How to find recurring chess opening mistakes in your own games
Use your Chess.com or Lichess history to spot opening choices you repeat and underperform, without treating a single engine label as a diagnosis.
A recurring opening mistake is a choice you face more than once and handle worse than your usual move from the same position. That is a different question from whether Stockfish prefers another move in one isolated game.
Start from a position you actually reach
Import a recent block of standard games — enough that the same opening trees appear, not so many that a rating climb from two years ago drowns the current habit. Chessigma reads public Chess.com and Lichess histories; it does not need your password.
Replay your own moves, not a book line. The useful comparison is the move you usually play in a position versus the move you chose this time, and how those two choices have scored for you.
- Compare the same colour only
- Stop comparing once the move sequences diverge
- Ignore one-off experiments until both moves have a real sample
What a useful finding looks like
A finding worth keeping is narrow: same position, two candidate moves, and a score gap large enough that a couple of flukes cannot explain it. “You blundered on move 18” is not that. “As Black in this Italian, you score 25% with Qb3 across four games and 64% with the castling move you usually play” is.
Silence is a valid result. Many games contain no repeated departure. Forcing a weakness onto a thin sample is how chess blogs invent problems you do not have.
What this method cannot tell you
Personal-history analysis does not calculate centipawn loss or name a blunder. Opponent strength, a rating change, and a short sample can all move the numbers. Treat a finding as a review prompt: look at those games together and decide whether the habit is still the one you want.
Frequently asked questions
Is this the same as a Chess.com game review?
No. A game review grades each move against an engine. This looks at whether a choice you repeat has scored worse than the move you usually make from the same position.
How many games do I need?
A few dozen can show broad colour or time-control splits. Repeated opening decisions usually need more. Imports here cap at 200 standard games so the newest habits stay visible.