Leela is quite good at defending the positions too I think - it definitely finds some of the thematic exchange/pawn sacrifices.

With many of the 7.. o-o lines you're right - white has to know quite a bit, and with 9 Qh5 over a huge number of lines too.

9 Bg5 is a bit of a worry in that black seems to only have about one viable way to play, and the position is somewhat under control.

Its a hard opening to write a book about!

Watson did one a bit back, main line vs 9 Qh5 shortly more or less refuted & various of his ideas vs 9 Bg5 with it. Berg 'lost' his main lines vs 9 Bg5 and a huge dent in his one vs the 9 Qh5 stuff. Awfully good books from excellent authors.

The truth is probably that it is a bit of a nasty struggle vs very accurate play.

LC0 seems quite balanced overall to me really about these positions. It loves space advantages and slow, controlled attacks so naturally very keen on some of the positions white can get.

Its one reason I'd be a little bit wary of those correspondence draws - I've a feeling that Leela and friends will handle these positions as white much better than SF etc. Might prove wrong if tested with an tournament of course!

Leon_Trotsky wrote on 10/07/19 at 00:16:53:


You can play around analyzing with Lichess's analysis tools, or search directly in the ICCF archives.

Leon_Trotsky wrote on 10/07/19 at 00:16:53:


I am not sure where you are going to find expert opinions, but correspondence games are usually quite close to the "truth" of an opening variation. In the past couple of years White has had an obscenely high plus score with 9. Bg5.

I am sure Aagard is aware of these recent developments, and agree with his assessment that Black needs to play 8...f5. Even then, the "main line" of 8...Nbc6 has been the main line for a reason: I wouldn't expect Black to be able to equalize with 8...f5 either.

Edit: Looking at the 2018 ICCF archives, it seems that that three correspondence games have reached the endgame with 15. Ba3 Qf5.

In one correspondence game, White played non-critically. Black was able to play ...a5, build a fortress, and hold the draw.

In two correspondence games, White played a brutally direct plan with f4, g4, and h4 and won the game. In one of these games, both players were rated above 2300.

Honestly, a lot of these correspondence games are freely available online through Lichess's Opening Explorer tool. Here is a link to the recent correspondence game with 9. Bg5:

https://www.iccf.com/game?id=934223

Now that I think about it, in the main lines with 13...Qf7, the plan with Qe2 and g4 might be the most dangerous continuation. Of course, all of this discussion of 9. Qh5 is beside the point if 9. Bg5 becomes a refutation of the whole line. At least, Miedema and Aagard seem to think so.

Well, if you look and it in depth and still like it then play it, but put a lot of work in first It very definitely isn't responsible for white not playing Qh5.

Last time I saw this line in the past few years, Black drew with 19...Rf7 in correspondence games. As of right now I do not know what the state of theory here is.

Last year there were some correspondence games where all were draws in 13...b5 by Berg and White played Negi's 15. Cg5 there. I think that Watson had already also did an update on this line. I recall it being very long analysis, but basically Black was fine if I remember.

Which lines do you think Black struggles in in Watson's updates.

Regarding 9. Ag5, I had analysed this for months and came up with some interesting replies. Although whether Black players in the 7...0-0 line have found them as well, who knows..