Another Lesson with Dan --- a few elements of positional analysis
Jul 11, 2013, 8:23 PM 2
More travails of a (wanna-be) improving chess player...
I just finished a lesson with Dan and, wow, was that fun! I never know what we're going to actually work on, but Dan always finds something interesting and pertinent to point out.
This time I came prepared with several questions, from the general to specific. But the most interesting thing that came up was talking about a misunderstanding (or mis-application of a general principle) a specific type of position.
Question 1 (Opening Repetoire):
I didn't know how to word this question so I made it so weakly general that Dan had an obvious answer --- and then we got to the heart of the issue.
Me: Should I think about changing my opening repetoire at some point.
Dan: At some point? Of course!
Me: Well, I don't want to spend a lot of time currently learning a lot of new opening theory.
Dan then pointed out that Mike Leahy (who created Bookup which is now Chess Opening Wizard) says that the typical 1600 rated players knows 60-100 opening nodes (a node would be a unique position in the tree of variations). Dan said that a page of MCO probably has a couple of hundred nodes. Ergo, an intermediate player knows very little of opening theory. At that level (and my level) there isn't much need to know more that a few moves and apply basic principles.
In other words (mine, not Dan's), Jeff you can quit being a wimp and go ahead and try other openings. And, I readily admit that I know that book knowledge isn't winning or losing games at my level. So Dan's take on this is quite reassuring.
Question 2 (Candidate Moves):
After my De Groot exercise in the previous lesson, I realized that I had some serious flaws in my thought process. And so I asked Dan a fairly general question about how to be better at generating a list of candidate moves and how identify the likely moves of my opponents (I seem to overlook sometimes what my opponent plays). To this Dan suggested that looking at checks, captures and threats is always the best place to start. Often times, you don't have to evaluate every possible response by your opponent because they are merely a positional move ("A grandmaster might think: 'he can play Rb1, but that's no big deal'"). But you need to try to identify your plan and your opponents plan. That points you to the candidate moves.
Question 3 (Computer Evaluation of Specific Positions from one of my recent games).
I then asked Dan to look at one of my recent games where I was trying to understand why the computer evaluation differed significantly from mine.
(a) In this position I played 8. e5 which Fritz rated must worse than it's suggested move(s). I told Dan: "But I thought this pawn push e4-e5 is quite common in the King's Indian Attack". He then took me through the start of a classic game where Fischer used the KIA and made this kind of pawn push. "What's the difference between the position in Fischer's game and your game?", he asked. Turns out there were lots of differences --- completely different pawn structure being one of them. Dan: "You are making the mistake of trying to apply an idea that works in one variation to a completely different variation." Point taken!
(b) Two moves later, I am in this position and I chose the completely natural (and almost automatic) move 10. c3
The computer prefered 10. dxc5, which I thought at the time was completely unplayable.
Dan: "Why is it unplayable?"
Me: "Because, I'm going to have trouble holding either the c- or e-pawn and my pawn structure is wrecked"
Dan: "How is black going to get the e-pawn?"
Looking at the position, I had to admit that black was hard pressed to pressure the e-pawn. And Dan pointed out that even if black manages to pick off the c-pawn, it would take time. Time that I would use in developing.
Dan: "And how is your pawn structure wrecked? How many isolated pawns do you have?"
Me: "errrrr... none"
Dan: "That's right. Your pawn structure is fine and you're up a pawn. Sounds playable!"
Dan has this annoying habit of being correct
(c) Two moves after that, I choose to play 12. Nh4
Dan: "Why would you play that?"
Me: "I wanted to get the bishop pair"
Dan: "Well, he's just going to move the bishop and your knight is in a poor spot, but his bishop will be fine. He can just play Bd3."
Me: "Well, I anticipated him playing Bd7."
Dan: "No, that would be a bad spot to place the bishop. Bd3 is good, and the next best spots are Be6 and Bc8."
And now, this is where we go spinning off into something that is just fascinating for me. I complain that Be6 looks completely wrong to me. Why? Because it blocks the e-pawn.
Dan comments that beginner are often given a general principle to not block pawns with their bishops. But, as you get better you begin to understand that in this case there is nothing wrong with that.
Dan then constructed the following position (suppose that a game begins with 1. e4 e5 2. Nf3 Bd6) to illustrate the difference between a case where this beginner's principle is applicable and to the position in my game where it isn't apt.
Dan asked my what was wrong with black playing the bishop to d6. There are 3 things wrong:
(1) Why develop pieces in the opening? To gain mobility. But the bishop gains no mobility by moving to d6 --- it can still move to 5 squares, the same as from f8.
(2) The bishop on d6 blocks the d-pawn from advancing, but d5 is a break move for black. Parking the bishop here prevents black from playing an important break move.
(3) Blocking the d-pawn also prevents the other bishop from moving out along the c8-h3 diagonal, forcing black to fianchetto that bishop. So, Bd6 removes options from black as to how to develop another piece. And that makes it possible for white to anticipate and take advantage of.
Then we went back to the position in my game:
Dan: "Now think back to those 3 issues that made Bd6 a terrible opening move for black. How does that compare with this position.
First, as to mobility, granted the bishop on e6 has limited mobility, except that it isn't required to guard the d-pawn in this case. So it can move at any time.
Second, black has already played one break move: c5. What is the other break move? Answer: f6. But this bishop isn't blocking that break move.
Third, does this move block the other bishop in? Not at all, the other bishop is already finachettoed and isn't constricted at all by the bishop on e6.
So, these two positions highlight why there can't be a general rule to never block your pawns with bishops. And Dan found a way to highlight exactly what makes one case bad and the other completely acceptable.
And then Dan, in his typical fashion, imparts some more wisdom. He asked me where he picked up this knowledge. I hummed and hawed a bit, and then the answer was simply: from hanging around good players and discussing games with them. You know how Dan encourages players to go over their games with their opponents afterwards? There's a reason for that!
Thanks Dan for another enjoyable lesson!