NeverGiveUp wrote on 06/02/11 at 08:16:00:


Yes, I certainly go through all my games with the computer, although not just for the engine analysis. The engine analysis did show a trend in some positions where I thought I was losing and went for a risky line and ended up losing quickly, whereas the engine was saying, huh, what's the problem? So I now try to defend these "bad" positions more doggedly and try to assess when I have real counterplay better.

I looked at all my games with a decisive result (either way) to determine what the main cause of the result was. It was something like 10% Opening errors, 50% tactics/combination errors, 30% strategic outplaying, 10% endings. So was all my time spent studying openings/endings only contributing to 20% of my results? Difficult to assess, as say tactical/strategic problems might have been a consequence of choice of opening, but I haven't looked more deeply into that. Also I found I was good at spotting tactical opportunities for myself, but not those of my opponents.

I looked at my openings (by using the repertoire feature of Chessbase on myself) and found some variations were scoring really badly for me. So I swtiched over to an alternate variation, rather than learning new openings. I also looked at my openings based on results against players within 100 ELO of me, <100 ELO and >100 ELO, and found some interesting patterns of which of my lines worked best against stronger players, and which against weaker.

I also had a look at my endings, based on games with just King, 0-1 other pieces per side and any pawns. Not too surprisingly, over 50% of these were Rook endings. I looked at the postion where the ending began to assess the likely result, and compared to the actual result. My Rook endings too often went the wrong way (e.g. to many drawing a pawn up, and losing at equal pawns), but I did much better in all the other endings. So improving my Rook endings would be of most value to me - this lead to me buying Karsten Müller's Chess Endings DVD Vol 2 (and then subsequently his whole series).

Did this improve my game? Still not sure on that, but maybe the accumulation of small improvements adds up.

CraigEvans, shouldn't this thread be moved to General Chess?

And here's another thought ... I have played the Leningrad for a long time now, approx 10 year. I intend to get all (or most of) my games played with the Dutch and take them through the computer. Analysing my own games is something I have hardly done until now. Surely this should increase my understanding of the opening?

Do you guys have experience with taking your own games through the computer? How do you go about doing this and what are your experiences? Does it improve your game?

Thanks a lot for this good advice guys.

After having read what you said I've decided to stick with the Dutch but learn more sidelines and variations so I can mix it up that way, and also increase my understanding of the Dutch of a whole.

This may not be relevant, but I've seen this on the other end of the board. For instance, I've encountered predictable players that are very stubborn in their approach and they never bother to seek meaningful improvements or consult current theory. If there is a new deviation, it's often inferior and not so difficult to handle. Usually this is because they are not familiar with the main line theory and this limits their approach. It's harder to understand the deviations well if you avoid the main lines, because there is often a thematic logic in the principal moves that pervades the opening.

I like to play openings where I feel have a lot of freedom to adjust my approach depending on circumstances. This is usually feasible with high quality openings. It may be controversial to say this, but I am not so sure I consider the Leningrad to be a high quality opening. (Maybe moderate quality?) I have considered the Leningrad for tournament play, but the character of the opening makes me very uncomfortable. If you like the play and believe in the opening, you should definitely seek improvements and prepare multiple approaches to trip up your opponent's preparations. If you are stumped in a problem line, it's hard to give meaningful advice without knowing the specific line. For instance, as a junior I used to play Ward's recommendations in the 9.Bc4 Yugoslav, but eventually I realized they were becoming outdated as theory evolved. That would be a good reason for learning new lines in the Dragon and enriching my approach, but NOT for learning a different opening to the avoid the challenges posed by the 9.Bc4 Yugoslav. If you are not actively seeking improvements and expanding your base of knowledge in the openings you already play, you are likely to commit the same methodological errors in other openings. 

I'm all for learning new openings, but I think it should done in the interest of facing new challenges rather than avoiding old challenges. If you still believe in the quality of an opening, then you should try to defend it. If things start to feel stale and you want to learn another opening to expand your horizons, I think that's good as well. IMO learning a new opening is a long term decision that shouldn't be made on the basis of short term results.

knightmare wrote on 05/27/11 at 14:20:31:


Well said.  Fischer never had a problem with predictability.  Once you really make the Leningrad "your opening,"  you'll be glad when people play into it.      

NeverGiveUp wrote on 05/26/11 at 15:55:35:


Preparation of opponents is something usual above a certain level IMHO. I'm (more or less exclusively)  playing the French for more than 20 years, and most of my opponents are prepared for it. But I have played hundreds of games with "my" lines, so my experience and understanding of the positions is better usually. And that's what you should do  IMHO. Enhancing the understanding. What helps a lot additionally, is to have small deviations "in the pocket". I remember playing vs a young guy last year, who blitzed the first 18 moves or so in a Winawer, following some correspondence games that I played 2 years earlier. Then I deviated (dunno what he had prepared). The character of the position changed, and he had no idea what was going on. And that was it. First he got time trouble, and then lost the game.
The same btw. happened to me the other way round in a game in the French Tarrasch this year, when I had to accept that my opponent understood the position much better than me.

I agree. Firstly, I think you should fix the lines that don't work. But having options to choose from make your game more flexible and I would say that you get better because you would know how to play more types of positions. Staying within the Dutch makes sense to begin with since you wouldn't have to learn about the various anti-main line tries all over (as opposed to switching to e.g. 1 d4 d5). The Stonewall is an option, or even Leningrad stonewall whenever you get the oppportunity (see e.g. Kindermann's book).

In general, though, I want to learn different systems, in order to profit from it in the long term. This might cause a drop in performance in the short term, but it doesn't concern me so much. And I am not sure your results will be much worse if you try something else. You still know how to play the endgame and many middlegame positions, right? This doesn't change just because you switch to another opening system.