You'll remember from Saturday that we were looking at a line that leads to the position immediately above.
1.d4 Nf6 2.Nf3 e6 3.e3 c5 4.Bd3 Nc6 5.c3 b6 6.Nbd2 Qc7 7.O-O Be7
We were proposing to compare it with the line
1.d4 Nf6 2.Nf3 e6 3.e3 b6 4.e3 Bb7 5. O-O c5 6.Nbd2 Be7 7.c3 Nc6
given by Cox (and others) to see whether or not the delay in developing the c8-bishop is an improvement for Black. I think it might be, as it appears to gain a tempo in the most crucial line, beginning 8.e4.
Now it might surprise you that I don't propose to spend much time investigating the alternatives for White on the eighth move: 8.a3, 8.dxc5 or 8.Qe2. This is for a couple of reasons, the first being that this is a suggestion, an idea, not a comprehensive theoretical treatise (I am good enough to come up with ideas, but not, I think, to write theoretical treatises) and the second being that none of them look particularly threatening. Playing the Black side of these shouldn't be so hard - no, really: provided you ask yourself what White is trying to do and take steps to interfere with it. In most cases we can simply play ...Bb7 and transpose back into lines that Cox considers perfectly fine for Black without looking as threatening as the main line we are trying to improve upon.
So 8.dxc5 I can't be bothered with at all, it's not a try for advantage. 8.a3 prevents the ...Nb4 manouevre that we are going to play if White tries 8.e4 - but it gives us a move for nothing. We can play 8...d5 if we want - yes, I know we were trying to avoid this but White has wasted a move and we have e5 nicely covered. The only way White can try and justify it is with 9.b4 which threatens to drive the knight away from c6 and thereby claim e5 for the knight.
But in truth that's not all that scary - we can play 9...Bb7 transposing into a line Cox gives as "at least equal":
10.Bb2 c4 11.Bc2 b5 12.Re1 O-O 13.e4 a5 14.e5 Nd7 and White needs to explain what role he is expecting the b2-bishop to play in the future (Guimard-Sunye Neto, Porto Velho 1988). Perhaps 10.b5 is rather more to the point but after 10..Na5 11.Ne5 c4 12.Bc2 O-O
White still can't play e4 and while he is trying to organise this Black has plenty of time to organise play against the knight on e5 and/or with ...a6.
But for that matter Black could hold back the bishop and play 9...O-O or could permit the e4 idea after all: Cox (with ...Bb7 on the board and the queen still at d8) gives a game Kovacevic-Seirawan* which went 8....a5 9.e4 d5 10.e5 Nd7 11.Re1 Ba6. Provided we think the queen would go to c7, we are, again, a move up on this line. So Black has all sorts of options.
8.Qe2 is also less than frightening as White still isn't threatening 9.e4. Simply 8....O-O and now if 9.Re1 it still isn't threatened due to the ...cxd4, ...Nb4 and ...Ba6 idea so simply 9...Bb7, transposing back into Cox, is fine for Black.
8.e4, as mentioned at the end of the first part of this survey, is precisely the idea against which our move-order is aimed. In the normal line, with ...Bb7 instead of ...Qc7 played, the line continues:
8.e4 cxd4 9.cxd4 Nb4 10.Bb1 Ba6 11.Re1 Nd3 12.Bxd3 Bxd3
after which 13.d5!? is the move that seems most threatening. Now in our variation, we would have a slightly different position after the same moves 8-12 and now 13.d5:
which seems at first sight just to put Black a useful tempo (...Qc7) up. However, it's not so simple, because 13.d5 actually includes the threat of 14.d6, which after a capture on d6 wins a piece with a fork 15.e5. For this reason Black (in the Cox line) normally continues 13...Qc8 after which 14.d6 Bd8 is at least a little awkward for Black even though he may be fine after 15.Ne5 Bc2 and an unravelling process involving moves like ....b5.
It was in seeking to find an improvement on this that I hit on the idea of delaying ...Bb7, and it's worth mentioning that in extremis Black can always play, however eccentric it seems, 13...Qc8 here too! So, having a transposition available to him, he is definitely no worse off than before. However, it would be nice to actually make use of that extra tempo, and while the queen is on c7 it has the merit of preventing the White knight coming to e5 - as it did in the line cited in the paragraph above.
However, it has the demerit of being vulnerable to the same d6 idea as came before, perhaps indeed strengthening it. So clearly either the queen or bishop must move, albeit Black could play 13...Bc2 before making the choice. The computer, reasonably enough, suggests 13....Bc5 which has the horrific threat of 14...Ng4 (not to mention 14...Bxf2+ since after 15.Kxf2 Ng4+ 16.Ke2 loses the queen). so it suggests further that White play 14.Nb3, allowing Black to win the e-pawn with 14...Bxe4.
Now this, to me, is the crucial position. Black is a pawn up, but is uncastled and the e4 bishop is loose. Meanwhile White has a dangerous rook on e1 and is close to completing his development. However, the d6 threat has been defused and indeed White is a bit short of centre pawns.
Has White got sufficient for the pawn? I'm not that interested in what computers have to say, unless they can offer a concrete line proving the case one way or another. Can they? Mine (Rybka) seems to think White has about half a pawn's worth of compensation, but that's no guide since my experience is that computers normally - though not always - underestimate compensation for material. Indeed it just backs up my gut feeling that this is exactly the sort of position where White might have enough for the material, but in the absence of concrete variations, I'm really not sure.
I'm much more interested in what readers have to say. What variations would you try? Do you think White has sufficient? Which side would you prefer to be on? If we get some decent feedback we can have a look at some of the ideas (and then conclude with a look at some move-order issues earlier on in the line). But first, I'd be grateful if you could ask yourself the questions above. What has White got? What can you suggest?
But I shall ask myself the question - would I like to defend the Black position against Robin Haldane?
[* = given simply as "Indonesia" without the year, which Chessbase Database Online reveals to be 1983.]
Post a Comment