The most popular defence to the Blackmar-Diemer Gambit is 5...Bg4, pinning the knight on f3, and keeping open the possibility of swapping it off at the right moment. I don't think it's one of the most critical, though. After 6.h3!, 6...Bh5 7.g4! Bg6 8.Ne5 tends to transpose into the variations with 5...Bf5 6.Ne5 e6 7.g4 Bg6, but denying Black the critical alternative 6...c6 7.g4 Be6. I think White has enough for the pawn here.
You might wonder, why play 6.h3, isn't it just a waste of a tempo to provoke Black into exchanging? Well, not quite, because if White just develops normally, Black is more likely to just leave the bishop on g4, keeping the knight on f3 in an awkward pin. Also, the move 6.h3 prepares g2-g4, which threatens to kick the black knight away from f6 at a good moment with g4-g5. Then in various lines the bishop on f1 can come out to g2, rather than its more traditional post on c4. |
More consistent is to meet 6.h3 with 6...Bxf3 7.Qxf3 c6 (7...Nc6 walks into an awkward pin after 8.Bb5, threatening 9.d5).
Then it's not clear if White can get enough compensation for the pawn with the routine 8.Be3, but White has a more violent alternative in 8.g4!?, sacrificing a second pawn on d4, with the idea of quickly bringing a rook to f1, playing g5 and then breaking through to Black's weak point on f7 (this highlights another useful point behind 6.h3). If Black takes on d4, White can play Be3 and 0-0-0 and gets very dangerous compensation, while if Black doesn't take, White tends to get into favourable versions of the 8.Be3 lines. |
5...e6 is a solid but passive defence to the Blackmar-Diemer, shutting in the bishop on c8, and aiming for rapid kingside development. If White plays 6.Bd3, 6...c5! is quite awkward to meet, so White does best to pin the knight on f6 with 6.Bg5. Then if 6...c5, 7.Bxf6, intending 7...Qxf6 8.Bb5+ followed by 0-0 and/or Ne5, gives White a dangerous initiative. Thus, Black tends to continue with 6...Be7 (diagram).
Here White has two main options. |
7.Bd3 is the traditional main line. Joe Gallagher and James Rizzitano have recommended 7...c5, but I think White can get away with 8.dxc5, followed by either 9.0-0, or 9.Qe2 and 10.0-0-0, with plenty of compensation for the pawn in either case. Also, Black should probably avoid the "automatic" 7...0-0, because then 8.0-0 followed by Qe1 and Qh4 gives White a dangerous attack, with the bishop on d3 pointing at h7, and the threat of Rxf6 sacrifices.
The most challenging reply to 7.Bd3 is rather 7...Nc6, with the idea of exchanging off the bishop on d3 with ...Nb4 and ...Nxd3. White can continue with either 8.a3 (preventing ...Nb4, but at the cost of time) or 8.Qd2 (allowing the exchange). Lev Zilbermints has favoured 8.0-0 Nxd4 9.Kh1, sacrificing a second pawn, but if Black isn't tempted by 9...Nxf3?! 10.Qxf3, which is dangerous for Black, Black has good chances of getting the upper hand. |
The other way to continue is 7.Qd2, preparing queenside castling. Here White can continue with Qd2-f4-h4, and get the queen across that way. White has had many nice wins where Black has played ...0-0 and ...h6, and fallen for a very strong Bxh6 sacrifice.
|