I've been having some trouble against the Benko in a couple of recent games, and since it is said (paraphrasing a little here) that if you lose three consecutive games in the same opening, you should give up chess, I thought I'd try to make a thread about the subject to avoid that. Since at least one other person was interested, I thought I'd kick off a discussion.

Really, this is more about the Benko middlegame than the opening, since looking back I didn't play the opening that badly, so I'll put in a couple of different opening options that I found reasonable.

The main problem is that Black's plan is just so much clearer than White's and so I'm spending a lot of energy (and time, which is already a problem) without making noticeable progress (and then eventually giving the pawn back to reach an inferior endgame well behind on the clock). Mainly, I'm looking for insight into where White's pieces belong.

One final remark -- I'm reaching these positions from an early Nf3 move order. I know that 30% of you will think that this is terrible and the reason for all of my troubles. It's hard for me to avoid this without a very serious overhaul of my repertoire, though.

Variation I -- White Castles By Hand
1.d4 Nf6 2.Nf3 g6 3.c4 c5 4.d5 b5 5.cxb5 a6 6.bxa6 Bxa6 7.Nc3 Bg7 8.e4 Bxf1 9.Kxf1 d6 10.g3 0-0 11.Kg2 Nbd7

This all makes sense to me. You'll see from this (and the next line I'll mention) that I'm not comfortable with Black targeting d3, so I prefer to play this way, and just get rid of the bishops, or keep the pawn on e2. The question is: what now? Black has a few obvious plans (as I see it): put major pieces on a/b files, shuffle the knights around to put one on e5, and play e6. I, on the other hand, don't know whether White should play h3 (to prevent Ng4), wait for Ng4-e5 and play f4 (but doesn't that weaken the king and encourage f5?), and definitely don't know where any of the pieces belong.

Variation II -- White Sticks with e2
1.d4 Nf6 2.Nf3 g6 3.c4 c5 4.d5 b5 5.cxb5 a6 6.bxa6 Bxa6 7.Nc3 Bg7 8.g3 d6 9.Bg2 Nbd7 10.0-0 Nb6

Similar comments as to before, except here, Black works to make the most of White avoiding e4, but playing quickly against d5. Black can force e4 by playing something like Bb7, but that's something of a waste of time...I think.

Of course, playing 5.Qc2 lines or 6.b6 lines is possible. But, come on now, that can't really be better than taking the pawn, can it? And there may be more modern ways of meeting the gambit (but do they work with an early Nf3?) which I will be interested in learning from you.

Looking forward to hearing your thoughts!

One variation to look at is line 1, but with h3 instead of g3. You need one more move to get your King out of the way, but h3 is a much more useful move than g3. I'll get back to the thread when i have more time, playing something decent against the Benko is on my agenda for a long time now.

Great suggestion, thanks. This idea is really appealing intuitively. I played h3 in that line, so isn't even a waste of a tempo, and the king structure/pawn structure would be much healthier (especially if f4 is contemplated). I'll take a closer look!

Played a lot of stuff (now back at 1.e4 ) vs benko, felt most comfortable with b6-lines from Squeezing the Gambits. Georgiev recommends the move order with 4.Nf3 to prevent 4.cxb a6 5.b6 e6

As someone who used to play the black side fairly regularly, I'd be much more comfortable in position 1 than position 2, mainly because in those lines the white-squared bishop often gets in the way of the attacking ideas (it blocks the a-file on a6, if it goes to b7 it blocks the b-file & encourages e4, while if it goes to c4 at the wrong time white can simply play b3 [a useful move in many lines] with tempo).
What are you trying to avoid by playing with a 2.Nf3 move order rather than 2.c4?
If you're using the 2.Nf3 move order, there are also some old lines with Nf3-d2 before playing e4 that can be interesting (and often more solid than some of the 'castle by hand' lines) as you can play Nxf1 & castle normally later, though I must admit to not having played against this for a while to know if its still ok theoretically.
Another thing to consider (given your thoughts about the ease of playing the middlegame as black) is why you're accepting the gambit? Often declining the gambit can be more problematic for black, if not from a material perspective, then at least from an ease of playing perspective.

Hmm, interesting conclusion. That book looks pretty good based on a quick look at the intro. At least, he seems to be answering my question. Thanks for the reference!

Nice explanation. I don't think I considered that, really, since Black's bishop, on its own, usually looks like the more useful the two light-squared bishops.

Yeah, that's the third vote for not accepting the gambit (fifth if you include the authors of various books on the subject), and I guess I am now in a position to accept that judgment.

Thanks for those thoughts.

*

I guess my answer has to be the Budapest (what else?) but I can't really recall. I got into the habit because I play Nf3 lines anyway, and this doesn't give Black any additional options (assuming he already knows I play Nf3 lines). So my comment seems a little odd now. (I might have been thinking about the choice on move 3 between Nf3/Nc3, but that doesn't really apply to early-g6 lines...)

Okay, apologies for this late reply! As a practioner of the Benko since my younger days, I'm also looking for something reasonable as White too. I'll only comment mostly on variation I because this is a tabiya that occurs somewhat frequently and I've played the Black side of this many times. Theoretically, Black isn't in any trouble except you might look at a recent game with Caruana as Black when faced with 12. a4 where there might be room for improvement for both sides and it's not obvious that 12...Qa5 is Black's best. The main moves are 12. Re1 ( which seems to almost never occur now ) and 12. h3, so it's best if you look at recent games with 12. h3 or consider 12. a4. Variation II seems to be hard to play as White since Black scores quite well in practice.

You could decline the gambit as others have mentioned, so maybe replace variation II with your favorite flavor of the gambit declined. I've looked at a number of offbeat lines here too, but nothing very attractive.

1.d4 Nf6 2.Nf3 g6 3.c4 c5 4.d5 b5 5.cxb5 a6 6.bxa6 Bxa6 7.Nc3 Bg7 8.e4 Bxf1 9.Kxf1 d6

I do not know if fits this position for white to be playin g3 and Kg2. That seems to be using up alot of time for not a greatly improved King safety.

I seem to recall some play where White leaves his King in the centre - uses H file as his plan of attack - and also transferring his a file rook via a4 Ra3 pretty quickly.

Maybe my memory is shot but they are some of my thoughts on position 1.

Yo - there's no way to create an h-file attack in this position - even if White gets to play like 15 moves in a row, I don't even see how he can even get to threaten mate.

One idea I have tried against the Benko is to play 1.d4 Nf6 2.c4 c5 3.d5 b5 4.ab a6 5.ba g6 6.Nc3 Ba6 7.f4 With the idea of Nf3 and e4 and taking back on f1 with the Rook. - I don't think it is objectively one of the best lines but it forces Black to solve some unusual problems.

Completing development (activating the rook) is at least as important as king safety. And I'm not sure I believe that can be done via the h-file--in that case, if Black opens the center with e6, White is going to have a mess on his hands.

I still like Noir_Desir's suggestion of h3 rather than g3 as the way to make space; it's typically pretty useful. It'd be interesting to see whether that gets White anything in the lines, e.g., bigpooch mentioned. If I ever test that out, I'll update the thread, although even with the extra tempo, I don't think White's play is so obvious.

AFAIK the typical plan is h3, run the K to h2, play Rh1-e1-e2 and if necessary Rc2 to defend the queenside. Then, try to pursue whatever white's plan is in the Benko.

I have a vague recollection of 7.-Bg7 8.Nf3 Qa5 being a really good line against this. Maybe it later involved getting the queen on a6 and trading on e2 for a really good version of a Benko ending. (benko=one of the few gambits in which endgames are often pleasant for the one giving the gambit)