Chess.com Game Review tells you what. A coach tells you why.
After a game on Chess.com you can run Game Review and get every move graded: best, excellent, good, inaccuracy, mistake, blunder, with an accuracy score at the top. Lichess gives you the same thing for free with its computer analysis. These tools are genuinely good. They are also the reason many players review every game and never improve.
What an engine review is good at
It is honest about the position. If a move lost two pawns of evaluation, it says so, and it shows the move that would not have. For checking a specific moment, “was my sacrifice sound?”, nothing beats it. The accuracy number also tracks reasonably well with rating over many games.
What it cannot tell you
It grades one game at a time, so it cannot see that the blunder on move 23 is the same blunder you made on move 19 last Tuesday and move 31 the week before. It names the move, not the habit. It says “Nf5 was a blunder, Rc1 was better,” and leaves you to work out that all three games were lost because you moved a piece away from the defence of another.
It also has no idea what is normal for you. An 1800-rated player and a 900-rated player get the same red marks for the same move, when the useful lesson for each is different.
- One game, not a pattern across many.
- The better move, not the reason you did not see it.
- No memory of last week, so no sense of progress or regression.
- No plan for what to practise next.
How to get the “why” yourself
Keep a one-line log. After each review, write the decisive mistake as a category rather than a move: hung a piece, missed a fork, bad trade, pushed a pawn in front of my king, ran out of time. After twenty games, read the log. The category that appears most is the reason you are at your current rating, and it is almost never the one you would have guessed.
Then stop reviewing every move. Review the two or three turning points per game and spend the saved time drilling the one category that keeps coming up.
Or let the coach keep the log
This is the gap Deepline was built for. It runs the same engine over your whole history, sorts the mistakes by cause instead of by game, and tells you which habit is costing you the most rating. The free scan at /try shows the idea on one game; the full version keeps the log for every game you play.