WC Intel Hub

Transparency

Prediction methodology

Our predictions combine a compact game-style team profile, recent form, expected goals, and tournament simulation. The model runs simulated outcomes for upcoming matches to estimate group and knockout probabilities. Predictions are estimates, not certainties.

What data is used

Four profile ratings per team (overall, attack, midfield, defence) plus recent form, tournament schedule and confirmed results. We do not use private information such as injuries, line-ups or training data — and the model never pretends to.

How probabilities are calculated

Expected goals for each side come from attack, midfield and defence matchups adjusted for form. A Poisson model turns expected goals into a probability for every scoreline, which gives win/draw/win odds and the most likely score. A Monte Carlo engine then plays every remaining match thousands of times and counts how often each team finishes 1st, 2nd, or qualifies among the 8 best third-placed teams.

Why probabilities change after results

Each final score removes uncertainty: completed matches become fixed inputs and the simulation re-runs on what remains. A single upset can move a group's qualification odds by 20 points or more — that is the tournament, not a bug.

How the bracket is projected

The Round of 32 starts from the published match-schedule slots for group winners, runners-up and eligible third-place pools. While the group stage is in progress, projected qualifiers are pulled from the current simulated table and best-third race.

Limitations

Ratings are simplified, knockout football is high-variance, and 90 minutes can ignore any spreadsheet. Treat every number as a structured opinion with error bars.

Predictions are statistical estimates for entertainment and analysis only. They are not betting advice.