moves to clarify the position: 10…Be7 11.O-O O-O 12.d3 f5 13.Nec3 Be6 14.Re1 Bd6 15.Bd2 Rb8 16.Bb3 Nxb3 17.axb3 Nb4 18.Na3 & it is difficult to see compensation for the pawn.
Pinski, The Two Knights Defence, evaluates 8.Qf3 as ?! This is, in large measure, the result of inferior over the board play and dubious analysis. Take, for example, the following game, Spiegel-Mari Arul, Calcutta, 1997:
8.Qf3 Rb8 9.Bd3 h6?! (This does not mesh very well with Rb8 because it forces White to do exactly what he wants to do mainly 10.Ne4 Nd5 11.b3 (of course) 11...Nf4 (The "threat" of doubling the d-pawns can be ignored here; in the above game White played the horrible 12.Bf1.
White should just hit e5 with 12.Bb2. After this move, Fritz 12 has this as +- & Houdini has this as +/= to +/- Now Pinski gives 12...Nxd3 13.Qxd3? (Why?) White wants to meet the defensive of e5 difficult e.g. 13.cxd3 f6 14.Qh5ch +/- to +-
Both Fritz and Houdini recommend removing the Rook from danger with 13...Rb7. White chops on 14.Bxe5 Re7 15.Qg3 & both Fritz and Houdini give a big advantage White. Black can't try to win a piece with 15...f5 because of 16.Qg6ch so White is simply better.