Skip to content
Forepitch
  • Season
  • Upsets
  • Beat the model
  • Mini-games
  • Accuracy
  • Strategies
Premium
EN
  • Русский
  • English
  • Português
  • Español
Sign in
  • All games
  • Football
  • Hockey
  • Tennis
  • Basketball
Sections
  • My Forepitch
  • Season
  • Upsets
  • Beat the model
  • Mini-games
  • Accuracy
  • Strategies
  • Premium
  1. Home
  2. Prediction basics

Prediction basics

Short explainers for anyone new to reading predictions: where the percentages come from, what they mean and what to expect from them.

Every number in these articles comes from our own check of the model and from the upsets of recent weeks, and it updates with the site.

  • What is xGxG measures how many goals a team’s chances were worth. How expected goals differ from real ones, where they fall short and how the Forepitch model uses them.Read →
  • How to read probabilitiesWhat “62% to win” means, why such a favourite fails to win about 4 games in 10, what calibration is and how a probability differs from bookmaker odds.Read →
  • Why favourites loseHow often favourites fail to win, from coin-flip games to heavy favourites, across 48,148 club football matches in the model’s check — plus recent upsets and why they are inevitable.Read →

This explains how probabilities work; it is not betting advice. 18+.

  • Model accuracy
  • How the model works
  • Prediction basics
  • Betting strategies
  • Season forecasts
  • Power rankings
  • Upsets of the week
  • The model’s week
  • Beat the model
  • Mini-games
  • Premium
  • About
  • Privacy
  • Terms of use
  • Widget for your site

18+Statistical estimates of a model: analytics only, with no guarantee of any result. Times are Moscow time (UTC+3). Updated at 21:52.

  • Русский
  • English
  • Português
  • Español

Cookies only keep you signed in. No ads or third-party trackers; for statistics your browser only remembers the dates of your visits, with no identifier.Cookies only keep you signed in; no ads or third-party trackers. May your browser remember the dates of your visits (no identifier)? It tells us whether readers come back. Details