I had mentioned in my last post that I had been doing well making databases that are opening specific. Shoemaker responded that he does not think of grouping "chess concepts" but rather makes databases for aesthetic value. When I want to see something of aesthetic value I pull out Kasparov's Greatest Games, Volume 2, I turn to a random page, and I try to sink in parts of the game. There's no way that I can begin to comprehend any of these games, but they are pleasing to look at. When I'm going through a database to work on my game, I will focus the database not on "chess concepts", but on specific opening lines. For instance with the Caro-Kann, I wanted to play a new line. I found that Kasparov had chosen predominantly the main line, I went over all the games in my database that Kasparov had played in that line, I would say I spent maybe 2 or 3 hours on creating the database and going over it a few times. Then I played it online a little bit. And then several months later I faced Grant Fleming and he played c6 and I decided to try it out:

1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Bf5 5.Ng3 Bg6 6.Nf3 Nf6 7.h4 h6 8.Ne5 Bh7 9.Bc4 e6 10.Qe2 Nd5 11.f4 Nd7 12.Bd2 Nxe5 13.fxe5 Qb6 14.0-0-0 0-0-0 15.Rhf1 Qxd4 16.Bg5 Qb6 17.Bxd8 Qxd8 18.Bxd5 cxd5 19.Rxf7 Bg6 20.Rf4 Be7 21.Qg4 Bxh4 22.Qxh4 Black Resigns 1-0

Now neither of us played very well, although we were both over 1800. But the first game against Parreira, in the Rough Day post is the only other time I have played this Caro-Kann main line in a tournament and I have been able to understand games in this line much better since going through this database. The most noticeable differential to me in between these games is the exchange of the bishop. It develops White's Queen for free and gets rid of the powerful albeit cornered black bishop, which has already moved 3 times and will now move a fourth. These are the things I look for when I study openings, it is impossible to play the middlegame without knowing what exchanges are beneficial and which are harmful. And it is true that there are other ways to evaluate this, and good players can find this information at the board. But where does such information come from. Study of endgames obviously helps in the later middlegame, but in the early middlegame oftentimes you can rely on empirical information, the games of very strong grandmasters in the same lines.

You can find them trying your idea and see the best opposition at the same time and I think that this is important. At least knowing what the best tries are, I oftentimes will dumb down my line because I don't think that my opponent will be able to keep up, but I usually know the game I'm giving up too. I think that in order to actually find the openings that I want to play I have to find the middlegame positions that I'm getting myself into and either get to know them or find an alternative. That's how I've built the repertoire that I've been playing and I have no holes. The English has been a catastrophe of late I will admit but I have a line. The Maroczy also I have several lines. I am closing holes before you can even find them, stacking up two or three alternatives in case we play a few times and I want to jump around, discuss a different topic. And if you find one, please believe you will never find me there again, and if we are going down the same path, walk slow and carry a very big stick.