Harding merely mentioned that 6.g3 "is also critical". I think Black's best in Moskalenko's line may be to "bite the bullet" with 6...Nc6 7.Nh4 Be6 8.Bg2 f5 (8...Nc5 9.b4 Nd7 10.exd6 Bxd6 11.Nd2 leaves Black with insufficient compensation, e.g.11...0-0 12.0-0 Re8 13.Nhf3 Qf6 14.Ra2 with advantage for White) 9.exf6 Nxf6 10.Nc3 Bxc4 11.Qa4 Be6 12.Nf3 Qd7 (envisaging ...Bh3, e.g. 13.0-0 Bh3 14.Bxh3 Qxh3, after which 15.Qb3?! 0-0-0 16.Ng5 Qh5 17.Nf7 Nd4 gives Black good compensation for the exchange). White's best is probably to prevent ...Bh3 altogether, e.g. 13.Ng5 Bf5 14.e4 Bg4 15.h3 Bh5, when White still has some advantage.
Recapping on the earlier thread, after 4...d6 5.Qc2 d5 etc, Craig Evans gave the following:
Quote:Moskalenko gives a game after 9...Bg4 where white blunders on the next move; whilst the work is a labour of love, his treatment of certain lines is very, very sketchy, and this is one. 10.Bf4 is far more critical, and he should have devoted more than a few lines to it.
10.Bf4 O-O-O 11.e3 and now he suggests 11...Be7, giving 12.Be2 g5! 13.Bg3 h5 14.h3 Be6 "and black has a dangerous attack for the pawn", and mentioning that 12.Bc4 Qe4 13.Be2 asks "another good question".
I would suggest that this is a critical position for the variation. Bizarrely, according to Chesslive.de at least, the move 11.e3 is a novelty, with white normally continuing with h3 or Rc1, which have not done well in practice. 11.e3 seems to make sense though, and after 11...Be7 12.Be2! (I consider 12.Bc4 Qe4 13.Be2 inferior, since black at some point will possibly want to play his queen here anyway, so why give him a tempo) g5 13.Bg3 h5 14.h3 Be6 15.Rd1 Qe4 and white has gained two tempi over the 12.Bc4 line. Now 16.Nd2! looks playable since my original thought of 16...Rxd2 looks insufficient after 17.Kxd2! Qxg2 18.Kc1 and white can untangle after 18...h4 Rh2!.
Therefore, after 16.Nd2, black should probably play Qa4. Here, if white wants to prove he is better, I think he has to man up and play 17.O-O!?, and the question is how dangerous black's attack really is after 17...h4 (17...g4 18.h4 doesn't accomplish too much) 18.Bh2 - and I'm not convinced black has much here.
I think in his line (which may well represent best play after 9.Qxc3, I don't see any obvious improvements for either side), Black has hacking chances on the kingside (e.g. 18...g4 19.hxg4 h3) but White certainly stands better. Btw 16...Rxd2 will leave Black with one pawn for an exchange with best play, e.g. 17.Kxd2 Qxg2 18.Kc1 h4 19.Rh2 Qe4 20.f3 Qg6 21.Bf2! (eyeing a7) 21...Qf5 rounding up the e5-pawn, but at the cost of exchanging queens, as 22.Rhh1 Nxe5? 23.e4! wins the a7-pawn, and 22...Kb8 23.e4 Qf4+ 24.Be3 forces Black to take with the queen anyway.
I can see this sort of thing working for Black in rapid games (it is easy for White to go wrong, e.g. at the end of the first line, the tempting 17.b4? h4! works nicely for Black) but it doesn't look fully sound. 4.a3 d6 5.Qc2 still looks to be the most critical.