Research method · corrected 30 July 2026

Football Prediction Math: A Plain-Language Guide

Ratings can summarise past performance; probabilities express uncertainty. Neither becomes trustworthy merely because it contains decimals.

Evidence status: legacy numeric claims withdrawn where no retained input table, raw output or version record exists.

What this page can support

A model description should name the target, input window, weighting, validation period and treatment of draws, extra time and penalties.

Ratings, probabilities and calibration are different

A rating orders or separates teams using defined inputs. A probability maps that difference to a specific event, such as a win in regulation. Calibration then checks whether events assigned about the same probability occur at about that rate in unseen matches.

Each step needs its own data, version and test period. A precise-looking percentage is not evidence that those steps were retained or validated.

Evidence checklist

  1. Define exactly what the probability predicts.
  2. Use out-of-sample matches for validation.
  3. Check whether stated probabilities are calibrated.
  4. Publish uncertainty and known blind spots.
  5. Label worked examples as illustrative.

Limit of the archive

No numeric example on this page is represented as a retained 26cup tournament run. The archive therefore explains what evidence a reproducible model would need instead of reconstructing missing inputs after the result.

Verify the tournament outcome

Use FIFA's official fixtures, results and final standings when you need the authoritative match record.

Open FIFA fixtures and results · Open final standings