1. ELO → EXPECTED GOALS
Current Elo plus the host edge becomes a goal expectation for each team.
λA = base × exp(clamp(eloDiff / 460))Methodology & data sources
Predicta now ships the completed 2026 World Cup as a versioned 104-match FIFA result snapshot. Finished matches are fixed facts in the tournament review; the International Match Lab reuses the calibrated team model for fixtures from other competitions. Match scores come from a joint Poisson grid whose expected goals follow Elo, with a Dixon–Coles low-score correction (ρ = −0.15). A deliberately small 10% attack/defense residual is learned only from earlier regulation-time matches and shrunk toward neutral. The automatic World Cup profile uses a 50-Elo host edge, K = 15 live Elo updates, σ = 40 rating uncertainty and a 0.98 knockout scoring multiplier. Every forecast runs 10,000 simulations; no model setup is required. Accuracy is reconstructed strictly in FIFA kick-off order. The five extra-time matches are excluded from 90-minute score evaluation because FIFA does not expose their separate regulation score. Real knockout participants and winners override simulated bracket approximations once results are final. Data: final fixtures and results from FIFA; starting Elo ratings as of June 11, 2026 from worldfootballrankings.com / eloratings.net. This is a calibrated forecast model, not a promise — that’s football.
LIVE & TODAY
Every match in play right now plus everything scheduled for today (your local time). Tap a match for live forecasts — win/draw/loss, most likely final scores, next goal, over 2.5 goals, both-teams-to-score and what's at stake in the tournament — recomputed from the current score and minute. While matches are live the app re-syncs with FIFA every ~75 seconds (otherwise every 5 minutes) and the match clock keeps ticking between syncs. Use ‹ › to browse other matchdays (preview upcoming fixtures and their forecasts) and Today to jump back.
VALUE BETS
Upcoming matches where this model's forecast diverges most from the live betting market. For each game we take the consensus of major bookmakers (EU/UK/US books via The Odds API), strip the bookmaker margin (de-vig), and compare it to the model's win/draw/loss estimate. A green edge means the model rates that outcome more likely than the market does — a possible value bet; EV uses the best price across books. Matches are sorted by how far apart the model and market are. Odds refresh on demand using your own API key. Markets move and the model can be wrong — this is a disagreement finder, not betting advice.
Enter a free API key from the-odds-api.com to load live bookmaker odds — 500 requests/month free, no credit card required.
GROUPS & RESULTS
Every fixture shows the predicted exact score beside the real result. Green marks final facts; the dimmed dashed chip preserves the honest pre-kick-off forecast. The points column counts finished matches.
KNOCKOUT BRACKET
Each slot shows the most likely team, its chance of being in that match, and its chance of advancing. Tap a slot for all candidates and the pre-match score forecast.
HEAD-TO-HEAD
MATCH LAB
vs
ACCURACY
Honest pre-kick-off scoring in strict chronological order. Full model versus Elo-only baseline; lower is better.
PREDICTION GAME
Turns the forecast into a prediction game: you bet one exact scoreline per match and earn points. For every match the app picks the scoreline that maximizes expected points under the rules below — which is often not the single most-likely score — so the season total is as high as the model can make it. Finished matches are scored against the pre-kick-off forecast (no peeking); knockouts are scored on the 90-minute result (a shootout counts as a draw).
| Outcome of your bet | Points |
|---|---|
| Exact score | 4 |
| Right winner and goal difference (non-draw) | 3 |
| Right winner, wrong margin | 2 |
| Both draws, different score (e.g. bet 1:1, result 2:2) | 2 |
| Wrong winner / wrong tendency | 0 |
MODEL TRANSPARENCY
HOW IT WORKS
Six concise layers turn ratings and real match data into calibrated forecasts.
10KSIMULATIONS
104MATCHES
ρ −0.15DIXON–COLES
K 15LIVE ELO
σ 40UNCERTAINTY
+50HOST ELO
2. POISSON SCORE GRID
Independent Poisson distributions create every scoreline from 0:0 to 12:12.
P(i:j) = Pois(i, λA) × Pois(j, λB)3. DIXON–COLES CORRECTION
A low-score correction with ρ = −0.15 recalibrates 0:0, 1:0, 0:1 and 1:1 outcomes.
4. LIVE UPDATE
Finished matches update Elo with K = 15. In-play forecasts keep the current score and simulate only the remaining time.
5. MONTE CARLO
Exactly 10,000 runs follow the real group rules, best-third-place allocation and knockout bracket.
6. HONEST SCORING
Accuracy uses only the forecast available before kick-off and compares the full model with pure Elo.