Published results
Guess distribution
“Suppressed” marks a low-count result. A second cell may also be hidden so subtraction cannot reveal it.
Opt-in player data
The Wordle Observatory
See how contributors solve Wordle. Games are added only when a player opts in after reviewing one. The dataset stores a few game facts, but no grids, answers, guess sequences, accounts, IP addresses or cross-game identifiers.
Data through
· completed through Wordle #1,905 worldwide
Generated .
This is the worldwide-completion boundary for the rolling aggregate, not a claim that every eligible puzzle has a public row.
26opt-in games
22puzzles observed
0qualifying recent puzzle rows returned
0thresholded opener rows
…solve rate
…average turns when solved
Contribute a finished game
Paste a share grid and review it first. Nothing joins the dataset unless you separately opt in.
Download and reuse
Four scoped exports and two embeddable charts
The files separate the overall aggregate, seven result cells, qualifying opener rows, and the latest qualifying puzzle rows. Every file repeats the snapshot ID, generation time, worldwide data-through boundary and 10-game publication threshold. Blank optional metrics are withheld values, not zeroes.
Overall CSVSample size, coverage, solve and optional aggregate metrics
Result distribution CSV1–6 and X with explicit published, suppressed or withheld status
Thresholded opener CSVOnly openers with 10 or more contributed games
Qualifying puzzle CSVLatest 50 eligible rows; live puzzles excluded worldwide
You may embed either SVG as an ordinary image. Please retain a source line and the self-selected-sample caveat. CC BY 4.0 requires creator credit, but no hyperlink is required.
Reproducible monthly record
Immutable monthly snapshots
Rolling files can change as people contribute or withdraw a game. A monthly snapshot freezes the exact public bytes, checksums, citation, licence, thresholds and caveats after a calendar month is complete worldwide. It is a cumulative publication snapshot—not a count of games submitted during that month.
No checked-in monthly snapshot is available yet. The publication job fails closed until the whole month is complete worldwide and the rolling JSON passes the same answer-free, threshold and timestamp validation used by the live exports.
Playing conditions
What contributors reported
- Hard Mode share
- 44.0%
- Average difficulty opinion
- 3.23 / 5
- Average decision efficiency
- …
- Minimum public subgroup
- 10 games
These are self-selected contributions, not a representative estimate of every Wordle player.
Hard Mode is provisional. Earlier answer-page contributions could not distinguish an untouched optional Hard Mode checkbox from No. New contributions preserve Prefer not to say, but the historical responses cannot be separated. Do not treat this share as a population estimate.
Method and privacy
How records are collected and published
Each contribution contains only the puzzle number, result turn count (or X), optional Hard Mode status, optional opener, optional 1–5 difficulty opinion, and optional decision-efficiency bucket. A random per-game token is hashed solely to make retries idempotent.
No answer, colored grid, later guesses, account, email, browser fingerprint, user-agent or reusable player identifier is stored in the observatory table.
Publication rules
- Contributing is off until a player ticks an explicit consent box.
- Live puzzles remain hidden until they are finished in every timezone.
- Puzzle and opener rows appear only at 10 or more games.
- Each optional average or share needs 10 actual responses of its own.
- The Hard Mode field is marked provisional because earlier answer-page records could not separate an untouched optional checkbox from No.
- Page summaries use only the published rows and cells.
- Low result cells remain suppressed, along with a complementary cell when needed.
- Thresholds guard each published snapshot; the rolling aggregates do not claim differential privacy.
- Decision efficiency is rounded to five-point buckets.
- Public CSV, JSON and SVG files contain thresholded aggregates only; answer words are omitted.
- Every rolling file states the same generated time, worldwide data-through date, snapshot ID and minimum sample for that response.
- Monthly snapshot directories are append-only and publish only after the named month has ended in UTC−12.
- Records are self-reported and cannot be independently verified; rate limits and per-game retry tokens reduce casual duplication.