The Ragozin variation of the Queen's Gambit starts with 1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Nf3 Bb4:

[StartPly "8"]
[FEN ""]

1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Nf3 Bb4

It seems that the Bb4 move is only sound in that position, and not if white played 4.Bg5 instead of 4.Nf3.

[StartPly "8"]
[FEN ""]

1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 (4.Nf3 Bb4) Bb4

This line is, in my experience, not even mentioned in common opening books, as if it were "obviously" wrong... I'm not sure why. If this is bad for black, which would be white's best answer?

4 Answers 4


I'm not an expert but my guess is black doesn't want white to develop his g1 knight to e2 in a variation he commits his bishop to b4.

For example after 1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Nf3 Bb4 5.Bg5 h6 6.Bxf6 Qxf6 7.e3, wouldn't you be more comfortable with your f3 knight on e2? You can get that in the other variation!

1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 Bb4!? 5.e3 keeps the g1 knight flexible. Now if 5...h6 6.Bxf6 Qxf6 7.Nge2 and white slightly improved his position. That is if you agree the knight is better on e2. There are more than hundred master games played in my database with 4.Bg5 Bb4 so I don't believe it's dubious, just slightly inferior.

  • Sounds reasonable. What bugs me is that opening books do not (to my knowledge) say a word about such a natural alternative. I would like to find some material, either from theory or from some games.
    – leonbloy
    Jul 10, 2014 at 19:09

Playing Bb4 before White puts his knight on f3 in the QGD isn't bad, but it can be uncomfortable for Black. Let's look at some lines:

[StartPly "10"]
[FEN ""]
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.a3 Bxc3+ 5.bxc3 d5

This is the Nimzo-Indian Sämich with an early d5 which is known to be comfortable for White. He has long term chances associated with exchanging on d5, pushing f3 and e4 and going for a king side attack as in the classic game Botvinnik–Capablanca, 1938 (from the AVRO tournament):

[FEN ""]
[Event "AVRO"]
[Site "Rotterdam, NED"]
[Date "1938.11.22"]
[EventDate "1938.11.06"]
[Round "11"]
[Result "1-0"]
[White "Mikhail Botvinnik"]
[Black "Jose Raul Capablanca"]
[ECO "E40"]

1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5
exd5 8.Bd3 O-O 9.Ne2 b6 10.O-O Ba6 11.Bxa6 Nxa6 12.Bb2 Qd7
13.a4 Rfe8 14.Qd3 c4 15.Qc2 Nb8 16.Rae1 Nc6 17.Ng3 Na5 18.f3
Nb3 19.e4 Qxa4 20.e5 Nd7 21.Qf2 g6 22.f4 f5 23.exf6 Nxf6 24.f5
Rxe1 25.Rxe1 Re8 26.Re6 Rxe6 27.fxe6 Kg7 28.Qf4 Qe8 29.Qe5 Qe7
30.Ba3 Qxa3 31.Nh5+ gxh5 32.Qg5+ Kf8 33.Qxf6+ Kg8 34.e7 Qc1+
35.Kf2 Qc2+ 36.Kg3 Qd3+ 37.Kh4 Qe4+ 38.Kxh5 Qe2+ 39.Kh4 Qe4+
40.g4 Qe1+ 41.Kh5 1-0

So playing Bb4 before White puts a knight on f3 gives White an extra option which Black probably doesn't want to do.

PS: You're unlikely to find the move order 1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Nf3 in a high level game since 4.cxd5 followed by Bg5+e3+Bd3+Nge2 and eventually castles and f3+e4 with chances for a king side attack is great for White who scores 60% or so in that system (IIRC).


After Bb4 white can play Qa4 forcing Nc6: here you can see a difference emerging. With the B in g5 you can play 5.Qa4 Nc6 6.e3 and on 6... h6 retreat in h4 keeping pressure. While in the Ragozin this seems less effective since after 5.Qa4 Nc6 6.e3 locks the B inside, while after 6.Bg5 h6 7.Bh4 dxc4 becomes interesting, hence 7.Bxf6 is seen played, where black seems fine. Nothing final of course, still a possible explanation.


If White plays his B to g5, the best place for Black's black squared bishop is e7, according to Capablanca (Chess Fundamentals), breaking the pin on the Knight. This move may reveal White's B as being overextended, and allow Black to simplify with e.g. Ne4 or Nd5, bringing about an early equality.

The Bb4 variation is "bad" against Nc3 because logically, Black should take the knight and double White's pawns on the c file. But doubled c pawns are good for White, because he can capture Black's d pawn with his c4 pawn, then have an extra "center" pawn at c3 (formerly on b2) supporting his d pawn.

If Black wants to play Bb4, he should refrain from advancing his d pawn to d5 so that White cannot exchange his c4 pawn for it.

Source: Fred Reinfield, "Attack and Counterattack in Chess."

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.