Skip to main content

Frequently asked questions


How fresh is the data?

LineupAPI publishes data as snapshots.

  • A new snapshot is created every 15 minutes
  • Fixtures are available up to 17 days ahead (forecast window)
  • Predictions refresh once per day per fixture (not every 15 minutes)
    As a rule of thumb: for fixtures inside the 17-day window, predictions update every 24 hours, counted backwards from kickoff time.

So when you call /fixtures, you’re getting the latest snapshot at that moment.
When you call /predictions/{fixtureId}, you’re getting the latest daily prediction update for that fixture.

Refreshing too often wastes credits

If you refresh faster than the update frequency, you’ll usually get the same data again.
For most apps:

  • refresh /fixtures every 15–30 minutes
  • refresh /predictions/{fixtureId} about once per day (or when you expect the next daily update)

Why are goals/assists/minutes high for an out-of-squad player?

Because most prediction groups are conditional on the player starting (being in the starting 11).
At 17 days out, “will they start?” is the biggest unknown — so LineupAPI separates availability from performance.

Read: Why predictions depend on starting


Why is a prediction group missing from the response?

Most of the time it’s one of these:

  • You didn’t request it via groups=...
  • There’s a typo in the group name (example: assist vs assists)

Do you provide historical predictions for backtesting?

Yes. Use the History endpoints to retrieve predictions “as they looked” at a past snapshot timestamp.
History coverage starts on 2023-07-25 (older timestamps are not available).

To make backtesting easier, every History response includes helpful timestamp headers, which advanced developers can find:

  • x-timestamp-requested — The snapshot timestamp you requested
  • x-timestamp-current — The snapshot timestamp actually returned (the closest available snapshot to your request)
  • x-timestamp-previous — The previous available snapshot timestamp (useful to step backward in time)
  • x-timestamp-next — The next available snapshot timestamp (useful to step forward in time)

How do I know how many credits a request cost?

Credits are calculated per request based on the endpoint and the prediction groups you ask for.
See Credits & usage and the endpoint documentation for the exact cost rules.

Advanced developers can check the response headers on successful requests:

  • x-credits-cost — credit cost of that request
  • x-credits-used — credits used since the last reset
  • x-credits-remaining — credits remaining until the next reset

You can also see your current credit usage and remaining credits anytime in your dashboard.