How Our Predictions Work
Every prediction on FootyTips.io comes from a transparent, quantitative model — no guesswork, no hype. Here is exactly how it works.
1. Elo ratings measure team strength
Each national team carries an Elo rating — a single number capturing current strength. Stronger teams have higher ratings; the gap between two teams' ratings is what drives the forecast. Ratings are seeded from international form and updated after results.
2. A Poisson model turns strength into goals
We convert the rating gap (with a small adjustment for venue) into an expected goals figure for each side. Football goals follow a Poisson distribution closely, so from those two expected-goal values we compute the probability of every plausible scoreline (0–0, 1–0, 2–1, and so on).
3. From scorelines to probabilities
Summing the scoreline grid gives everything on a match page: win / draw / loss probabilities, the most likely score, and the chances of over 2.5 goals and both teams to score.
Honesty about uncertainty
These are probabilities, not guarantees. Upsets are part of football — and part of any honest model. We settle predictions after each match and publish our real hit rate rather than cherry-picking wins.
FAQ
How accurate are the predictions?
No model is a crystal ball. We publish probabilities — a 65% favourite still loses about one time in three. After matches finish we settle every prediction and publish our real hit rate, so you can judge the model honestly.
Do you guarantee winners?
No. Anyone guaranteeing football results is selling something. We provide probability estimates from a transparent model — useful for context, never a sure thing.
What data powers the model?
Team strength (Elo), home/neutral-venue adjustment, and expected goals derived from those ratings. Fixtures and squads come from public football data sources.