Austin GP 2026: practice & qualifying telemetry and race prediction for Verstappen, Antonelli and Leclerc
All charts are interactive: hover for values, drag to zoom, click legend entries to hide a driver.
Before tonight’s United States Grand Prix (21:00 CET) I ran the whole weekend’s timing and car telemetry through my prediction model. This post covers the three drivers the model rates highest for the win: Max Verstappen (Red Bull), Kimi Antonelli (Mercedes) and Charles Leclerc (Ferrari).
All lap times below exclude laps deleted for track limits (taken from race-control messages). That matters a lot at Austin this year: race control deleted 55 lap times in FP1, 96 in FP2, 32 in FP3 and 7 in qualifying. Data source: the F1 live-timing archive (via FastF1’s API), including car data and X/Y positions.
Best laps per session
| Session | Verstappen | Antonelli | Leclerc |
|---|---|---|---|
| FP1 | 1:36.136 (+0.084) | 1:36.533 (+0.481) | 1:36.052 (P1) |
| FP2 | 1:35.480 (+0.189) | 1:35.333 (+0.042) | 1:35.291 (P1) |
| FP3 | 1:35.061 | 1:35.196 (+0.135) | 1:35.127 (+0.066) |
| Qualifying | 1:34.214 (P2) | 1:34.035 (pole) | 1:34.198* (P4) |
*Leclerc’s 1:34.198 was set in Q2. His best Q3 lap was 1:34.267, which put him P4. Verstappen’s last Q3 lap (1:34.482) and Hadjar’s (1:34.363) were both deleted at Turn 19, so Verstappen’s front-row time came from his first Q3 run.
Leclerc topped both Friday sessions, and Verstappen was the quickest of the three in FP3. In qualifying, Antonelli found the extra 0.18 s when it counted.
Sectors and speed trap
Best qualifying sectors (s):
| S1 | S2 | S3 | |
|---|---|---|---|
| Verstappen | 25.254 | 37.475 | 31.481 |
| Antonelli | 25.292 | 37.185 | 31.428 |
| Leclerc | 25.253 | 37.430 | 31.436 |
Speed trap maximum (km/h):
| FP1 | FP2 | FP3 | Q | |
|---|---|---|---|---|
| Verstappen | 322 | 331 | 322 | 326 |
| Antonelli | 328 | 328 | 328 | 330 |
| Leclerc | 309 | 333 | 324 | 318 |
Antonelli won qualifying in sector 2, where he was almost 0.3 s up on Verstappen. In qualifying Leclerc’s Ferrari was among the slowest cars through the trap (318 km/h, 20th in the field). That points to a higher-downforce set-up, and it makes him harder to defend against on the long back straight.
Qualifying lap telemetry
On the back straight, Antonelli and Verstappen hit 341 and 339 km/h on their fastest laps, while Leclerc reached 326 km/h. Leclerc gets the time back in the slow corners. The delta line comes from integrating speed over distance (aligned via the cars’ X/Y positions) and is scaled to the official lap times, so treat it as approximate (about ±0.1 s mid-lap). Zoom with the range slider; hovering shows all three cars at the same point of the lap.
Minimum speeds at the slow and medium corners the script detected automatically (km/h, distance from the line):
| Corner at | Verstappen | Antonelli | Leclerc |
|---|---|---|---|
| ~615 m (T1 area) | 83 | 84 | 89 |
| ~1830 m | 145 | 147 | 154 |
| ~2510 m | 76 | 75 | 78 |
| ~3710 m | 88 | 89 | 90 |
| ~3940 m | 98 | 99 | 101 |
| ~5260 m | 100 | 103 | 100 |
The Ferrari carries the most minimum speed through nearly every slow corner, and the Mercedes and Red Bull are faster on the straights. That is the classic downforce-versus-drag trade-off.
Who was fastest where
The map at the top splits the lap into 25 equal mini-sectors and colours each one by the fastest of the three on their best qualifying lap. Antonelli wins 11, Verstappen 8 and Leclerc 6. Hover any segment for all three times, or press ▶ to replay the three laps on the track from X/Y position data. Mini-sector splits are reconstructed from ~4 Hz telemetry, so close calls (a few hundredths) can flip either way.
Long-run pace
There is honestly little to show here. Once deleted laps, traffic and cool-down laps are filtered out, the timing data contains few clean 5+ lap runs for these three:
- Verstappen, FP1: 5-lap run averaging 1:41.87 with about +0.38 s/lap degradation (probably fuel and tyre effects mixed together).
- Leclerc, FP2: 5-lap run averaging 1:40.86 with only about +0.02 s/lap.
- Antonelli: no clean run of 5+ laps passed the filter.
Tyre compound wasn’t available for these stints in the timing feed, so read these numbers with care.
Conditions
The track was hot on Friday, peaking at about 43 °C with air around 28-30 °C, and dry all weekend. Qualifying ran at 36.5-40.7 °C. Tyre degradation in the heat is one of the main race risks.
Starting grid
Antonelli starts on pole, with Verstappen, Hadjar, Leclerc and Hamilton behind him. Lindblad drops 5 places to P14, Colapinto takes a 10-place penalty and starts P21, and Bortoleto starts from the back. The grid graphic is schematic: slots are spaced 8 m apart behind the line.
Prediction
The model combines practice pace, qualifying, car/team strength, pit crew and reliability. It runs with its default settings (heat feature off).
| Driver | Grid | Win | Podium | Top 10 |
|---|---|---|---|---|
| Verstappen | P2 | 28.9 % | 60.9 % | 84.1 % |
| Antonelli | P1 | 28.5 % | 64.9 % | 94.7 % |
| Leclerc | P4 | 12.3 % | 40.5 % | 83.9 % |
| Hamilton | P5 | 9.1 % | 35.1 % | 89.5 % |
| Hadjar | P3 | 8.6 % | 33.2 % | 87.7 % |
Predicted top 10: 1. Verstappen, 2. Antonelli, 3. Leclerc, 4. Hamilton, 5. Hadjar, 6. Russell, 7. Norris, 8. Piastri, 9. Lawson, 10. Hülkenberg.
Verstappen and Antonelli are essentially a coin flip for the win. Antonelli is the likelier podium finisher because of his lower retirement rate.
Key risks:
- Safety cars: these reshuffle strategy, and the model handles them poorly.
- Track limits: deletions and penalties are likely given this weekend’s record (Turn 19 especially).
- Tyre wear at ~41 °C: long-run data is thin, so a team that manages the heat better could beat its qualifying pace.
- First-lap incidents: Turn 1 at COTA is wide and uphill, so the front row is exposed.
This is a model forecast, not a certainty. In backtests it picked the winner about 61 % of the time and had the real winner in its top 3 about 94 % of the time. I’ll score it against the actual result after the race.