Game-BasedPuzzles
Your own blunders, handed back as puzzles. The loop finally closes.
I rebuilt it inside a replica of Chess.com, but only Game-Based Puzzles runs end to end. The surrounding pages are there for context.
Context
Review already works.
It just ends too early.
You see the mistake once, understand it, and move on. Nothing brings that exact decision back for practice.
- 01Play
- 02Review
- 03Understand
- 04Move on
The shift
Close the loop.
- 01Play
- 02Detect
- 03Practice
- 04Feedback
- 05Progress
- 06Return
Every Chess.com puzzle (Daily, Rush, Battle) draws from other people’s games. Not yours.
One blunder, end to end
Watch it happen.
Blundervs. M. Kowalski · RapidMedium
bookA rapid game, move 14. Stockfish hasn't looked at it yet.
01 · Play
A real game, from your archive.
Not a curated library position. Yours.
02 · Detect
Move 14: Rd3
A forced mate was on the board. After Rd3, Black plays Rxd3 and takes the rook instead.
03 · Rewind
The board rewinds.
Back to the instant before the decision. Red on the bar is what you gave away.
04 · Decide
Find it again.
Legal dots, a ring for the capture. Hints cost your clean solve, never your progress.
05 · Feedback
Checkmate on the back rank.
Green sweeps up the bar. You won back exactly what the blunder cost.
06 · Return
Home updates instantly.
One shared progress state. The two screens can't drift.
The best puzzle for you is one you’ve already failed: under a clock, in a game that mattered.
Interactive
Your turn: three puzzles.
Live, right here: a back-rank mate, a knight fork and a hanging queen. Each state is explained as it happens.
Missed TacticPuzzle 1 / 3blunderOn move 14 you played Rd3 and let a forced mate slip. It is still there, so start it.
M1 → −5.0The hard part was never generating puzzles. It was generating fair ones.
The finding that reshaped the product
How the set gets built
Rating decides how wide we cast the net.
Prototype heuristic: a lower-rated player’s games may hold more teachable moments, while a stronger player’s games may need a wider search to find worthwhile ones. So the depth of analysis is read off the player’s Elo, rather than fixed at one number of games.
An exploration rule, not a validated formula. The bands below are a starting point to test, not a measured result.
The set size is still never padded. The net widens, the quality gates don’t move. If the games only held eight fair puzzles, the set is eight.
Two versions, one switch
Which games become
which puzzles, and when?
Two answers were arguable. Rather than argue in the abstract, both were built, so they can be tested on real players rather than decided in a room.
Your last N reviewed games
Everything mineable goes into one pool; the best of it becomes a single set that’s always there.
- Zero navigation: open the page, puzzles are waiting
- Best-of selection from a wide pool
- No sense of time or rhythm; it’s a bucket, not a record
- Once solved, there’s nothing to come back to
Puzzles from the day you played
Grouped by date, with a stepper and a month calendar. No target, no padding. A day is as long as that day’s chess deserved.
- The calendar is the progress record, so streaks and gaps become visible
- An uncleared day sits there, visibly uncleared
- Play nothing, get nothing: empty days are real
- A navigation decision before you can solve anything
Both are live in the prototype behind a switch, so the same account can be run on either. V1 optimises the session: always full, no decisions. V2 optimises the return: an uncleared day pulls you back. Which one wins is a research question, not a taste one.
Built pieces
The parts that
carry the feature.
Real screens from the prototype. One line each.
Game Puzzles
4/12 completed
Home hero card
Reads the same progress state the solver writes to, so it can never drift.
Evaluation bar
The red band carries the size of the drop, or the mate thrown away. No subtraction required.
Brilliant
Great
Best
Excellent
GoodBook
Inaccuracy
Mistake
MissBlunderPuzzles are mined from the bottom four.Classification badges
Official art, one colour each: the taxonomy the queue is built on.
missed♜xd8#. Checkmate on the weak back rank. Spot a forced mate in one and start it with the right move.
The coach
Two halves: what happened, then the pattern to take away.
Get Unlimited Puzzles!
You’re all caught up!
Completion: free vs premium
The set is never trimmed. Free members see all 12 and exactly where the wall sits.
♜xd8#is correct!Solver rows
Figurine notation, because “Rd3” reads wrong in a Chess.com context.
Product decisions
Five calls that
changed the product.
Decision 01Brilliant
The engine’s best move is not a puzzle.
brilliantNot every engine disagreement is teachable. Three gates now stand between a raw disagreement and a puzzle: findable, singular, worth it. Eight solid puzzles beat fifteen with five duds.
Free vs premium
The wall is the pitch.
The set is never trimmed for free members. They see all twelve, and exactly how far in the wall sits.
Locked puzzles stay visible, greyed.
A queue that looks three long has nothing to sell. The puzzles you can see but can’t reach are the argument, stated as a fact, not a sales line.
Retry 3 · Next theme · Solve again.
The end card is about improvement, not celebration. Actions are ordered by usefulness and the default focus moves down the list as options disappear, so the most useful remaining action is always under your finger.
- 01Hit the wall at puzzle 3
- 02Upgrade in place, no reload
- 03Short celebration
- 04Resume at puzzle 4
The hypothesis
What I’d want to measure.
Do players make fewer of the same mistakes after drilling them?
Solved without hints or the solution?
Do players leave when a puzzle feels unfair or irrelevant?
Do they come back for more of their own puzzles?
How many reviewed games yield one worth practising?
A prototype, not a launch: outcomes to validate, not claim.
What I’d carry forward.
- Spaced repetition
- The prototype builds the puzzles; it doesn’t yet schedule their return.
- Named weaknesses
- “Blunder” is a severity, not a lesson. Cluster into back rank, hanging pieces, endgames.
- Queue first, diary on top
- The queue is the better product, the diary the better habit. Ship the queue as default.