How we model a World Cup match — sample, filters, confidence, limits.

The baseline for every market is the combined record of the 2018 and 2022 World Cups — 128 matches. We chose tournament play specifically because it is the closest reference to 2026: neutral venues, knockout pressure, condensed schedule, mixed-confederation matchups. Club-league data was deliberately excluded.

On top of the historical baseline we apply five adjustments, all of them visible on the market hubs and per-match pages:

• Stage adjustment. Group, Round of 32, last-16, quarter-final, semi-final and final each carry their own historical multiplier (intensity, fouls, shots on target rise into knockouts; goals tighten).
• Possession profile. Possession-dominant sides generate more throw-ins. We nudge throw-in projections up when both teams skew possession-heavy.
• Deep-defender bias. A team that defends in a low block historically draws more fouls in their own third. We raise foul projections for matches with a clear deep-defender side.
• Strength mismatch. When the FIFA-rank gap is wide we apply a capped nudge — large enough to matter, small enough that one rating point does not flip a projection.
• Climate downgrade. For matches flagged as high-heat or high-altitude, tempo-sensitive markets (corners, shots on target, throw-ins) get a confidence downgrade rather than a speculative point estimate.

Every projection is tagged High, Medium or Low. The ladder is colour-coded across the entire site so it is impossible to miss. The rating reflects three things: historical reliability of the market, agreement between the 2018 and 2022 sub-samples, and how many filters fired against the projection.

Throw-ins are currently our most consistent market — the 2018 and 2022 sub-samples agree within roughly two percentage points. Goals markets are noisier and more often land in Medium.

Offsides was tested as a market and dropped: the historical signal did not survive cross-tournament robustness testing and the per-match variance was too wide to support a calibrated number. Publishing it would have meant publishing a guess. We would rather ship five honest markets than six with one we cannot defend.

• Two World Cups is a small sample. We treat it as the best available reference, not a law of nature.
• The 48-team format is untested. Group dynamics, third-place qualification and the new Round of 32 may shift baselines in ways we cannot measure ahead of time.
• Heat and altitude effects are modelled from prior tournament data, not measured for this event. That is why we downgrade confidence rather than push a precise number.
• Squad changes inside the tournament (injuries, suspensions) are not yet modelled at player level. Where it materially matters we annotate it as a caveat on the match page.

After each matchday we will publish a projected-versus-actual review on Insights. The reliability rankings on /stats update as new evidence lands. If a market drifts out of calibration we will say so on the market hub before we say it anywhere else.