Loading...
Powered by Elo: pairwise driver ratings across 75+ years and 1,100+ races. Era-normalized, teammate-adjusted, updated after every race.
We adapt the pairwise rating system originally designed for chess to multi-competitor races: after every race from 1950 to present, each finisher is compared head-to-head against every other finisher, and ratings move up or down based on expected finishing order versus what actually happened. Every driver starts at a baseline of 1500.
Average ratings have drifted upward over F1's 75+ year history simply because the field and the sport have changed, not because drivers got better relative to their contemporaries. We strip that drift out by re-centering each era's mean rating to 1500, so a peak rating from the 1960s is directly comparable to one from the 2020s.
A driver's single best rating can be a one-season outlier. Sustained Greatness blends 60% peak rating with 40% the average of their five best seasons, rewarding drivers who stayed at the top for years over one exceptional but brief peak. It requires a minimum of 50 races for statistical significance.
After every race, once official results are in: every finisher's rating is recalculated against every other finisher from that session.