Documentazione
Come funziona il modello
Il riferimento metodologico dietro ogni probabilità. Ogni documento viene presentato come sommario e campione.
How we make predictions
How match previews work
Every fixture page on the site opens with a detailed match preview — predicted formations, key players, strategic factors, and expected scorelines. Here's how we generate them and what data feeds into
How our 2026 World Cup prediction model works
Our 2026 FIFA World Cup forecasts come from a statistical prediction model that blends three approaches — an Elo rating system, a Dixon-Coles Poisson goals model, and a hierarchical Poisson model — in
How We Rate Team Strength
How do we figure out which teams are strongest and most likely to win any given match? Our model considers two main approaches. **Goal-based models** simulate how many goals each side will score, then
In-tournament calibration
Rolling Brier score, log-loss, and ECE measured against observed 2026 World Cup match outcomes. Empty before kickoff; populates as fixtures are played.
What we predict and how
For every prediction target — match outcomes, goal totals, scorelines, individual player events — there's a standard modelling approach and a set of input variables. This page catalogues all of them i
Why trust these numbers
A probability publication is a credibility game. Anyone can publish numbers; the question is whether those numbers track outcomes once the matches finish. This page collects the discipline, the archit
Players and managers
Expected goals, assists, and player actions
In football analytics, "expected value" refers to a family of metrics that assign a numerical value to events, game states, or players based on the probability of contributing to a goal (or preventing
Manager experience and history
A head coach's experience matters — tournament pedigree, tenure with the national team, and whether they rose through the federation's youth system. This page documents the manager profile dataset for
Measuring player quality
How do you put a number on how good a footballer is? This page covers the approaches we use beyond match actions — rating players' overall skill level, accounting for position, age, and the league the
What shapes a match
Home advantage, fatigue, and other match factors
Beyond team strength and player quality, match outcomes are shaped by factors like home advantage, travel fatigue, rest days between games, weather, and altitude. Each effect is small on its own, but
How each team plays
Is a team high-pressing or deep-defending? Do they build up patiently or play direct? This dataset captures each WC 2026 team's tactical fingerprint across 12 playing-style dimensions.
Squad workload and minutes played
How many club minutes has each nation's likely starting XI played this season? This dataset captures the workload each team's players carry into the tournament.
Travel distances and rest days
How far does each team travel to reach their match venue, and how many days of rest do they get between games? This page documents the dataset that quantifies these logistical factors for every WC 202
What data goes into the model
Our predictions draw on dozens of data sources — from 49,000 historical international match results to individual player statistics, manager profiles, and team playing styles. This page is the complet
Who takes corners, free kicks, and penalties
Set pieces decide tournaments — roughly a third of World Cup goals come from dead-ball situations. This page explains how we identify each team's primary corner, free-kick, and penalty takers.
Behind the scenes
Building the player database
To predict the World Cup, we need data on every player who might take the field. Here's how we identify, collect, and structure player-level data for all 48 qualified nations.
Data sources at a glance
A quick-reference table of every data source, what it covers, and how accessible it is. Companion to the full narrative in "Where our data comes from".
Our research roadmap
This is the plan we followed to build the prediction model from scratch — mapping the research landscape, building a baseline, and iterating. The pipeline is now live; this page explains the thinking
What football can learn from other sports
Most of the statistical techniques we use weren't invented for football — they came from basketball, baseball, tennis, and American football. Here's what each sport contributed and how we've adapted t
Where our data comes from
The quality of any prediction depends on the data behind it. This page maps every data source we use — from free public archives to commercial feeds — and explains what each one provides.
Why we research what we research
Each area of our research was chosen for a specific reason. Here's why each topic matters and how it connects to producing reliable predictions.