People graded below 2000 get better compensation because at those levels, attack tends to be easier than defence, and so Black misplays the position more than White does.  Above 2000, defensive techniques improve and so White finds it harder to generate good compensation.

Even wholly sound gambits, e.g. the Evans, or the Two Knights' Defence for Black with 4.Ng5 d5 5.exd5 Na5, score better for the gambiteer at lower levels than at higher levels.

The BDG scores phenomenally well on the databases, so regardless of its objective soundness (which even I can't vouch for), it works well as a winning weapon at the mean level of play in those databases.  You don't always win by playing the objectively best moves, but rather by playing the moves that most unsettle your opponent.

Dragonslayer wrote on 02/23/09 at 16:26:43:



I played it for 20 years from 1986 to 2006 and maintain the positional compensation in any given Benko is the same for players of any rating.

The position is the position is the position.  Rating level does not change the level of compensation in a position.  It may effect a players ability to understand and exploit said "compensation", but that's about the extent of it.

And I further maintain that any opening that is good below 2000 and not good above 2000 is a most dubious opening, bordering on unsound, as defined my modern Chess Praxis, which seem to be based on the play of people that know what they are doing instead of patzers.


The arguments proffered by BDG Alchemists are just bizarro and are true in some Non-Euclidian reality that I have never visited, but they do not seem to apply to the world in which I am currently spending most of time......

flaviddude wrote on 02/22/09 at 23:51:10:



This must be my problem.


Interesting that people below 2000 get better compensation than those above 2000 in the exact same position.   This is a very cool phenomena that I did not realize existed in chess.

What opening can I play as above 2000+ that will give me compensation while not allowing those under 2000+ to have it?

In my opinion it's probably better to use the Blackmar-Diemer as a part of your repertoire alongside other openings.  There's a temptation to try to get it in nearly every game (which is possible, as ArKheiN points out) but as with any opening, playing just the BDG all the time gives a limited range of positions.

Alongside one or two other 1.d4 lines, though, I don't see a problem with using it.  I personally use a different approach- I play 1.e4, but sometimes transpose to the BDG from 1.e4 openings (eg 1.e4 d5 2.d4, 1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.f3).  It can even be used against some lines of the French (e.g. 1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.Bg5 dxe4 5.f3; also 3.Be3 dxe4 4.f3 is playable)

I don't think you need to learn much theory- at least not at club level- as the ideas are pretty easy to grasp.  The most popular approach is indeed 0-0 Qe1-h4, and as Tim McGrew mentions, sacrifices on h6 (after Bg5, Qh4, ...h6, Bxh6) are common as well, as in practice Black often puts the question to the white bishop at an inopportune moment.  The g-pawn push is also common.  If Black does not play ...h6, then h7 is often an important attack point, and the rook on the f-file opens up possibilities of Rxf6 sacrifices followed by crashing in on h7.

I often use an alternative approach of queenside castling, and playing for the d4-d5 push as well as on the kingside (it has the disadvantage of not utilising the f-file as strongly and harder to play Qe1-h4, but it makes the kingside pawns more free to roll at the enemy king).  In my experience it's especially effective against the ...g7-g6 lines, when the h-pawn push is effective, but maybe less so against ...Bf5 lines.

Statistically the BDG scores very well- you'll only find trouble getting compensation for the pawn if you significantly exceed 2000 Elo, I reckon.

Some questions for BDG players:

I want to try out this gambit in order to improve my tactics. What is the best way to learn this opening, and does it require a lot of theory? Any good books, maybe?
And how often can you actually get a BDG? If many players choose to decline of enter a sideline, then I might find it easier to get tactical positions with another opening.

It's hard to assess every positions of the Zilbermints gambits. I think it's unsound but Black must be cautious. For example: 1. d4 Nf6 2. f3 d5 3. e4 dxe4 4. Nc3 exf3 5. Nxf3 e6 6. Bg5 Be7 7. Bd3 Nc6 8. O-O Nxd4 9. Kh1 c5 10. Ne5 O-O 11. Qe1 Nd5 12. Bxe7 Qxe7 13. Ng3 f6 14. Ng6! is arealdy better for White. But ok Black may have improved somewhere, but even a computer don't see that strong 14.Ng6 immediately. 9..c5 is probably playable.

See my analysis on the other thread- I cannot believe that 3...f5 is Black's best move, as it tends to transpose to the Staunton Gambit which is sound enough, if insufficient for advantage against best play.

Looking over Lev's post, perhaps I should get 1.e4 d5 2.exd5 Nf6 3.Nc3 Bg4 4.Be2 h5 (which I played a few times at online blitz) named after me.  Or at least Lev seems to like naming every move he plays after himself if it doesn't already have a name.

The gambit with 8.0-0 offers practical chances, I don't deny that, but the line with 8...Nxd4 9.Kh1 c6 doesn't give White enough compensation.  We've discussed the Euwe on other threads and the general consensus was that 7.Qd2 first was better.

IM Christoph Wisnewski wrote on 02/07/09 at 20:03:19:


I look forward to purchasing it, using it and perhaps learning something from it (I am sure I will!).

How can we get you to maybe next look at King's Gambit?