How good are these forecasts?
Every forecast scored on seasons the model never saw while its settings were chosen.
A driver's rating is two parts added together: the driver, and the car, which is the team's strength that season. Teammates share the car, so the gap between them is what separates driver from machinery.
The title odds play every remaining Grand Prix and sprint with today's strengths and the real points (25-18-15-12-10-8-6-4-2-1, and 8 down to 1 for a sprint), on top of the points already scored.
Retirements count as finishing behind every classified car, whatever the cause, so a mechanical failure costs the driver as well as the car.
Grid penalties and the track's character are not in the model yet. Rain is only a general loosening of the odds (tested on the model page), not a driver's wet-weather skill.
Before the entry list is out, the next race's field is the last race's.
The title odds were replayed race by race over 2014–2025, each from a fit that saw only earlier races. When the favourite was 95% or more (78 times) they won every time, and at 85–95% (43 times) they won 95%. In closer fights the favourite was too short: at 50–85% (136 times) they won 55% against about 68% forecast. A chance of retiring in each race, a drift in pace from race to race, sharper per-race luck and wider season-long doubt were each tested; none scored the champion better over those seasons, so the simulation is unchanged.