Hanging material
A piece or pawn was takeable, or a tactic won material immediately.
Blunder analysis
Find the Blunder Pattern You Repeat Most
A blunder count tells you how often. ChessLogix shows what kind, where it happens, and which positions to train first.
Free · Sorting, ranking and training your blunders costs nothing
- No Chess.com or Lichess password — a public username is all it needs.
- The learning platform is free — AI Move Explanation and Meta-Coach are an optional membership.
- Every recommendation comes from your own games and Stockfish analysis.
- Every large swing is sorted into a category, not just counted.
- Categories are priced in win probability and ranked, so one of them is the place to start.
- Each category links back to the positions in your own games where it happened.
Why "Fewer Blunders" Is Not a Plan
Every player already intends to blunder less. The instruction contains no information about what to do differently on Tuesday.
A blunder count tells you the frequency, but not the cause. The useful version is narrower: which kind of blunder, in which phase, at which time control. Hanging a piece in a sharp middlegame and mishandling a won rook ending are both blunders, and nothing about training one helps with the other.
Once your blunders are sorted this way, one kind almost always stands out from the rest — and that is the one worth your time first.
One Kind Rises to the Top
Every big swing is sorted by what actually went wrong on the board, ranked by what it cost you, and stacked up across a batch of games. The one at the top is where to start.
- When one kind leads by a clear margin, that is the one worth an hour this week.
- Error types are counted separately underneath: three missed tactics, one piece left hanging.
- Training Focus turns the top category into a set of exercises rather than a resolution.
What Gets Classified
Categories describe the mechanism visible on the board. A move that allowed a forced mate is its own case, and a swing with no identifiable tactical mechanism is reported as a positional error rather than pushed into one of these.
Missed tactic
A tactic existed for one side and the move played does not account for it. Detected from the engine lines before and after the move.
Missed defensive resource
A defence was available and was not played. Most common in positions that feel already lost.
Unsupported attack
An attacking move the position does not support — a sacrifice or advance the engine lines do not justify. Distinct from a sound attack followed up badly.
Time-trouble errors
Swings on moves played with little time left on the clock, reported separately from the rest.
Conversion mistakes
Technique problems at the point of converting an advantage — usually endgame knowledge, and usually trainable quickly.
From Classification to Training
-
Analyze a batch of games
Import from Chess.com or Lichess and analyze. One game shows you one blunder; a batch shows you which kind you keep making.
-
Read the ranking
Kinds of blunder are ranked both by how often they happen and by what they cost, each shown with the games behind it. Start where the two agree.
-
Train the positions themselves
Each category links to the positions in your own games where it occurred. Those become puzzles directly, and then similar positions test whether the idea transfers.
-
Check it on new games
Import the games you played after the training and see whether that kind still turns up as often. What counts is the change in your games, not the hours you put in.
Frequently Asked Questions
What counts as a blunder here?
A move that costs you a serious share of your chances of winning the game, compared with the best move available. Smaller losses are labelled mistakes and inaccuracies, so a normal wobble in a balanced position is not called a blunder. Because everything is scaled by your chances of winning rather than by raw engine numbers, a big engine swing inside an already decided position counts for very little — which is how it should be.
Does every blunder get a category?
No, and it says so rather than guessing. A blunder is only given a name when something specific went wrong that can be pointed at — a piece left hanging, a tactic missed, a defence overlooked, an attack the position did not support, a won ending mishandled. Anything else is reported honestly as a positional error, and still counted in your totals.
Which platforms can I analyze?
Import games directly from Chess.com or Lichess, or upload a PGN. Classification works the same way regardless of source.
Will fixing my top category fix my results?
It should mean that kind of blunder happens less often, and you will be able to see that in your later games. Results depend on far more than one category, so ChessLogix shows you what actually changed rather than promising a rating.
Is blunder analysis free?
Yes. Importing your games, engine analysis, sorting blunders into kinds, ranking them, reviewing the positions they came from and turning them into puzzles are all free. If you want a particular blunder written out in plain English — what your move was going for and what it allowed — that is AI Move Explanation, which comes with the monthly AI membership.
Does it distinguish blunders from time trouble?
It reports swings that cluster in low-time situations separately, which is usually enough to see whether a problem is chess or clock.
Find Out Which Kind You Actually Make
Import a batch of games from Chess.com or Lichess and let ChessLogix sort your blunders into kinds. Sorting them, ranking them, reviewing the positions and turning them into puzzles are all free.