The API

Ask with a plain GET; no key, no sign-up. Any site may call it from a browser (CORS is open).

curl https://collegefootballdigest.com/api/v1/teams/michigan

Every answer comes in the same envelope. attribution lists each source, its license and the fields it covers; links.html is our page for the same thing.

{
  "data": { ... },
  "meta": { "version": "v1", "stats_version": 412, "as_of": "2026-10-14T10:52:03Z", "ttl": 86400,
            "lists": { "series": { "total": 31, "returned": 10, "truncated": true } } },
  "attribution": [
    { "source": "CollegeFootballDigest, computed from game results by CollegeFootballData.com",
      "license": "CC BY 4.0", "license_url": "https://creativecommons.org/licenses/by/4.0/",
      "fields": ["record", "elo", "badges", "series"] },
    { "source": "Wikipedia: Michigan Wolverines football", "license": "CC BY-SA 4.0",
      "license_url": "https://creativecommons.org/licenses/by-sa/4.0/",
      "url": "https://en.wikipedia.org/wiki/Michigan_Wolverines_football",
      "fields": ["wikipedia.titles.conference"] }
  ],
  "links": { "html": "https://collegefootballdigest.com/teams/michigan",
             "series": "https://collegefootballdigest.com/api/v1/teams/michigan/rivalries" }
}

Errors carry their HTTP status and a body like

{ "error": { "code": "not_found", "message": "No team at /teams/michiganx" } }

A list the answer caps or pages says so in meta: total (null where it isn’t counted), returned and truncated. Paged lists take ?limit= and give the next page as links.next; pass it as it is (its cursor is opaque). Lists cut short inside an answer, like a team’s ten biggest series, are under meta.lists.

{
  "data": { "measure": "...", "series": [ ... ] },
  "meta": { ..., "total": 4210, "returned": 100, "truncated": true },
  "links": { "next": "https://collegefootballdigest.com/api/v1/rivalries?limit=100&cursor=100" }
}

Every answer has an ETag: send it back as If-None-Match and an unchanged answer comes back as 304 with no body. Browsers may keep an answer five minutes, shared caches until the next update. An answer that lives at another address (a team’s old address, the other order of a rivalry, ?week= on a season) answers 301 to it. Anything but GET, HEAD and OPTIONS is 405 with Allow.

Ids and addresses are the site’s own: a team is its address (/teams/ohio-state is ohio-state), a game its number. Every team reference also carries id, the team’s number in the game records (it stays put when an address changes), and links: html, our page, and api, its address here (null for games and players, whose pages have the rest). Scores aren’t in the API, only who won; links.html goes to the game’s page.

We serve what we compute; we don’t re-serve CollegeFootballData.com’s own feeds.

The game results behind the site come from CollegeFootballData.com. From them we compute our own numbers, and those are in the API: Elo ratings and ranks (each season, going into each week, and a program’s whole history), the research desk’s notes, series totals as played and official, winning streaks, who won a series’ last meeting, who holds each rivalry trophy, badges, and each program’s record on file. They’re ours, under CC BY 4.0, credited as “CollegeFootballDigest, computed from game results by CollegeFootballData.com”.

What CollegeFootballData.com publishes as it is stays out: no lists of games, no scoreboards or schedules, no scores or box scores, no play-by-play, no player stat lines, rosters, polls, recruiting or betting lines, and none of its ratings (SP+, SRS, FPI, PPA). A game’s score is on its page; links.html and each game’s url go there.

Not in the API

• CollegeFootballData.com’s own feeds: game lists, scoreboards and schedules, scores and box scores, play-by-play and drives, player and team stats, rosters, polls, recruiting, the transfer portal, NFL Draft picks from 1967, betting lines, and its ratings (SP+, SRS, FPI, PPA). Its terms don’t allow its data as a standalone dataset, and we don’t offer feeds that are the same as its own. Game pages on the site show them; the API links to those pages.
• Massey Ratings rankings. Shown on the site by Kenneth Massey’s permission, which covers the site only.
• Outlets’ headlines and photos. They belong to the outlets; the site links to their stories.
• Bluesky posts, YouTube highlights and view counts, podcast episodes. Their platforms’ terms cover showing them on the site, not handing them on.
• Stadium outlines (OpenStreetMap). The Open Database License would carry over to anything built from them.
• ProPublica’s nonprofit totals (2011-17). Its terms bar redistributing them on their own.
• Team logos. They’re the schools’ trademarks.
• Anything about minors. Recruits and high-school players stay off the API.

The full list of datasets with their sources and row counts is at /api/v1/datasets, and every dataset can be browsed at /data.

1. Credit what you use. Print each source the attribution block lists for the fields you show, with its license, and link to CollegeFootballDigest (links.html where there is one). Each entry’s credit is a line you can use as is.
2. Wikipedia’s fields keep their license. Everything in a wikipedia block is CC BY-SA 4.0: credit the article (the entry’s url) and share what you build from it under the same license. We add no conditions of our own to it.
3. Our own data is CC BY 4.0. Elo ratings and ranks, notes, badges, series totals and streaks, winners, trophy holders, records on file and heat-board ranks: use them for anything, with credit.
4. Public-domain data stays free. Newspapers, NOAA and NWS weather and the federal school data carry no restrictions; a credit is courteous and the College Scorecard asks for one (CC BY 4.0).
5. Takes are excerpts. The feeds carry each take’s title, excerpt and link; the take itself stays on this site.
6. Be gentle. 60 requests a minute per network; past that the answer is 429 with Retry-After. Cached answers come back fast and don’t count. There’s no promise the API stays up or unchanged; v1 changes only by adding.

• /api/v1/takes: every take as a JSON Feed 1.1 (title, excerpt, link), kept 5 minutes. RSS too.
• /teams/<team>/feed.json: a team’s takes and our notes on its season, e.g. /teams/indiana/feed.json.

Two embeds, open data only, for any site to frame:

<iframe src="https://collegefootballdigest.com/embed/team/michigan" width="360" height="420" style="border:0" title="Michigan on CollegeFootballDigest" loading="lazy"></iframe>

<iframe src="https://collegefootballdigest.com/embed/on-this-date" width="360" height="520" style="border:0" title="On this date in college football" loading="lazy"></iframe>

Questions, or a source you think we’ve got wrong: admin@collegefootballdigest.com.