← Back home

About PatternChess

Chess training glossary

Blunder
A move that sharply worsens your position. PatternChess flags a move as a blunder when it loses roughly 15% or more of your winning chances, measured with the Lichess winning-chances model rather than raw engine centipawns.
Woodpecker Method
A training technique from the book by Axel Smith and Hans Tikkanen: solve the same set of tactical positions over and over, each pass faster than the last, until the patterns become instant recognition instead of slow calculation.
Spaced repetition
Reviewing the same material at widening intervals so it moves into long-term memory instead of fading. PatternChess reschedules each blunder you drill on an expanding ladder of days.
Winning chances
A 0–100% estimate of practical winning chances derived from the engine evaluation using the Lichess formula. It captures how costly a move really was better than centipawns, because the difference between +2 and +4 matters far less than the difference between 0 and +2.

Why I built this

PatternChess mascot

I built PatternChess because I wanted to test an insight I had from my own training. I started studying chess seriously as an adult, and the only method that got me results was spaced repetition. I'd print out a set of puzzles and review them every night until I could solve a set of 200-300 puzzles in one sitting.

The issue was, these positions taught me how to win. They didn't teach me how to stop losing. Without an expensive coach, no one could tell me exactly where my blindspots were, and I had no way of training the exact positions I was losing in. As a professional software engineer, I thought I might finally be able to come up with a solution. So I built one.