Loading the driver page…
Loading the driver page…
Championship
P4
319 pts
Races
24
of 24 · 24 classified
DNFs
0
none
Avg qualifying
4.4
Avg grid
4.4
pit-lane starts excluded
Avg finish
4.5
retirements at their place
| Points | Scored | Expected | Vs expected |
|---|---|---|---|
| Grand prix points | 289 | 254.5 | +34.5 |
| Sprint points | 30 | 26.2 | +3.8 |
| Total | 319 | 280.7 | +38.3 |
| Race gain, all races ± SE | 1.26 places better than the model expected ± 0.45 |
|---|---|
| Classified races only (runs high) | 1.26 places better than the model expected |
| Qualifying and grid gain | 0.03 places worse than the model expected |
George Russell's race pace rank (model) at round 24 is median 11 of 20 (80% range 5–16); over all races this season, the finishes averaged 1.26 places a race better than the model expected (± 0.45).
The rank is lap-time pace, separate from the car, and moves slowly; the gain is how the driver's finishes compared with the model's expectation, which already includes the rating. So a mid-table pace rank can sit beside a large gain: places won through starts, strategy and tyres, or a rating still catching up.
2025 is a development season: the rating model was tuned on 2022–2025; only 2026 was held out.
One race is noise, and a lasting gain can be the driver or the model under-rating them. “All races” includes retirements, mechanical ones too, and is the one to quote; “classified races only” leaves retirements out, so it runs high for nearly everyone (the expectation allows for a retirement).
Best race vs model expectationR17 · Azerbaijan GP
Finished P2 against a pre-race expectation of 6.8: 4.8 places better than the model expected.
Worst race vs model expectationR12 · British GP
Finished P10 against a pre-race expectation of 6.5: 3.5 places worse than the model expected.
Largest rating moveR4 · Bahrain GP
Race rating moved from -0.259% to -0.174% between the round 3 and round 4 snapshots (0.085 slower; lower is faster).
Largest qualifying gap to teammateR14 · Hungarian GP
Qualified 1.551% faster than teammate Antonelli in Q2 (dry qualifying, the segment both set a time in).
Key moments are detected automatically from timing data and race-control messages; they state what happened, not why.
Beta: in the latest independent check (4 races, 28 moments) 25 were judged correct and none stated a false fact; the other 3 were misleadingly worded and have since been fixed. Key moments are being checked again on the last three races of 2026.
The estimated pace vs the mean driver (model), by round, at each event's pre-race snapshot: a percentage of a lap, separate from the car, lower (higher on the chart) is faster, the band is ±1 sd. It is learned mostly from teammate gaps, so it moves with the teammate's; a gap within about 1 sd is not a clear difference. The car's pace is on its own panel: it has a different scale and zero.
| Round | RUS | ANT |
|---|---|---|
| R1 | -0.164 ± 0.198 | 0.260 ± 0.174 |
| R2 | -0.228 ± 0.167 | 0.231 ± 0.158 |
| R3 | -0.259 ± 0.160 | 0.228 ± 0.154 |
| R4 | -0.174 ± 0.156 | 0.208 ± 0.151 |
| R5 | -0.179 ± 0.153 | 0.243 ± 0.149 |
| R6 | -0.100 ± 0.150 | 0.192 ± 0.147 |
| R7 | -0.128 ± 0.149 | 0.207 ± 0.147 |
| R8 | -0.093 ± 0.148 | 0.227 ± 0.146 |
| R9 | -0.093 ± 0.149 | 0.233 ± 0.147 |
| R10 | -0.083 ± 0.148 | 0.228 ± 0.146 |
| R11 | -0.081 ± 0.147 | 0.227 ± 0.146 |
| R12 | -0.052 ± 0.147 | 0.201 ± 0.146 |
| R13 | -0.052 ± 0.147 | 0.191 ± 0.146 |
| R14 | -0.047 ± 0.146 | 0.193 ± 0.145 |
| R15 | -0.082 ± 0.146 | 0.233 ± 0.145 |
| R16 | -0.050 ± 0.146 | 0.197 ± 0.145 |
| R17 | -0.036 ± 0.146 | 0.181 ± 0.145 |
| R18 | -0.043 ± 0.146 | 0.181 ± 0.145 |
| R19 | -0.091 ± 0.145 | 0.219 ± 0.145 |
| R20 | -0.081 ± 0.145 | 0.209 ± 0.145 |
| R21 | -0.044 ± 0.145 | 0.167 ± 0.145 |
| R22 | -0.046 ± 0.145 | 0.172 ± 0.145 |
| R23 | -0.041 ± 0.145 | 0.167 ± 0.144 |
| R24 | -0.008 ± 0.145 | 0.133 ± 0.145 |
| Round | RUS | ANT |
|---|---|---|
| R1 | -0.306 ± 0.174 | 0.300 ± 0.164 |
| R2 | -0.354 ± 0.154 | 0.242 ± 0.150 |
| R3 | -0.310 ± 0.150 | 0.214 ± 0.148 |
| R4 | -0.261 ± 0.148 | 0.195 ± 0.146 |
| R5 | -0.262 ± 0.146 | 0.205 ± 0.145 |
| R6 | -0.190 ± 0.144 | 0.109 ± 0.143 |
| R7 | -0.258 ± 0.144 | 0.163 ± 0.143 |
| R8 | -0.251 ± 0.144 | 0.182 ± 0.143 |
| R9 | -0.254 ± 0.143 | 0.169 ± 0.142 |
| R10 | -0.268 ± 0.143 | 0.166 ± 0.143 |
| R11 | -0.270 ± 0.143 | 0.173 ± 0.143 |
| R12 | -0.267 ± 0.143 | 0.159 ± 0.142 |
| R13 | -0.268 ± 0.142 | 0.179 ± 0.142 |
| R14 | -0.280 ± 0.142 | 0.192 ± 0.142 |
| R15 | -0.282 ± 0.142 | 0.190 ± 0.142 |
| R16 | -0.249 ± 0.142 | 0.153 ± 0.142 |
| R17 | -0.209 ± 0.142 | 0.109 ± 0.142 |
| R18 | -0.211 ± 0.143 | 0.108 ± 0.142 |
| R19 | -0.221 ± 0.142 | 0.119 ± 0.141 |
| R20 | -0.205 ± 0.142 | 0.102 ± 0.141 |
| R21 | -0.135 ± 0.141 | 0.038 ± 0.141 |
| R22 | -0.181 ± 0.142 | 0.081 ± 0.142 |
| R23 | -0.189 ± 0.141 | 0.091 ± 0.141 |
| R24 | -0.188 ± 0.141 | 0.091 ± 0.141 |
| Round | Team | Quali | Vs teammate | Grid | Grid − quali | Finish | Expected | Vs model | Teammate | Pts | Sprint | Telemetry |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| R1 Australian GP | Mercedes | 4 | −0.724% in Q1 | 4 | 0 | P3 | 5.7 | 2.7 places better than the model expected | ANT P4 | 15 | — | laps, round 1 |
| R2 Chinese GP | Mercedes | 2 | −0.417% in Q3 | 2 | 0 | P3 | 4.4 | 1.4 places better than the model expected |
The same car, so the cleanest comparison: every driver who started a grand prix for the same team in the same round, over those rounds only (RUS first).
| Measure | RUS–ANT |
|---|---|
| Weekends together | 24 |
| Qualifying ahead | 21–3 |
| Race ahead, both classified | 17–3 |
| Race ahead, finish order (a DNF counts as behind) | 21–3 |
| Sprint ahead, finish order | 5–1 |
| Points (grand prix and shared sprints) | 319–150 |
| DNFs | 0–4 |
| Season | Teams | Pos | Pts | Races | Avg finish | Gain vs model, all races ± SE | Classified only (runs high) | Race rating ± sd, − = faster | Quali rating ± sd, − = faster | Quali vs teammateQualifying head-to-head against teammates, this driver first. |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026 | Mercedes | 3 | 262 | 20 | 6.7 | 2.11 places worse than the model expected ± 1.41 | 0.48 places worse than the model expected | +0.104% ± 0.157 | −0.109% ± 0.155 | 9–11 |
| 2025 | Mercedes | 4 | 319 | 24 | 4.5 | 1.26 places better than the model expected ± 0.45 |
Norris passed Russell for P3 on lap 18.
Russell passed Leclerc for P2 on lap 39.
Leclerc passed Russell for P2 on lap 43.
Leclerc passed Russell for P3 on lap 38.
Telemetry: LEC lap 38 vs RUS lap 38, R5 · Saudi Arabian GP →
Safety car on laps 46–53; 9 cars pitted under it; from just before it to the end of lap 53, Russell gained the most (P9 to P6) and Hamilton lost the most (P5 to P7).
Russell won with a 13% win probability in the backtest pre-race forecast; the favourite there, Piastri, had 41%.
Russell passed Leclerc for P3 on lap 62.
Russell won with an 11% win probability in the backtest pre-race forecast; the favourite there, Piastri, had 34%.
Norris passed Russell for P2 on lap 34.
| Round | Mercedes |
|---|---|
| R1 | 0.463 ± 0.181 |
| R2 | 0.398 ± 0.149 |
| R3 | 0.421 ± 0.146 |
| R4 | 0.287 ± 0.145 |
| R5 | 0.338 ± 0.144 |
| R6 | 0.352 ± 0.140 |
| R7 | 0.470 ± 0.141 |
| R8 | 0.646 ± 0.140 |
| R9 | 0.822 ± 0.143 |
| R10 | 0.525 ± 0.141 |
| R11 | 0.561 ± 0.141 |
| R12 | 0.606 ± 0.142 |
| R13 | 0.934 ± 0.140 |
| R14 | 0.834 ± 0.140 |
| R15 | 0.901 ± 0.141 |
| R16 | 0.818 ± 0.141 |
| R17 | 0.676 ± 0.141 |
| R18 | 0.478 ± 0.142 |
| R19 | 0.427 ± 0.139 |
| R20 | 0.370 ± 0.140 |
| R21 | 0.265 ± 0.139 |
| R22 | 0.286 ± 0.141 |
| R23 | 0.246 ± 0.138 |
| R24 | 0.332 ± 0.140 |
| ANT P6 |
| 15 |
| P4 · 5 pts |
| laps, round 2 |
| R3 Japanese GP | Mercedes | 5 | −0.271% in Q3 | 5 | 0 | P5 | 5.8 | 0.8 places better than the model expected | ANT P6 | 10 | — | laps, round 3 |
| R4 Bahrain GP | Mercedes | 2 | −0.226% in Q3 | 3 | +1 | P2 | 4.6 | 2.6 places better than the model expected | ANT P11 | 18 | — | laps, round 4 |
| R5 Saudi Arabian GP | Mercedes | 3 | −0.522% in Q3 | 3 | 0 | P5 | 4.6 | 0.4 places worse than the model expected | ANT P6 | 10 | — | laps, round 5 |
| R6 Miami GP | Mercedes | 5 | +0.132% in Q3 | 5 | 0 | P3 | 5.5 | 2.5 places better than the model expected | ANT P6 | 15 | P4 · 5 pts | laps, round 6 |
| R7 Emilia Romagna GP | Mercedes | 3 | −0.578% in Q2 | 3 | 0 | P7 | 4.8 | 2.2 places worse than the model expected | ANT DNF · Unknown | 6 | — | laps, round 7 |
| R8 Monaco GP | Mercedes | 14 | −0.519% in Q1 | 14 | 0 | P11 | 8.1 | 2.9 places worse than the model expected | ANT P18 | 0 | — | laps, round 8 |
| R9 Spanish GP | Mercedes | 4 | −0.365% in Q3 | 4 | 0 | P4 | 6.2 | 2.2 places better than the model expected | ANT DNF · Unknown | 12 | — | laps, round 9 |
| R10 Canadian GP | Mercedes | 1 | −0.689% in Q3 | 1 | 0 | P1 | 4.6 | 3.6 places better than the model expected | ANT P3 | 25 | — | laps, round 10 |
| R11 Austrian GP | Mercedes | 5 | −0.786% in Q3 | 5 | 0 | P5 | 6.3 | 1.3 places better than the model expected | ANT DNF · Unknown | 10 | — | laps, round 11 |
| R12 British GP | Mercedes | 4 | −0.404% in Q3 | 4 | 0 | P10 | 6.5 | 3.5 places worse than the model expected | ANT DNF · Unknown | 1 | — | laps, round 12 |
| R13 Belgian GP | Mercedes | 6 | −0.348% in Q1 | 6 | 0 | P5 | 7.3 | 2.3 places better than the model expected | ANT P16 | 10 | P12 | laps, round 13 |
| R14 Hungarian GP | Mercedes | 4 | −1.551% in Q2 | 4 | 0 | P3 | 6.8 | 3.8 places better than the model expected | ANT P10 | 15 | — | laps, round 14 |
| R15 Dutch GP | Mercedes | 5 | −0.259% in Q2 | 5 | 0 | P4 | 6.7 | 2.7 places better than the model expected | ANT P16 | 12 | — | laps, round 15 |
| R16 Italian GP | Mercedes | 6 | −0.054% in Q3 | 5 | −1 | P5 | 6.7 | 1.7 places better than the model expected | ANT P9 | 10 | — | laps, round 16 |
| R17 Azerbaijan GP | Mercedes | 5 | +0.347% in Q3 | 5 | 0 | P2 | 6.8 | 4.8 places better than the model expected | ANT P4 | 18 | — | laps, round 17 |
| R18 Singapore GP | Mercedes | 1 | −0.423% in Q3 | 1 | 0 | P1 | 4.7 | 3.7 places better than the model expected | ANT P5 | 25 | — | laps, round 18 |
| R19 United States GP | Mercedes | 4 | −0.309% in Q3 | 4 | 0 | P6 | 6.1 | 0.1 places better than the model expected | ANT P13 | 8 | P2 · 7 pts | laps, round 19 |
| R20 Mexico City GP | Mercedes | 4 | −0.110% in Q3 | 4 | 0 | P7 | 6.1 | 0.9 places worse than the model expected | ANT P6 | 6 | — | laps, round 20 |
| R21 São Paulo GP | Mercedes | 6 | +0.369% in Q3 | 6 | 0 | P4 | 5.7 | 1.7 places better than the model expected | ANT P2 | 12 | P3 · 6 pts | laps, round 21 |
| R22 Las Vegas GP | Mercedes | 4 | −2.725% in Q1 | 4 | 0 | P2 | 5.1 | 3.1 places better than the model expected | ANT P3 | 18 | — | laps, round 22 |
| R23 Qatar GP | Mercedes | 4 | −0.230% in Q3 | 4 | 0 | P6 | 5.0 | 1.0 place worse than the model expected | ANT P5 | 8 | P2 · 7 pts | laps, round 23 |
| R24 Abu Dhabi GP | Mercedes | 4 | −0.421% in Q2 | 4 | 0 | P5 | 5.3 | 0.3 places better than the model expected | ANT P15 | 10 | — | laps, round 24 |
Vs teammate (qualifying): the gap in the last segment both completed (Q3, Q2 or Q1 time), in % of the teammate's lap; negative = faster. Grid − quali: + means a start further back than qualified (usually a penalty). Expected: the model's expected finish before the race; vs model: expected minus finish (+ = better than expected). Points: grand prix only; sprint points in the sprint column. Telemetry opens the lap comparison with both teammates selected.
| 1.26 places better than the model expected |
| −0.008% ± 0.145 |
| −0.188% ± 0.141 |
| 21–3 |
| 2024 | Mercedes | 6 | 245 | 24 | 6.8 | 0.56 places worse than the model expected ± 1.01 | 0.09 places better than the model expected | −0.098% ± 0.250 | −0.184% ± 0.249 | 19–5 |
| 2023 | Mercedes | 8 | 175 | 22 | 8.1 | 1.39 places worse than the model expected ± 1.23 | 0.22 places better than the model expected | +0.037% ± 0.245 | −0.028% ± 0.243 | 11–11 |
| 2022 | Mercedes | 4 | 275 | 22 | 5.0 | — | — | −0.068% ± 0.241 | −0.031% ± 0.237 | 9–13 |
2022 has no forecasts before each session, so there is no expected finish, no gain vs model expectation and no expected points for it.