Friday, March 28, 2014


White to play
JMGB v J. Briggs, London League 19.03.2014

To play 12 Bg5 or not to play 12 Bg5? That is today's question.

Comfort zones. That is today’s theme.

A pair of rook endings. That is today’s bonus.


I’ve had this position on several occasions over the years, but I’ve never played 12 Bg5 before. The first time because I didn’t know what I was doing. The others because, even though I’d researched the line and knew that it was the book move, I just didn’t understand it.

Usually I played 12 Bd2 instead. I got good results with it (an advantage out of the opening every time and a score of 2.5/3) and I knew that GM Adgestein had played it at least once. You finish development, bring your queen’s rook to the e-file and play f2-f4. In a lot of lines you can bring your bishop to c3 and challenge Black on the long diagonal. What’s not to like?

After one game I mentioned to my opponent that I whilst I was aware that Bg5 was the approved move, it made no sense to me at all.  "Well," he said with the absolute certainty of somebody who was still in his teens and therefore knew everything, "obviously White’s swapping a bishop that’s doing nothing for Black’s best piece."

Obvious enough that even I had grasped that much. What I found harder to fathom was what came next. White pretty much has to go 13 f2-f4, I think.

Partly because it’s the sort of thing you want to do in such positions and partly because you don’t want Black to play ... g5 and ...f4 himself. Which is all very well, but now Black will play 13 ... exf4 opening the line for the bishop.

I couldn’t understand why White hadn’t just given Black a powerful unopposed piece? Doesn’t Black sometimes give up an exchange to get rid of a fianchettoed bishop’s opponent? For example as Azmai tried to do against Karpov last week (Azmai Could Play)? How could White’s strategy possibly work here?

I had another look. The line is covered extensively in Avrukh’s GM Repertoire Volume Two, not to mention Janjgava’s much older book on fianchetto set-ups against the Grunfeld and the KID.

There’s no shortage of information out there. Everyone seems to give 12 Bg5! Twenty years ago both Karpov and Kasparov had played it. I was obviously missing something, I just didn’t know what.

Eventually it dawned on me: playing this does give Black a strong bishop. It’s just that White gets a lot of other advantages - space, lead in development, none of Black’s other pieces being particular good - which more than compensate for this.

And yet for all the strengths of White’s position I simply couldn’t get over that bishop. OK, I knew, in theory that I could play around it. I just didn’t trust myself to do it without messing up for a whole game.

Another opportunity to play 12 Bg5 came around. I ducked it and stuck with 12 Bd2 again anyway.

Worth a punt

For a whole bunch of reasons I’ve been thinking about comfort zones a lot recently. In terms of chess that means taking a risk or two. Moving beyond what you normally do. Breaking the rules. Breaking your rules.

And 12 Bg5 does break the normal rules. You’re not supposed to give up bishop for knight. You’re supposed to challenge a fianchettoed king’s position with your own bishop. You don’t leave yourself with a bishop locked in behind central pawns and you certainly don’t give the enemy a free run on the dark squares.

Except some times you do. I knew that, I just didn’t believe it. Something else I knew: I never was going to believe it until I tried it in a game.

So next time around - which happens to have been a week ago at Golden Lane - I did.

And guess what?

Nothing terrible happened and nobody died.

BORP? Index


Anonymous said...

If you check the position with an engine, you find that the evaluations of the candidate moves, namely Bg5, Bd2, f3 , f4 and exf5 are pretty similar. It isn't then a seminal moment in the game. The suggestion would be to play a line and subsequent moves you are comfortable with. If it isn't the line suggested by some authors, so much the better, particularly if they condemn it as bad.

Bg5 has become popular in recent years, but is this the result of people knowing what's written in books, or judging the position for themselves?


Anonymous said...

If you're planning f4 soon then the long diagonal may well open for Black's g7-bishop. In that case you definitely want to preserve your own dark-squared bishop. So I prefer to delay developing the bishop and play f4 here.

ejh said...

I know I've had the position, but as I can't remember how many times, or exactly what I played, knowing this doesn't help. I think that on one occasion I may have played 12.Bg5 h6 13.Bd2 on the grounds that I was then going to attack the pawn on h6, but I doubt that this was very good. Undoubtedly I was concerned about apparently exchanging bishop for knight for no obvious reason.

It's possible that one reason to get rid of the bishop is that it doesn't have any obviously good square and isn't going to get one. Mind you I nearly always think this about my c1-bishop any time it goes anywhere but b2.

Jonathan B said...

@Roger: Yes I definitely think it’s better for amateur chessers to play what they’re most comfortable with rather than blindly follow what they know to be - or think is - 'correct'. What I’m also trying to get at today was the idea of having to play oneself into comfort in some lines/variations.

@anonymous: " If you're planning f4 soon then the long diagonal may well open for Black's g7-bishop. In that case you definitely want to preserve your own dark-squared bishop."

I fully understand this logic. Indeed i felt exactly the same thing. You could perhaps go on to explain how it is, then, that both Ks played Bg5-xf6 and only then played f2-f4.

@Justin: I suspect giving Black ... h6 for free is not a good idea because it helps with a pawn storm.