SWJediknight wrote on 09/15/14 at 12:36:11:


11.g4 (Watson) is almost winning.

I think 5.g3 Nge7 is sound, but that White can return the pawn and get a slight edge (6.Bg2 Ng6 7.0-0 Nxe5 8.Nxe5 Nxe5 9.b3 as played by Ivanchuk against Morozevich in 2012)- this avoids Black's various attractive attacking lines that follow the attempt to disrupt Black's efforts at regaining the pawn with 7.Bg5.

The problem with 5.g3 Bg4 in my opinion is not 6.Nbd2 but rather 6.Bg2 Qd7 7.0-0 0-0-0 8.Qb3, intending 9.Rd1 rounding up the d4-pawn, and if 8...Bh3?! then 9.e6 follows, intending 9...Bxe6 10.Ne5 and pressure mounts against c6 and b7.  Partly for this reason, I prefer 5...Bf5, whereupon in the above line, 8.Qb3 can be met by 8...a6 since 9.Rd1?? Na5 wins the queen.

As far as I'm aware 5.Nbd2 Bg4 6.g3 is likely to lead to the popular line 6...Qd7 7.Bg2 0-0-0 8.0-0 Bh3- I don't deny that this is theoretically better for White, but I am more than happy with Black's kingside hacking chances there.  5.Nbd2 Be6 6.g3 is likely to come to the same thing.

I don't trust 5.Nbd2 Bf5 any longer because of the first line above: 6.Nb3 Bb4+ 7.Bd2 Be7 (or 7...Qe7 8.a3, and after 8...Bxd2+ 9.Qxd2 the bishop on f5 is vulnerable to Qd2-f4) 8.Na5 (Justesen-Anderson, email 2011) and I don't think much of Black's compensation after 6...Qe7 7.Nbxd4, e.g. 7...0-0-0 8.e3 Nxe5 9.Bd2 (the bishop on f5 proves to be a target).

Therefore I think Black's best options are:

A) 5...Bg4 6.a3 (6.h3 Bxf3 7.Nxf3 Bb4+ 8.Bd2 Qe7, 6.Nb3 Bb4+ 7.Bd2 Qe7) 6...Qe7 7.h3 Bh5 (7...Be6 is well met by 8.b4).  However, White can get a queenside attack rolling here with 8.Qa4, e.g. 8...0-0-0 9.b4 Nxe5 10.Nxe5 Qxe5 11.Bb2.

B) 5...Be6 6.a3 (again 6.Nb3 Bb4+ 7.Bd2 Qe7 is an improved version for Black over 5...Bf5 6.Nb3) 6...Nge7 7.Nb3 Nf5.  White has the better chances with accurate play (8.Qd3 and 8.g3 are among the most critical responses) but the position is complicated.

Looking at those positions after 5.....Nge7.  I would have to go with 5....f6.  I have found a few games from 50 years ago where black play 6......Qf6 and seems to get a decent game.  Opocensky played it once and it is called the Janowski variation.

After 1.d4 d5 2.c4 e5 3.dxe5 d4 4.Nf3 Nc6 5.Nbd2 I have used 5...Bg4 consistently with good results. Some of the theoretical lines used to dismiss this after 6.a3 Qe7 are old and not particularly convincing. 

I don't know, honestly, what the true verdict should be on this line, so all that I can offer is my practical experience: my last two games using this line for Black were comfortable draws against mid 2300 level opposition. After the last one (where I was probably winning but took a draw because it was getting late and I wanted to sleep), my opponent told me that he was now doubting Avrukh's analysis. However, since I don't own Avrukh's book, I can't confirm this.

Yes ! I do agree with you ! In fact, this game doesn't help me to find a final answer to my initial question : what to do against 5.Nbd2 ... ?
1) to accept this ending with 8.... KxQd8 even if i'm not totally convinced by the resulting situation.
2) to try 5.... Bf5, 5.... f6 or 5... Be6 ?

And now, a recent game with one of the best french woman player defending the white side ! At the moment of writting, she is playing the French National championship at Pau (France) from the 14th of august to the 24th of august 2012.
This game is quite strange ! White seems to hold a clear advantage without too many problems ... Black has won but i cannot understand why !
Some ideas ?!