Score by opening and variation
Every figure is labelled for what it measures: "Your score: 26%", "Played in 19 games", "Reached in 62% of your games". No bare percentages whose denominator you have to guess.
Opening Lab
Find the Exact Branch Where Your Opening Results Drop
Follow your results beyond the opening name, find the branch where performance begins to fall, and prepare a practical alternative for your rating and time control.
Free board, no account · One full analysis free · Chess.com and Lichess
- No Chess.com or Lichess password — a public username is all it needs.
- Your first full analysis is free, and no card is required.
- Every recommendation comes from your own games and Stockfish analysis.
- Score counts a win as 1 point and a draw as ½.
- Every percentage carries the number of games behind it, so three games read as three games.
- Then follow the tree down to the reply where your score actually separates.
See Your Real Opening Performance
Before it recommends anything, Opening Lab reports your own record — with labels that say exactly what each number counts.
White and Black kept apart
Repertoire problems are colour-specific. A line you are comfortable meeting is not the same as a line you are comfortable playing, and the two records are never merged.
One account at a time
If you play under several Chess.com or Lichess handles, filter to one. A serious rapid account and a casual blitz account produce different repertoire evidence.
Filtered by time control
Bullet, blitz, rapid, classical and correspondence can be selected independently, because a line that is fine at 15+10 may be unplayable at 3+0.
Two Answers About the Same Position
"You score badly in the Italian Game" is not something you can act on. The Italian is a tree, and your record is almost never uniform across it — so exploration walks the branches you actually reach and stops where your score separates from what the position is worth.
- Engine, on the left: this position is worth +0.2 at depth 26. That is soundness.
- Human results, on the right: c3 is played in 29% of 13 million games and scores 52%. That is practical difficulty.
- The two are shown side by side and never merged — a line can score well for humans and still be unsound.
Find Practical Alternatives
Exploration searches human continuations for the rating range and time control you choose, and uses Stockfish to tell a playable practical weapon from an objectively losing trap.
Three sources, clearly separated
Compare the same position against your own games, the Lichess database filtered to your rating range and speeds, or Masters games. Which one you are looking at is always on screen.
What opponents actually play
Each reply carries how often it is played at your level, so preparation time goes to the moves you will meet rather than the moves theory considers critical.
Why each move is here
Every candidate says what put it on the list — an engine top choice, or a move that is simply common in practice at this level. A move is never called "best" only because humans score well with it.
Punishable mistakes
Replies that are frequently played but inaccurate are flagged with the concrete way to meet them, which is where most practical points at club level come from.
Choose the Kind of Line You Want
Exploration starts from an explicit question — which opponents, which speeds, and how much objective ground you are willing to give up — and every line that comes back carries those terms with it.
- 1. State the question: objectively strong, or practical against this rating band?
- 2. Read the answer: practical score, opponent mistake rate, and objective cost, kept apart.
- 3. Keep what you chose, with the reason it was recommended attached to it.
The Full Opening Lab Workspace
The same screen intact, at a width where the repertoire column and the explore controls are readable. Tap it to enlarge.
- The middle column is the line you are building and the recall cue you will need three weeks from now.
Keep It as a Repertoire, Then Rehearse It
Exploration that is not saved is exploration you will repeat. Decisions you make in the Lab become a repertoire with practice attached.
Recall cues, not just move orders
Attach the reason you chose a move to the move itself. The cue is what you will need three weeks later; the move order is not the hard part.
Only the decisions that are due
Practice surfaces what is slipping rather than replaying the whole repertoire, and tracks first-attempt accuracy separately from eventual accuracy.
Repertoire and sparring
Repertoire mode checks recall along the lines you saved. Sparring plays on past the end of your preparation, which is where preparation usually fails.
Frequently Asked Questions
What does "Your score: 26%" mean?
It is your score as a fraction of available points in that position — a win counts one, a draw counts a half. It is shown next to the number of games it was computed from, because a 26% score over four games and over ninety games are very different claims.
Where do the numbers come from?
Two separate sources, always labelled. Your personal record comes from your own imported Chess.com and Lichess games. Population-level frequencies and results come from the Lichess opening database, filtered to the rating range and time controls you select. Masters games are available as a third comparison. Engine evaluations come from Stockfish and are independent of both.
Is a high human score the same as a good move?
No, and Opening Lab never merges the two. A human score says how players in a rating band did from this position with a clock running — practical difficulty. An engine evaluation says what the position is worth under deep search — soundness. The disagreements are the informative part: a line that scores well and evaluates as equal is a practical weapon; one that scores well and evaluates as losing is a trap that works until it does not.
Do I need an account to use it?
The board is free to explore without logging in, including the human database comparison and the engine evaluation. An account is needed to see your own game statistics and to save preparations.
Does it handle transpositions?
Yes. Exploration and the saved repertoire are both kept as a graph rather than a list of lines, so positions reached by different move orders are recognized as the same position, and backing up to try a different order does not lose the work already done.
Why show Masters games at all if I am rated 1400?
Mostly as a sanity check on soundness rather than as a guide to what to play. If a line disappears entirely at master level while scoring well in your rating band, that is usually a sign it works because the refutation is hard to find, not because the line is good.
Start With the Opening You Score Worst In
Connect an account, let Opening Lab rank your openings by result, and follow the worst one down to the move where the trouble actually starts.
Analyze My Games Free