Wednesday, June 11, 2008

Excelling at Chess - again

You know it's strange but I find I've been finding it next to impossible to give a crap either way about the great grading debate so today I've decided to write about something else and I'm returning to Jacob Aagaard's book Excelling at Chess.

I started off taking the piss out of it (never let it be said I won't try to hit a soft target for a cheap laugh when there's one available) but more recently had a look at the central thesis of the early part of Aagaard's book - the idea that it is a superior understanding of the positional fundamentals of the game that is most likely to secure us victories.

Rivas Pastor-Akopian, Leon 1995
White to move

Today's position is taken from a game cited by Aagaard in support of his claim.

How would you assess this position? Would you say

  1. White is winning
  2. White is clearly better
  3. White is slightly better
  4. Equal
  5. Black is slightly better
  6. Black is clearly better
  7. Black is winning

or, do you want to play the chess book author's universal 'get out of jail free' card, and cop out with 'unclear'?

I'll come back to what Aagaard has to say about this position at a later date. In the meantime, please indulge my curiousity by answering the following too ...

How did you come to your conclusion? Did you calculate specific lines or are there certain features in this position that suggested the answer to you?


ejh said...

So is anybody actually going to be the first to make a fool of themselves? (I'm not.)

Dave said...

Black looks a bit better to me, nice Bishops and Rooks and the strong Knight can easily be kicked. No variations checked.

Tom Chivers said...

Jonathan's shown me this position before. I recall saying after a minute or two that black was clearly better, and white ought jettison a pawn by moving the queen from the e-file to give himself the best fighting chance. I don't remember which square I plumped for, nor how I came to the conclusion.

ejh said...

My gut feeling is that Black's better, certainly.

Anonymous said...

Well i've heard of the black player so it's probably good for black.

In the London League with the time control i'd take white and play f4.


Anonymous said...

OK, it appears that just loses a piece, so Black is clearly better.


Jonathan B said...

OK. Black's better appears to be the consensus of opinion on the whole.

So how much better is Black?

Jussi said...

They seem to be equal

True, Black has connected rooks and the bishops, while white's knight can be booted. But black's bishops are buried because there are too many pawns, and it's not clear how his rooks can attack

An enterprising white could win this, if he played to get a free pawn on the e-file

Tom Chivers said...

Given the diversity of opinion it's obvious that Aagard chose this position very well!

Anonymous said...

In my opinion, White seems to be a little better taking into account the more static features. I don't think Black's control of the c file counts for much as the entry squares are covered. However, Black's bishop on b7 in combination with his pawn on d5 are not so good. BUT, White's knight on e5 is pinned and Black is threatening to play f6 so what does White do about that? If White solves the pin problem and then retreats his knight to d3 after Black's f6 then Black can play Ba6 and swap off his bad bishop.

Play might go 1 Rfe1 h5!? (to stop Ng4) 2 f4 f6 3 Qxh5 g6 (at a cursory glance, this looks better than fxe5) 4 Nxg6 (threatening Rxe7) Qf7 and then maybe 5 f5... I don't have time to analyse further so I'll say this in unclear.

Anyhow, I'm thinking the key point is: can White avoid having his knight swapped by Black's light-squared bishop? If he can then he might be better.

How's that? Cobblers?


ejh said...

I never calculate any lines in a game normally unless I think I'm losing, so I'm damned if I'm going to here....

Mike said...

Black is clearly better. (I recognise this position: I'm white and I'm losing.)

Mike G.

Jonathan B said...

I'll be coming back to this position next week.

In the meantime, I'd thought Black was a little better in this position but Aagaard says,

"If you cannot fully appreciate that Black is a lot better ... you might want to have a discussion wiht yourself and/or a friend in order to acquire a better understanding of the differences between the respective set-ups."

Tom Chivers said...

Angus - if 1. Rfe1 then I think just 1...f6 2.Ng4 Qxe2 3.Rxe2 Ba6 when all black's pieces are active... That endgame was the sort of thing I was thinking about when I wanted to sacrifice the pawn I think.

Anonymous said...

Now the author's opinion has been revealed, you might want to check the evaluations of a chess engine. At no particular depth Rybka suggests White's next as Qd2 and considers Black slightly better (-0.60). That this loses a pawn to Bxe5 doesn't bother it as Black is left with an IQP. In the IQP continuation,it considers White as worse but defensible (-0.38)

If you have the whole game, Rybka prefers 17 Ne5, hitting the rook, as equal instead of 17 cxd5.


Jonathan B said...

I'm not sure the computer's assessment is correct here. To put it another way, I rather think Aagaard's assessment - even back then when he was 2300 odd and not British Champion - is more likely to be accurate than mine.

The whole game will be appearing in the next post. Aagaard is very critical of cxd5.

Tom Chivers said...

If I remember rightly from our conversation Jonathan, Aagard doesn't consider the pawn sacrifice. There are three possible reasons for this. 1. He didn't think of it. 2. He did think of it, but it's a distraction from his point. 3. He did think of it, but all it does anyway is confirm his point - if white's best is to sacrifice a pawn, then black is clearly better - so why bother to mention it.

The position after the pawn sacrifice is one where white will draw a certain % of the time, and black will win a certain % of the time. A computer assessment in fractions of pawns is not especially meaningful. The real question is what are the two %s above. Having computer programmes play one another in this position a number of times would be a better way to get a grasp on the chances it offers.

Glenn Wilson said...

I was looking for some fantasy line like:
1. f4 f6 2. Qh5 g6 3. Nxg6 hxg6 4. Qxg6+ Qg7 5. QxBd6. But it is just a fantasy as the pawn at f6 is in the way and Qxd6 is not a possible move. :(

If you think it appropriate I'd appreciate a link to my chess blog Houston Chess where I've been blogging about Chess in Houston (and the world) since 2003. Also, please consider checking out and linking to ChessFlash -- a new Flash Based Chess Viewer than can be embedded in your blog. Thanks.