Zukertort Opening — Win

Thursday, October 8, 2026 · vs Kotomina · View game on lichess ↗

Twice this game you had the engine's favorite move sitting right there after a long think — most notably move 32, where 104 seconds still produced rook to f6 instead of the crushing f4 — so the fix isn't thinking longer, it's specifically scanning for forcing shots (checks, captures, pawn breaks) before settling on the tidy developing move.

• ›You took a long 38-second think here and came out with a6, but the engine wanted bishop to d5, centralizing the piece into the heart of the position. A6 doesn't do much — it let White start generating real counterplay for the first time since that early blunder, basically handing back a chunk of the advantage you'd been sitting on.
• ›Knight to e5 instead of knight to e7 was the pick here, and it cost a full point and a third on the eval. The good news: White immediately returned the favor with h3 instead of grabbing on e5, so this one basically canceled itself out — a mutual slip rather than something that got punished.
• ›White had just blundered with queen to f2 instead of h4, practically begging you to play f4 and crash through. You spent a full 104 seconds thinking here — so it wasn't a speed thing — but came out with rook to f6 instead, letting a golden punishing shot slip by. Worth noting for next time: long thinks only help if you're searching for the active, forcing idea, not just the safest-looking one.
• ›This one went the other way — only 7 seconds on rook to f7, when c6 was the move the engine wanted. It's a smaller slip than the others, but it's the same theme: a quiet, natural-looking move instead of the one that actually presses the position.