How good are these forecasts?
Every forecast scored on seasons the model never saw while its settings were chosen.
A rider's rating is two parts added together: the rider, and the bike, which is the team's strength that season. Teammates share the bike, so the gap between them is what separates rider from machine; satellite teams are rated apart from the factory teams on the same make.
The Saturday sprints (since 2023) start from the same grid as the Grand Prix, so they train and are scored in the model like any race; on the held-out seasons that beat leaving them out. The title simulation plays every remaining race and sprint with today's strengths and the real points (25-20-16-13-11-10-9-8-7-6-5-4-3-2-1, and 12 down to 1 for a sprint).
Crashes and retirements count as finishing behind every classified rider, whatever the cause.
Injured riders' replacements start from the average; the model doesn't know who is riding hurt.
Before the entry list is out, the next race's field is the last race's.