Black sets up a Caro-Kann Defence formation and maintains the c8-bishop's flexibility, since 6.Bd3 can be met effectively by 6...Bg4 although White can then offer the unclear sacrifice of a second pawn with 7.h3!? intending 7...Bxf3 8.Qxf3 Qxd4 9.Be3.
More normal is 6.Bc4 Bf5 which I examined in a previous blog article and I don't have a great deal to add to what Stefan Bucker analysed in his article, How to Detect a Novelty. My detailed analysis of the BDG main lines features a game in the line I gave in the other article, 7.Bg5 Nbd7 8.Qe2 e6 9.0-0-0 Be7 10.Rhf1 (Guido de Bouver has blogged about the interesting alternative 10.Bxf6, but I prefer the text move, bringing a rook to the half-open f-file) 10...Nd5 11.Bd2 Qc7 12.h3 0-0 13.g4 Bg4 14.h4.
This position led to a wild battle in P.Grott-W.Hort, email 2001, in which White missed a good opportunity to generate a strong kingside attack around move 22, instead grabbing a queenside pawn and losing.
Boris Avrukh gave one important line against one of Stefan's ideas: 7...e6 8.Nh4 (8.Qe2 is met by 8...Bb4! if Black goes for this move-order) 8...Bg6 (if 8...h6 9.Bxf6 Qxf6 10.Nxf5 Qxf5, then 11.Rf1 is quite promising for White) 9.Nxg6 hxg6 10.Qd3 Be7 11.0-0-0 Nbd7 12.h4 a5 13.Bb3 Nb6 14.a4 Nbd5 15.Kb1 Qd7.
Here Stefan Bücker gives 16.Rhf1 Nxc3+ 17.Qxc3 Ne4 18.Qe1 Nxg5 19.hxg5 Bxg5 20.d5! with dangerous compensation for two pawns, while I suggested 16.Ne2!?, aiming to create some trouble by landing the knight on f4.
Of course, at the club level, it is unlikely that either side will have a good enough memory or play accurately enough to head into these lines, and the 7...e6 8.Nh4 Bg6 line remains untested at high levels of play. However, it seems to me that 5...c6 is not quite as big a challenge to White as previously thought.
So is Black's best bet 5...g6? Maybe if White goes in for the Studier Attack (6.Bc4 Bg7 7.0-0 0-0 8.Qe1 Nc6 9.Qh4 Bg4) but in the line 5...g6 6.Bf4 Bg7 7.Qd2 0-0 8.0-0-0 a5 9.Bh6 a4, Mark Nieuweboer has provided some convincing analysis in support of 10.a3:
White blocks Black's a-pawn advance and refrains from making a premature exchange on g7, and as he pointed out, White is doing OK after 10...Bxh6 11.Qxh6, e.g. 11...Bg4 is simply met by 12.Be2, and in the line I briefly gave (10...Bf5) White could instead consider the standard plan of harassing the bishop with h3 and g4 followed by h3-h4-h5, e.g. 11.h3 Ne4 (anticipating the bishop being kicked) 12.Nxe4 Bxe4 13.h4.
Despite the loss of a tempo with h2-h3-h4, White has good attacking chances in this position.
The Summing Up
Therefore, my conclusion is that the most problematic line for White from a theoretical (and also practical) perspective in the Blackmar-Diemer is 5...Bf5 6.Ne5 (6.Bd3 Bxd3 7.Qxd3 c6, intending ...e6, ...Be7 and ...Nbd7) 6...c6 7.g4 Be6. I have nothing to add to what I said in Part 3 of this coverage of the BDG and I am not as convinced by White's compensation here as in the 5...c6 and 5...g6 lines examined above.