Chess Coach

The learning path

Paste a game

Find the turning points in a game and turn them into practice. Pick any way in:

Import from chess.com

Bring in your game history straight from chess.com. Your games go to your Coach's Database, ready to train on.

Import from lichess

Bring in your game history straight from lichess. Your games go to your Coach's Database, ready to train on.

Games you've imported before are skipped, so running this again later only adds new ones.

Flag this answer

Also saved on this device under Settings → Flagged answers. What's sent: the position, the imbalance, both answers and your note. No name or account.

Your flag was reviewed

Thanks for flagging it. Here's what the coach decided.

Flagged answers

What's your rating?

This sets which imbalances you practice and how many pieces are in the positions. Use your chess.com or lichess rating if you have one.

Pick a look

Tap one to see it on the app right away.

You can change both any time in ☰.

Which player is you?

Check every name you play under. Games with your name go to My games; the rest go to Other games.

Promote to

How this works

Strong players don't just look for tactics. They first list the imbalances: every way the two sides differ. The plan follows from that list. This app trains that habit, and the heat map shows you the evidence.

The eight imbalances at a glance

Modes

• Train: you get a position from a master game or one of your own. Decide who each imbalance favors, then check. Every answer comes with an explanation and a "Show on board" button.
• Weak Piece: first tap the opponent's weak piece, then make a move that attacks it safely. Find it the way strong players do: scan for targets before looking for moves. A piece is hanging if nothing defends it, balanced if it has as many attackers as defenders (one more attacker tips it), and a piece alone on a line to its king can be pinned. Any safe attack scores 1 point, the strongest one scores 2. Hints get more specific each time you ask, and you can retry as often as you like.
• Analyze: step through any game (master games are built in, or upload your own PGN). Turn on imbalances or quiz yourself on any move.
• Explore: set up any position by dragging pieces, or paste a FEN.
• Train: training games against the engine or a friend on this device, with the heat map live, or over a real board with Friend + Coach and Scoresheet + Coach.

The eight imbalances

Reading the heat map

The heat map shows where each side's pieces exert influence: each square is tinted by how many pieces attack it. It is evidence, not a verdict. A square you attack isn't automatically one that matters, so use it alongside the imbalances. Your side's attacks (the side at the bottom of the board) use the first color, the opponent's the second. "Both" blends them in proportion to the number of attackers. Turn on Blue/orange if red and green are hard to tell apart.

A note on the answers

The imbalance answers come from rules a coach would use (counting attacked squares, pawn weaknesses and so on), not from an engine. They are signals, not verdicts: every explanation says what was noticed, why it might matter, and what plan it suggests, with a note when a rule is known to be shaky. Close calls are marked Even. Treat the answers as a second opinion to argue with.

Game reviews say where each finding comes from. The quick built-in engine only says a move "looks weaker"; when Stockfish has checked it, the review says so.

Piece set: Colin M.L. Burnett, CC BY-SA 3.0 (via lichess). Move rules: chess.js by Jeff Hlywa, BSD license. Typeface: Atkinson Hyperlegible by the Braille Institute, SIL Open Font License. App icon: chess knight from Font Awesome Free, CC BY 4.0. Curriculum positions: real rated rapid and classical games from the lichess.org open database (public domain, CC0), with lichess's Stockfish evaluations. Coaching puzzles: the lichess.org puzzle database (public domain, CC0). The "In real games" figures come from rated rapid and classical lichess games: about 71,000 from 2015 and 12,000 from September 2025, with both eras weighted equally. Foundations positions come from the 2025 games. Deep analysis: Stockfish 18 (stockfish.js by Nathan Rugg), free software under the GNU GPL v3; source at github.com/official-stockfish/Stockfish and github.com/nmrugg/stockfish.js.