One Zillion Moves

The 2006 United States Chess Championship is being played through Sunday at the McMillin Event Center, 2875 Dewey Road, San Diego. A field of 64 is on hand. The youngest entrant and last to qualify is Elliott Liu of Encinitas, age 16.

I have him on the phone. "How was Friday's game?"

Liu says, "Pretty cool. He [Gregory Kaidanov, 46, Ukrainian-born grandmaster] is number three in the nation. I was nervous because he's someone everybody knows, and he's won many tournaments. My strategy was to give him a fight, keep the game very simple, and make sure he didn't have any chances."

Sounds good. "Walk me through the game."

"Let's start at the beginning," Liu says. "I spent a lot of time preparing for him. The problem with these strong grandmasters is they play different moves in order to avoid this preparation. So, in the database, [Kaidanov] played several lines, and I had to prepare for all of them. Of course, at the game, he plays a line he's never played before -- a really rare line, it's only been played a few times. So, early on, I was nervous knowing he knew what he was doing and I didn't.

"The line he used was played last year between two of the world's best grandmasters, but I didn't remember what happened. I thought for 40 minutes on my next move. It was probably five moves after that I got a feeling where the game was going. I wasn't losing. My position looked okay; I realized I could probably hold as long as I didn't mess up somewhere. And then I traded everything off and it simplified very nicely and was relatively easy to draw."

"When you say you thought about a move for 40 minutes, what do you think about?"

"With chess, it's not just making a move." Liu says. "Not only do you have to calculate and foresee positions 12 to 24 moves ahead, but you also have to decipher what he's doing; you have to calculate his plans. All the lines have branches and branches, and you have to keep all that in your head and evaluate the final position of all those lines. It takes a surprisingly long amount of time to make a good move.

"Against a grandmaster, in order to draw, you can't make a mistake; your game, literally, has to be perfect. A lot of people don't understand that in order to lose a chess game, you don't have to lose pawns and pieces. If there's a weak square, just one, we're all good enough to use that square and win the game based on that one weakness. In order to draw one of these guys, you have to be perfect throughout the game."

"How many openings, how many lines are there?" I don't actually expect to hear a specific number.

Liu laughs, "I have no idea. People think, 'An opening is his first move.' No, an opening is one line, the line can go to move 25. In those 25 moves there can be 20 variations of that same line. The tree branch goes on, almost to infinity. No one knows all the lines. You can't know them all. You have to specialize. So, for example, my opponent, against my first move, only plays pawn to E5. His whole life, he's only played that. But, after that he uses several lines and other variations depending on what his opponent does."

"How do you prepare for that?" My usual solution, spend more money, does not, at first glance, appear to apply.

"You're going to prepare for the individual guy, but you know he's going to change his stuff around because he knows you're preparing for him. There's a lot of reverse psychology going on. Mainly, you're retooling your own stuff. There is only so much you can do to prepare for your opponent because he's not going to play into something that's already in his database."

I ask, "When you're changing your stuff around, are you trying to create something brand new or are you trying to bring back to life something that's obscure, a move somebody made 100 years ago?"

"It depends, but I will definitely change something," Liu says. "I won't play directly into what he's preparing. I know what's in my database, I can work around that. I'll either have an improvement from last time or I'll pick a line that's very rare or maybe even use a computer move. There are chess computers now. Just to make sure your new move isn't bad, you always plug it into the computer and see what it says. Sometimes your computer can spit out a move that's actually better than what's ever been played."

Hmm. "You keep saying 'database'. How big is the chess database?"

Liu laughs again, "All I know is that the first game in the database was played in 1605."