How to read your chess review
A ChessReviewMaxx report has a lot in it. This page walks through each section in the order you meet it, and what to actually do with it.
Last updated: September 7, 2026
1. Your reports page
The site opens on your reports — every analysis you have run or opened, kept in this browser so it survives closing the tab. From here you can create a new report, load a local report from a file, or tick reports to merge or delete them. Reports are matched to the account they belong to, so merging is only offered within one player.
2. Setting up the analysis
Choose a platform (Chess.com or Lichess) and enter a username. Use Check to confirm the account exists. Then narrow the set of games:
- Time controls — bullet, blitz, rapid, daily. Analyse one at a time; your rapid mistakes and your bullet mistakes are different problems.
- More filters — colour, result, rated vs casual, your rating range, opponent rating range, date range, and newest-vs-oldest order.
- Games to analyze — a fixed number, or tick Analyze all games matching the criteria for the whole history.
The Stockfish engine settings panel controls analysis depth (higher is stronger and slower), the number of parallel engine threads, and how many candidate lines the engine keeps. The defaults are a good balance for a large batch. Everything runs on your machine, so more games and higher depth mean longer waits — keep the tab open while it works.
3. The report dashboard
When analysis finishes you land on the results screen. The top of the page restates the full set of criteria that produced this report, so you always know what you are looking at.
Move quality and accuracy
Counts and percentages for each move class — Brilliant, Great, Best, Excellent, Good, Book, Inaccuracy, Miss, Mistake, Blunder — plus an accuracy figure for you and for your opponents. What each class means is explained on its own page. Watch the ratio of Inaccuracy / Mistake / Blunder more than the absolute numbers.
You vs. your opponents
Your accuracy edge over the specific players you faced, their average rating, and a scatter of your accuracy against opponent strength. If your accuracy collapses against higher-rated opponents, that is a composure and time-management signal, not a knowledge gap.
Decision and piece profile
How well you handle forcing vs quiet vs defensive positions, conversion accuracy when you are winning, and accuracy and usage for every piece. Missed mating lines, missed checks and captures, and castling outcomes are all counted and compared with the opponents from the same games.
Charts
Accuracy over time (you vs opponents), a results donut, accuracy by phase, and rating progression. Every point has a hover tooltip; on the rating chart, big swings are annotated with what stood out in those games.
4. The improvement plan
This is the part to act on. It is a short, prioritised list of concrete actions built from your recurring mistakes — not "study tactics" but "in the Najdorf you leave theory on move 8 with a6; the move is Be7, and here is the line." Each item links straight to the position in your game.
- Must-study opening lines — for each repertoire, the exact point where you leave theory, the move that went wrong, and the engine's fix plus continuation.
- Must-study endgame theory — the endgame types you score worst in (rook endings, king and pawn, opposite-coloured bishops…), each with the specific theory to learn (Lucena, Philidor, the opposition, the wrong rook's pawn…).
- Structures and positional patterns — the pawn structures that recur in your games (isolated queen's pawn, hanging pawns, majorities…) and the plans that go with them.
5. The single-game review screen
Click any game to open the move-by-move review.
- Board and badges — each move carries a move-quality icon; the move list also shows the clock time for every move.
- Eval bar and eval graph — the graph has a colour-coded dot for every move's quality. Hover for the move and its evaluation; click to jump there.
- Per-move feedback — click a move for an explanation plus the engine's best line as a clickable variation you can play through.
- Free piece mobility — drag or click-to-move any piece to explore your own ideas. Stockfish evaluates whatever position you reach, live. "Back to game" returns to the real continuation; arrow keys navigate.
- Opening name — recalculated as the line on the board changes, so transpositions are named correctly.
6. The training course
ChessReviewMaxx turns the moments you misplayed — but only where you were in an equal-or-better position — into puzzles. Consecutive mistakes are merged into multi-move exercises. Hints are staged so they never hand you the answer, and completed lines are explained in full for both sides. Where several moves are equally good, the board returns to the branch point so you have to find each one.
7. Working with a results set
- Automatic saving — a report is kept in this browser as soon as it finishes, and updated whenever it changes. Nothing is lost if you forget to save.
- Save analysis — write the report to a JSON file you keep. Once a report has a file, saving again rewrites that file rather than making a new copy.
- Load a local report — reopen a saved file instantly, with no recomputation.
- Add games — fetch and analyse more games matching the original criteria (skipping ones already done) and merge them in. Analysing an account you already have a report for offers to add only its new games rather than starting a rival report.
- Merge reports — tick two or more reports of the same account on the reports page. Overlapping games keep whichever copy came from the more thorough engine pass.
- Filter these games — narrow the analysed set by the same criteria; every stat, chart and study line updates to the subset.
A realistic workflow: analyse 100–200 recent rapid games, read the improvement plan, pick the one opening line and one endgame at the top of it, work through those puzzles, then re-run the analysis in a month and check whether that leak closed.