To find your blunders, sort your moves by how much they cost, take the worst three or four, and for each one work out whether you missed the move entirely or misjudged it. Then save the position before the blunder and solve it again a week later, because repetition is what builds recognition.
Every player knows they blunder. Far fewer know which kind of blunder they make, and that is the only version of the question that can be fixed.
Sort by cost, not by order
A move list in playing order hides the important information. Sort by how much each move cost and the worst few jump out. In most games there are only two or three that really matter, and the rest is noise.
Cost is measured in centipawn loss: the difference between the evaluation after the best move and after the move you played. A move costing a hundred centipawns has thrown away a pawn.
Classify the cause, not just the move
Now the part that actually changes behaviour. For each of your worst moves, decide which of these it was.
| Cause | Tell-tale sign | The fix |
|---|---|---|
| A missed check or capture | The tactic was one move deep | Scan forcing moves before every move |
| A loose piece | Something was undefended or overloaded | Check undefended pieces at the end of your move |
| A back-rank or mate threat | You were concentrating elsewhere | Ask what your opponent's last move attacked |
| A wrong plan | No tactical error, just a slow decline | Study typical plans for the structure |
| Time pressure | The mistake is in the last few moves | Improve opening familiarity to bank time |
A five-second checklist before every move
- What did my opponent's last move threaten or attack?
- Are any of my pieces or pawns undefended?
- Does my intended move leave a check, a capture or a mate available in reply?
- If I play it, what is my opponent's most forcing answer?
Four questions, a few seconds, and it catches exactly the class of mistake that shows up as a blunder in a review. The point is not to calculate deeply, it is to make sure you looked.
Build puzzles from your own games
Generic tactics puzzles teach you tactics. Puzzles made from your own blunders teach you the specific mistakes you actually make. The method is straightforward: for each blunder, save the position one move before it, then solve that position again a week later without the answer in front of you.
- Save the position before your move, not after it.
- Note what you had considered at the time, if you remember.
- Return to it after several days, not the same evening.
- Mark the ones you get wrong again, because those are the durable weaknesses.
Look for patterns across games, not within one
A single game tells you about that game. Ten games tell you about you. If most of your blunders come after move thirty, that is a time or concentration problem rather than a knowledge problem. If they cluster in one opening, that is a theory problem. If they are mostly one-move tactics, that is a habit problem.