30 December, 2011

Playing against the Caro-Kann with Qf3

Its always good to play against someone who "sticks to their guns", consistently playing ( and believing ? ) in the same openings : you can look forward to a testing game.

In GM terms, a clear example must be Sveshnikov with his c3 against the Sicilian, and the same man with the Sicilian "Sveshnikov Variation" against 1.e4 !

I have no idea if my opponent is quite so principled as Sveshnikov in his chess beliefs, but he does seem to stay with the Caro-Kann against 1.e4, so when facing him. it gives me a choice of sticking to main lines ( and probably his better preparation ) or going with something different.

In this game, I chose something different ( after reading about it on the brooklyn64 site ), and played 2.Nc3 and 3.Qf3 against his expected 1...c6 and d5 and ended up here a few moves later...

11. 0-0-0
What appealed to me about the variation in question, was that it didn't seem like the normal Caro-Kann style, plus it has a long-castled king, something I am used to in other openings that I play.

I had in mind a king-side pawn attack, with the likelihood that Black would attack on the queen-side. Who would be quickest and most effective, is the question usually posed in such situations ?

By adopting this sort of setup, which seems a little slow to start,  I gave Black an an opportunity to attack quickly, and after a5, b5 ( where I miss a very simple tactic to win a pawn ! ) I expected a queenside build up with Bb7 or Bd7 in response to my bishop retreat to d3, but instead am presented with Qd5 as a direct threat.
13. ...Qd5
I took quite some time here, believing it to be a key position.

There are a few options that I considered

14. Bg5 leaves d2 free for the King in the event of Qxa2 and Qa1+, and White doesn't lose a pawn as there is Qxc6, although the threat of Bxf6 makes it interesting

14.Nc3 provokes the Queen exchange and leaves White with doubled pawns. One suggestion is 14...Qxf3 15.gxf3 b4 16. Ne4 Bb7 ( or Nd5 )

14.Kb1 is a sensible and solid move, protects the a-pawn, and probably avoids the complications of the Queen-side attack that actually followed. This, like Bg5, was also an option on subsequent moves. A possible line is 14...Qxf4 15. gxf3 Ng5 16. Rg1 Rd8 ( or c5 )

I eventually chose 14.Qxd5, objectively the best move for White, but somehow I completely ignored the reply of Nxd5, instead expecting cxd5 or exd5. However, once the knight was on d5, I felt ( possibly incorrectly in hindsight) that it exerted a powerful influence in the attack on the queenside, combined with  Be7 and the advance of the a- and b-pawns.

Psychologically this knight, and it's placement, proved a huge influence on the game, with this only disappearing when Black refused to exchange it later !

Following 14...Nxd5 the Black Queenside attack followed, but with White's a3, c3, and Be4, even though Black can re-organise his pieces it seems that White has enough to stop the attack.

Here is the position a few moves after after 18.g4

18. g4
I had played 17.Be4 with the intention of exchanging the Bishop for Knight, or at worst pushing the knight away, which is what happened, 17...Nb6.

The plan now was to gain space on the king-side and advance the pawns. Although Nc4 was seen, the intention was to meet it with Be1 ( to hold the defence on c3/b4 ) and have the Bishops co-ordinate against the queenside. I also needed to play f3 to allow the dark-squared bishop to escape after Nc4. Having the e1-h4 diagonal free for the bishop allows it to target the queenside again.

Play proceeded as expected 18. g4 Nc4 19. Be1 Rfd8 20. Bd3 Na5 with Black doggedly hanging onto his knight as I attacked it. I thought it had done its work and that 20... c5 21. Bxc4 bxc4 22. f3 Bc6 was better.

The next stage of the game saw Black continue with the attack on the Queenside with 21. f3 c5 22. Be4 Nb3+ ( not the best. Perhaps Rac8 or Ra6, so there are still options for the Knight)  23. Kb1 Ra6 24. Bf2 b4.

The diagonal a3-e6 is now blocked so the dark-squared bishop is free of it's defensive role and Black's attack stalls as 24...b4, instead of c5, allows the tempo-gaining Bd3 and encourages Blacks mistake of 25...Rb6?, the worst place to put the Rook !

I don't miss the combination that puts me a pawn up 26. dxc5 Bxc5 27. Bxc5 Nxc5 28. cxb4 and with 28... Nb3 instead of Nxd3 removing the strong and well-placed Bishop, a second mistake presents a winning combination...
30. Bxh7+ !
The discovered check gains the Rook,and following a later doubling of the rooks, Black resigned.

This was a tense and enjoyable game with plenty of good positions for in-game and post-game analysis.

I'm not an advocate of unorthodox openings, or weird sidelines, but I have to say that, at my level, playing a sideline like this that avoids mainline theory, but isn't too far of the beaten track allows both players to engage in playing chess rather than regurgitating moves.

In this case it led to an interesting position at move 13, which shows how different plans can be formulated and chosen, with wildly different  games resulting.

Chess is indeed a fascinating and rewarding game !


Anonymous said...

Thanks for posting this, I hadn't seen the 3. Qf3 variation before.

In general I dislike leading with ..Nd7 in these types of Caro-Kann positions, which are broadly similar to the main line ones. The move shuts in the light-squared bishop and White seems to get a nice initiative for free. An advantage of Qf3 is that it prevents the bishop from reaching f5 and g4 early on, so Black has to find some other way of getting it out. A disadvantage is that Black can sometimes use the White queen as a convenient target. In any case, the sideline seems fully playable.

lefthandsketch said...

Ha it's flattering to see you try out an idea you saw on brooklyn 64- thanks for the reference and the link! I got the idea from Asa Hoffmann.

I had luck with this for a while, but lately have been using the King's Indian Attack against the Caro Kann and am happy with my results in blitz.

It's really hard to knock a well prepared caro kann player off balance, but this system with Qf3 usually accomplishes that- some of the faces people have made when i play that move haha...love it.

Signalman said...

Thanks for the comment.

I've played a couple of games with this, and I would certainly say that it is playable for me.

I understand your comment about Nbd7, but I think its a fairly standard move in the Caro-Kann, and even in this variation ( I can check my database later ).

It wouldn't bother me too much, since I play Nbd7 in the Hanham Philidor, and am used to those sorts of positions.

Anonymous said...

..Nbd7 is indeed a standard move (including in this variation), but that doesn't mean I have to like it. :)

In this particular variation, the improvement is playing 5..Ndf6 instead of ..Ngf6, which as you demonstrated in your game leads to more problems for Black.

It's also worth noting that the 4..Nd7 variation of the main line is not played much any more. I've always played the Classical (4..Bf5) line, myself.