GMTonyKosten wrote on 11/01/07 at 00:09:18:


Let's stick with the plan, still play Rd1, if Bb3 sac the other rook on b3. White will have 2N vs R and Black's b pawns are weak. I prefer White. 

Not exaclty the piece up easy win though.

Thanks, some great suggestions!
Firstly, I thought 17. Bb2 Qxd3 18. Nde4 (or the other way around) would be simply met by 18...Rfd8 and I wasn't sure what the bishop was doing on b2, apart from defending a3, for instance 19. Qxd3 Rxd3 20. Na4 Bf5 looks at least equal for Black.
17. Nde4 Rfd8 18. Be3 Qxd3 19. Qa4 Qc4 20. Rfc1 b6!? does look quite promising for White, but I'm still not too sure where the white knights are going.
Instead, I thought that 18. b6 cxb6 19. Be3 Qxd3 20. Bxb6 would simply be met by 20...Qxd1 21. Rxd1 Rxd1+ 22. Rxd1 Bxa3 and I didn't look any further.
However, I never considered 19. Qc2 at all, but of course it makes perfect sense now, White leaves his bishop on c1 to defend the a-pawn, so 19...Qxd3 20. Qxd3 Rxd3 21. Rxb6 when 21...Bc8 looks pretty forced, then 22. Rd1 Rxd1+ 23. Nxd1 has to be better for White - this is the only line where Black has to make a real concession, with the bishop on c8 I would feel a lot happier as White.
So maybe it is a mistake, after all!

My two penn'orth:

Certainly trickier than all the English books I have read tend to make out.  In fact, I suspect that it is partly how natural this 'trap' is, that attracts some W players to this line!

I think that W is definitely better after some of your suggested improvements over the game, though, and I can't imagine many Black players going willingly into this.

A key decision for W is whether to play b5-b6 or not.  Playing it leaves Black with a choice of approaches - either capture as your opponent did, or play ...c6, preventing the b-file opening up.  Either way, W can target b7, either frontally (after ...cxb6) or via Nd2-e4, Nc3-a4, and then Nc5.

Personally, it looks more natural to me to hold that option in reserve though and just play - as you suggest - 18.Be3 Qxd3 19.Qa4!?  This has the advantage of keeping c7 as a target for the time being as well.

Yes, that's a good idea!

I was playing against the young Indian player Das yesterday, and the opening was a Reversed Dragon:
1 c4 e5 2 g3 Nf6 3 Bg2 d5 4 cxd5 Nxd5 5 Nc3 Nb6 6 Nf3 Nc6 7 0–0 Be7 8 a3 0–0 9 b4 Be6 10 Rb1 f6 11 d3 a5 12 b5 Nd4 13 Nd2 The mainline:
 

13...Nd5 A blunder, according to Motwani 14 Bxd5 The 'refutation' 14...Bxd5 15 e3 and the knight can't retreat because of e4. My opponent had a little think, and then coolly played 15...Be6 16 exd4 Qxd4
And here it was my turn to have a good think. Why is White winning? Black will win the d3-pawn and have two pawns for the piece, on top of that his light-squared bishop is extremely powerful and White's knights have no decent squares!
17 Nde4! Best 17...Rfd8 (17...f5? 18 Be3) 18 b6!?
18 Be3 is the alternative: 18...Qxd3 19 Qxd3 (19 Qa4!?) 19...Rxd3 20 Rfd1 Rxd1+ 21 Rxd1 Bxa3 22 Bc5 Bb4! Black has 3 pawns now, and in Vaulin,A (2510)-Karpeshov,O (2295)/Pavlodar 1991 White was happy to draw, despite the 200 point rating gap!
18...cxb6 19 Nb5?!
19 Be3 was my original intention, when 19...Qxd3 20 Rxb6 Qxd1 21 Rxd1 Rxd1+ 22 Nxd1 Bg4 23 Ndc3 Bxa3 24 Rxb7 Bf3 looked likely, but I started having my doubts about this position, Black can simply push his passed a-pawn, supported by his powerful bishops, and what is White doing? Still in retrospect this must be best.
19...Qxd3! 20 Qxd3 Rxd3 21 Nc7 Ba2! The saving resource! 22 Nxa8 Bxb1 23 Nxb6 Bc2!
and if anyone is better it is Black, and I eventually lost!! When I returned back to my computer to find the Suba,M - Garcia,S, Malaga, Spain 2001 game to see where I had gone wrong, I was astounded to see Black had resigned on move 15! Paul Motwani: "Black resigned, realising too late here that when his knight retreats, White wins material after 16 e4"
The only blunder in that game was Black's resignation!