Driver evaluation in F1 is emotional by default — fans remember passes, mistakes, and radio messages. Performance analysts start elsewhere: rolling windows of classified finishes, pace relative to machinery, and repeatability. Debrief surfaces four core metrics derived from historical race results.
Race pace score
Race pace converts mean finishing position into a 0–100 scale (inverted so higher is better). It rewards regular top finishes rather than occasional wins mixed with DNFs. A driver averaging P4 across 12 races scores higher than one with two wins and eight P15s.
0–100
Pace scale
Higher = better mean finish position
Consistency
Consistency uses the coefficient of variation of finishing positions over at least three races: 100·exp(−CV). Drivers who land P5–P7 every week beat volatile performers for championship points even when peak pace is similar.
Reliability
Reliability is the classified-finish rate: finishes counted toward the championship divided by starts. Mechanical DNFs, accidents classified as non-finish, and exclusions reduce the score. It separates driver-induced retirements from car failures only when status codes allow.
Teammate delta
Teammate delta is mean finishing-position difference versus the other driver in the same car, shared races only. Negative delta (finishing ahead on average) in equal machinery is one of the strongest signals of outright performance.
| Metric | Question it answers | Blind spot |
|---|---|---|
| Pace | How fast are they on average? | Ignores grid and strategy |
| Consistency | How repeatable? | Penalises bold risk-taking |
| Reliability | Do they finish? | Mixes car and driver DNFs |
| Teammate delta | Are they beating their benchmark? | Needs comparable machinery |
Key takeaways
- Use rolling windows, not single races, to judge form.
- Teammate delta is the fairest cross-driver comparison in the same team.
- Consistency wins championships when pace gaps are small.
- Reliability is a points metric — not just a mechanical one.