15. Qc2 is a good try for an advantage (17... cd 18. ed is more pleasant for white). Anka didn't look at it that deeply. 

Miroshnichenko's 19...f5 looks positionally dubious (19. Qc2 seems more promising than Qf4 at first glance) but he only lost as he blundered in the endgame (30... Rdxf2?? when 30... Kg8 was the only move and equal).

Well, Frank Marshall played it first in 1912. 

I didn't mean that Ponomariov had come up with the novelty, but that 15.Qc2 seems to be the new way to play for an advantage. I should have been more clear.

I'd be curious to see these equalising paths for Black.

Btw, I found the Topalov-Naka game, but it took me a while. I don't know why it didn't show up in my previous searches.

Still, it looks like white retains a small advantage even in the Topalov-Naka game. The question is whether it was enough to bother trying for more. Ponomariov seemed to think it is.

Here's one way to play for an advantage:



The new move is 15.Qc2. Before that, they followed Topalov-Anand (World Championship, 2010) where Topalov played the "normal" 15.Be4.

There's also a game that Ponomariov won fairly recently against Ne4. There's still hope for a White advantage in the QGD. But it really is getting harder and harder. Especially at the top levels and in correspondence chess.   

Where are the fun QGD lines where we can sac material to reach positions that the engines just don't understand, as in the King's Gambit lines?  I don't mean for OTB chess, but for correspondence chess. 

(I used to believe the King's Gambit was only good for blitz and an occasional surprise in OTB tournaments. But now, thanks to the really high quality posts on this site, I've become a believer.)

MNb wrote on 07/25/11 at 10:17:44:


huibui wrote on 07/25/11 at 14:03:12:

I play 3.Nc3 and after Bb4 4.e3. Sure Black is in good shape, but there is a lot of play left. I've won a few games already in corr.chess. The theory buffs can consult Vigorito's book on 4.Qc2.
I'm not trying to prove a theoretical += in all lines. 1.e4 c5 is already equal too, but it's not the reason I switched to 1.d4. Unbalanced equality is good enough for me.

MNb wrote on 07/25/11 at 10:17:44:



Unfortunately both the exchange variation and the lines that postpone the development of the Ng1 can be avoided by the 1.d4 Nf6 2.c4 e6 3.Nf3 d5 move order.

3.Nc3 is possible of course, but the Nimzo seems to be in good shape at the moment (see Dortmund).

That's my impression too, despite seriously lacking theoretical knowledge.
Besides the Catalan there is also the Exchange Variation.
The weird thing is that I played the latter twice and ended up drawing slightly inferior positions. Then I tried 4.Bg5 twice too and won. So maybe all hope is not lost for us amateurs.
Anyhow, an idea that intrigues me is 4.Bg5 Be7 5.e3 h6 6.Bh4 0-0 7.Rc1 evt. developing the Knight to e2.

Markovich wrote on 07/25/11 at 01:29:59:


I honestly feel that's what chess theory is approaching in the QGD. Black can choose equalizing lines analyzed out to the endgame in both the Lasker and 5. Bf4 0-0 6. e3 Nbd7 variations, and white's alternatives are sorely lacking.

From a practical perspective I think the Bxf6 lines (instead of Lasker/Tartakower) are a good try, but even the line you mention with 0-0-0 and pawn storming the kingside is easily defanged (perilously so - once white's pawn storm is stopped - and it can be - black's begins).

White needs something major, otherwise all that's really left is the Catalan.

I hope I'm wrong.