A depth-first approach to improvement

This will be my training strategy going forward. Go over games with a database, and also try to analyze the middlegame more deeply. Probably this means setting up a board and a PC – board so that I can look at board without being tempted to see the answer on the screen first.

If there is an opening line that you don’t know, but want to play, my advice would be to play it. If you can’t figure out why it is bad, look, you aren’t dropping a piece by playing it so what the heck, play it. You can always look at an openings encyclopedia afterward, if one isn’t built into your software, but when you study deeply, you don’t need it so much for your own games as to examine other possible ways of playing that position/separate strategies.

Middle-game analysis probably needs to deepen and become more accurate, in order to make Expert strength. Tactics will be there one way or the other, offensively/defensively, that is basically unavoidable, but it is one more thing to examine correctly OTB, and just as importantly find!

Some games may not need quite as much work as others, but in any game, the missed tactics could stand to be studied and not merely “Tell me Fritz! tell me! tell me!” before trying to find it oneself (and letting the engine notify you that it’s there, perhaps the first move only). Unless there is this sort of commitment to each game, Expert strength is probably going to be elusive or unmaintainable. I’m saying particularly once the 1900+ stage has been reached, to be able to progress from there.

As far as openings, I am going to stick to the ones, variations, that I already use. If I am not sure what to play in some opening I don’t see often, fine, I’ll just play something and figure that part out later when I am going over the game and looking for ways to improve it. Basically, I am going to study what I play, and only perhaps a key main line in some opening that I don’t see much of, like Bb4 in the Scotch, where I was kind of stuck, having bad internet results – but that will be more limited. I am not going to try and study every line in the Sicilian, for example, because a move-order can negate a whole variation, and it can get depressing trying to find an advantage, if you look at it to much. If anything, that approach seems to lead to yet even more variety and risky lines, because you want to see something different after a while – which is not a bad thing, it’s simply an openings thing. And what I am saying is that, either way, good or bad, it’s going to be the middle-game which determines the result.

Another thing I figured out is that there is a reason we like to play our variations. I could pick up and have cursory knowledge of other variations, but when I study the main lines, sometimes it looks as if both sides are goofing-off, willy-nilly pointless attack/counterattack. If I played them, I would probably care more deeply, but there is a style element involved. If one of the responsibilities once a person reaches A level is to find a “style”, then so be it, style usually means openings choices. For example, I may play French Advanced, but there are still many lines within that one opening, if you let there be.

Side note: I may be moving to Colorado very shortly and not playing chess for a while – until I get a permanent job there. So this may be it for a while, but I will update eventually, when I get there and get my feet back in the ground. 🙂

Advertisements

4 thoughts on “A depth-first approach to improvement

  1. linuxguy – if you lost your way, using the same old map will not get you far, better ask somebody or get your GPS from the trunk. If you missed something during the game, chances that you or your opponent will see it during postmortem are low, this is my experience. The same when you come home. You can spend, of course, several hours trying to find out if you missed any 2-3 move tactics, but do we have that time? I think more productive way is when Fritz (Crafty) finds it quickly and then you work on it. It’s like when good, high qualified coach explains you new stuff or corrects your mistakes, then you proceed with practice.
    I think openings are always work in progress, updating often lines in your variations, periodically trying new ones and once in half-year/year thinking about new opening(s).

    Hey, don’t disappear, you can always leave a comment on my blog.

  2. hehe. Thanks! 🙂

    Crafty finds a lot of nice things after the game. so many combinations are almost intuitively arrived at/analyzed to that just following Crafty helps a lot. Yeah, it would eat up major time to try and say “Okay, something is here” and then try to find it without Crafty’s help.

    I just played a game and found the first move, sacked the exchange, but the second move was tricky, it was Bh4, and then his queen has no good retreat squares and he is forced to trade stuff.

    I was just talking with someone in Colorado and their internet was out, so it’ll be interesting what kind of internet service that I find over there.

  3. Col-o-ra-do-?-?-?-? What the heck does somebody thinks to find there???

    Anyway, i hope the move goes smooth so that the chess bloggers dont notice that you were gone for awhile.

    First analyse the game yourself, only then let your chess engine run over it. That way the variations the engine gives will sink easier to your brain because you have tried to understand the same positions beforehand.

  4. hehe. Thanks, Chesstiger.

    Snow! and rain! Try finding either of those in Los Angeles! Oh and cold air. I am hoping that there is a tech jobs market in Colorado Springs and not just Denver or I would have to relocate up north eventually.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s