Data Sources and Editorial Policy About the data

• 2,537 men's Tests, 1877 to March 2024, with 3,271 players.
• 4,981 men's one-day internationals, 1971 to August 2026, with 3,151 players.
• 19 IPL auctions (2008 to 2026, 3,836 lots) and every IPL season's player statistics.
• 15 Big Bash League seasons, with 526 players.
• 13 men's Cricket World Cups, 1975 to 2023, rebuilt from the scorecards.

All of it is men's international and franchise cricket. Women's cricket, domestic first-class cricket and T20 internationals are not covered yet, and nothing on this page is presented as more complete than it is.

The Test archive holds a scorecard for every men's Test from the first, at Melbourne in March 1877, to the end of March 2024: teams and captains, the toss, the result, every batting and bowling card, fall of wickets and partnerships. The scorecards come from published match records; where a match also appears in Cricsheet's ball-by-ball data, we use it to confirm totals, extras and declarations.

Known gaps. The archive stops at 30 March 2024; later Tests are not included. Player of the match was not awarded in early Test cricket and is recorded from the mid-1970s. Debut lists are present for about 61% of matches, so each player's debut is worked out from his first appearance instead, which agrees with every debut list we have. Innings totals were rebuilt to include extras, and a small number (20 of more than 9,000 innings) could only be recovered as a lower bound; those are marked and kept out of records.

The ODI archive starts with the first one-day international, at Melbourne in January 1971. It joins two sources: published match records to early 2024, and Cricsheet from 2024 onwards. The two spell some teams and players differently (the United Arab Emirates appears as "U.A.E." in one and in full in the other), so teams are merged under one name and players are matched by identifier, never by name alone.

Known gaps. Cricsheet does not publish Afghanistan's ODIs after 12 March 2024, so the archive has none either; this is the source's policy, not an error we can fix by downloading again. Debut lists are present for about a third of matches, so debuts are again worked out from first appearances.

The Test and ODI player databases link each appearance in the archive to one player. That works for 99.8% of Test appearances; the rest are players who share a name with a contemporary and cannot be told apart from the scorecard alone, and we leave them unlinked rather than guess. Careers computed from the archive match the official career figures for nearly every retired player.

The venue database merges the 282 ground names found in the scorecards into 245 grounds, because grounds are renamed and sponsors change. Every merge is listed with its evidence, and grounds that might be the same place but cannot be shown to be are kept apart. The World Cup pages are rebuilt from the real scorecards, and every match that was scheduled but not played is listed with the reason.

IPL and Big Bash season statistics come from published season records. The IPL auction archive is built from the published lists of each auction's signings (including Wikipedia's, under CC BY-SA), reconciled lot by lot; where a list is wrong and a second source says so, the correction is recorded with that source. The ball-by-ball detail on IPL player pages comes from a reconstruction of 1,243 IPL matches cross-checked against Cricsheet, and is shown only where it agrees exactly with the season records.

Big Bash players are contracted by their clubs, so there is no auction history, and neither league's salaries are shown.

The archives, the player databases, the venue pages, the World Cup and auction histories, the guides' figures and the research are real cricket. The games are simulations: a match you play, a career's future fixtures and the ratings the games give players are part of the game, not the historical record, and they are never mixed into the archive. How CricArcade Works compares simulated matches with real ones.

Every figure in the site's writing is computed by a script over these records, not typed by hand, and the site's tests fail if a published count drifts from the data. The records themselves are checked against each other: innings totals against the result margin, debuts against debut lists, careers against official figures, auction lists against a second record lot by lot. Where a check fails, the record is corrected only when a second source gives the right value; otherwise it is flagged and kept out of records and leaderboards.

CricArcade is written and edited by Vishal Jaiswal, who is responsible for every published word. AI tools help with the work: drafting from the figures our scripts compute, checking prose against the scorecards at a scale one person cannot manage, and tightening wording. Nothing goes live until its figures have been checked against the records and Vishal has decided to publish it, and every research piece says how it was made.

The writing describes real cricket and what the games simulate, in cricket terms. It does not publish the games' formulas or settings, for the reasons given on How CricArcade Works.

If a number looks wrong, tell us which page and what you think the right figure is, with a source if you have one. We check it against the scorecard, correct the page if we were wrong, and update its date.

Cricsheet data is used under the Open Data Commons Attribution License 1.0; data from cricsheet.org. Wikipedia text and lists are available under CC BY-SA.