Prepping for Philly

Mar 22, 2015, 8:18 PM |

In about ten days I'll be heading east to compete in the seven round Philadelphia Open for the second year in a row. Last year, it was the first time I had ever traveled out-of-state for a chess tournament, and I had a blast. Playing seven slow chess games in four days. Watching top U.S. players, including GM Gata Kamsky. Exploring the historic city in my off hours. What's not to love? 

Unfortunately, my chess results last year were a disappointment. I scored 3.5/7, even though I was one of the top seeds in the U1600 section. In fact, in one game I fell for an opening trap and was checkmated in just 8 moves!

Not this year. For the first time ever, I've decided to put considerable effort into prepping my openings for this tournament. I plan to play my three standards: Torre Attack as white, Caro-Kann as black vs e4, and QGD Tartakower as black vs d4.

Differently than in the past, for the past few weeks I've been memorizing dozens of lines in each of my openings, going 8-12 moves deep. The interesting thing I'm finding--especially in the Torre Attack and QGD--is that I didn't know nearly as much about these openings as I thought I did. I've read several opening books over the years on my openings, I've played them, I've analyzed them in post-mortems. But, there were so many opening ideas and plans that weren't in my toolkit. I had either never been exposed to them or I had simply forgotten about them after reading about them in a book.

Here's the thing. At least for me, there seem to be only two ways I thoroughly learn an opening. I need to either brute force memorize it or I need to play the opening in a game, mess it up, then look it up and figure out what I need to do better next time. The latter works well but could take years and there is no guarantee you'll cover important lines.

In order to memorize opening lines, I've been using a few key tools.


I use SCIDvsPC (the Mac version) as my chess game database. It's fantastic free open-source chess software. Over the past year, I've put together a nice assortment of game fragments that represent my repertoire. Most of this came from books or independent analysis with the Toga II engine, or recommendations from coaches and friends.

With SCIDvsPC I can slice and dice my repertoire. I can view it in a tree view. And I can look for discrepencies. For instance, when there is an identical position that is reached from two different lines, do I make the same move in that position? If not, my repertoire should be adjusted so that any given position should only have one correct move, regardless of how I got to the position. Doing this really helps with KISS (Keep it Simple, Stupid).

Anki is a spaced-repetition learning tool (originally designed for learning things like vocabulary) that is similar to electronic flashcards. I have been putting my opening positions into Anki (from my SCIDvsPC repertoire database) so that I can memorize positions over time using spaced repetition--a much more time efficient way to memorize data.

In this way I am focusing on learning the right move for a position, not the right sequence of moves for a variation. This makes much more sense to me. If I get to a position via an unstudied transposition, it should still trigger my memory.

Through the process of entering the positions into Anki and trying to memorize them, I have had to look deeply at positions that are very similar to other positions and figure out why, for example, I play Bxf6 in one and c3 in another. There must be a reason and figuring out the reason leads to better understanding overall. This has been the key to this process for me.

Perfect Chess Trainer
The final tool that I have been using is the opening trainer that is included in the Perfect Chess Trainer Android app. PCT's opening trainer allows me to run though each of my opening lines, in sequence. Using SCIDvsPC I exported a PGN file for each of my repertoires (white, black vs e4 and black vs d4). The file contains one opening variation per pgn game. After loading this file into PCT, I can go through the variations and PCT tells me which ones I make mistakes on and which ones I get right. It then advances me to the next line. Using PCT, I can run through my lines very quickly, while keeping track of the ones that I need to put more work into.

I've always been a skeptic of spending too much time on rote memorization of opening lines without trying to understand the underlying ideas. I am still a skeptic.

But, thorough memorization of opening lines with the associated understanding of the underlying ideas, I hope, will help elevate my game in the upcoming Philadelphia Open. I'll be moving up to the U1800 section this year, so I'll need every advantage I can get.