CHESS
How I play, according to the data.
943 rapid and blitz games on Lichess since 2025, pulled through the public API and read for patterns. Below: what the numbers say about my style, and a bot that plays my openings at my rating.
STYLE PROFILE
As white I open 1.e4 in 81.3% of games. The main road is the Italian Game: 92 games, 59.2% score, my best weapon by a distance. Most of the rest is 1.g3 (10.7%), a slow fianchetto setup for days when I want a position, not a theory duel.
As black I answer 1.e4 with 1…e5 53.1% of the time: classical, symmetrical, no tricks. Against 1.d4 the numbers drop: the Queen's Pawn Game scores 40.1% over 71 games. If the data points to one piece of homework, it is that.
62.4% of my games run into an endgame. My score in games decided in the middlegame is 54.2%, in the endgame 50.8%. I know how to build an edge; converting it is where points leak. Average game: 33 moves, and 29.2% go past move 40.
The clock is on my side: 40 wins on the opponent's flag against 21 losses on mine. Checkmate is the other way round: I delivered 136, I got mated 163 times. I finish fewer attacks than I should, and I do not resign lost positions early.
Over 370 games Lichess analysed, average accuracy is 75.7% with 3 blunders per game. Castling: 64.9% short, 13.4% long, 21.7% never. The rating went from 872 to 1581 since 2025, peaking at 1597. Blitz sits far below at 1181: I play better when I have time to think.
PLAY ME
A bot built from my games.
The first moves come from an opening book made of my own games, weighted by how often I played each line. When the book runs out, Stockfish takes over, limited to my rapid rating.