Sunday, September 11, 2005

BCC Sept. TNS Rd. 1 - Part 2

So here's some analysis on the position I posted on Friday. But first, thanks to CelticDeath (CD) and Tony for playing along in the comments. As you'll see they ultimately found much of what I'm planning to share today. Of course, there are still some problems left to solve.

Second, a disclaimer: I plan to share mostly what I was thinking about during the game and a few things my opponent and I looked at during the post-mortem. I haven't submitted the position to Fritz since I think this makes a more useful exercise without computer assistance. As a result, I'm sure their are flaws in some of the lines or assessments. Your improvements and opinions are welcome, but please don't berate me with stuff like "you patzer, Fritz claims White is better by .16 pawns in the final position." If you prefer closer-to-the-truth analysis, I suggest you read The Chess Mind (not that Dennis depends on computers for his analysis, but that he is a much better player than me and therefore his analysis is of higher quality).

Black to move

OK, let's start with the three retreating moves for the Knight. I instantly rejected 1...Nd8 since it breaks the communication between the rooks and virtually all the squares it might move to next are already occupied by other black pieces. I don't think I gave 1...Na7 much of thought either -- while it does attack the pawn on b4, after 2.a4 it looks pretty silly in the corner with nowhere to go. That leaves 1...Nb8 which does have a few positive attributes. For example: 1)if White maneuvers his c3-Knight to c5 Black might be able to play b6 since the Nb8 covers a6; 2) if White gets that same Knight to b6 Black might be able to extricate it with Nd7 (after moving the Rc8 of course); and 3) the Knight might be able to redeploy itself through d7 to b6 to c4. All that said, 1...Nb8 is a fairly passive move and I'm quite sure that if it is indeed the best move for Black then White clearly has some advantage in the diagrammed position.

Obviously not satisfied with any of those options, I focused my energies on 1...Na5. My immediate reaction was not pleasant -- "it just drops a pawn to 2.Nxd5." But as I looked deeper, I saw that there was much more to this particular variation. So let's see how the game progressed:

1...Na5!?

We'll call it interesting for now until we reach a firmer conclusion about its merits. Black threatens to invade at b3 and c4. If White opts to avoid the tactics beginning with 2.Nxd5 then the most plausible continuation is something like 2.Qd3 Nc4 3.a4 -- during the game I assessed this position as slightly better for Black (and certainly no worse).

2.Nxd5

At first blush this certainly looks like it wins a pawn. However, from the Black side I was able to generate an alternative perspective -- for a pawn I was going to be able turn my bishop which was functioning as a pawn back into a bishop. So in some sense I was giving up two pawns for a bishop, a trade that makes sense most of the time. Of course, the tactics needed to work also.

2...Bxd5 3.Qxa5 Bxf3 4.gxf3

White is still up a pawn, but this forcing sequence has left his kingside pawn structure a shambles and Black with the initiative.

4...Qg5+

I'm fairly certain this is better than CD's suggestion of Qh4 since White's response is forced, leaving him no time to get his Queen back into the game.

5.Kh1 Qf4

This leads to the next critical position (which Tony also reached in his analysis). Black is attacking a couple of White pawns and there are perpetual check opportunities in the air.

After 5...Qf4

Questions:
  1. What is White's best move at this point (include the expected continuation)?
  2. Given your choice, what is your assessment of the position in the second diagram?
No bonus question this time. Both CD and Tony answered correctly that this game started as a French Defense - Advance Variation.

Again, no computers please.

No comments: