Calibrated probability
A probability that matches reality in the long run: events given a 20% chance happen about 20% of the time.
A model can be confident and wrong. Calibration is the check that its stated probabilities mean what they say. If RaceModel gives ten different drivers a 10% win chance across a season, about one of them should win.
RaceModel calibrates its podium and DNF probabilities on held-out races and publishes the outcome, so a probability on a race page can be taken at face value rather than as a vague signal of confidence.