Skip to content
Reference Lap

Liam Lawson's 2026: 82 grand prix points where our model expected 32.8

A year ago Liam Lawson finished about where our model expected. This season no driver has beaten its expectation by more. How the season built up: race day rather than qualifying, the record against Lindblad in the same car, the weeks at Red Bull, and a pace rating that is still catching up.

Reference LapWatch the video

Card: Liam Lawson, 82 grand prix points scored against 32.8 expected by the model (backtest expectations)

Liam Lawson has scored 82 grand prix points this season. Add up what our model expected before each race and you get 32.8.

Race by race, that's 2.54 places ahead of the model's expectation on average over 20 races, more than any of the 22 drivers, and it has carried Lawson to 9th in the championship on 89 points. (For these rounds, the expectations come from the model's held-out backtest.)

A year on

In 2025, Lawson drove 24 races for Red Bull and Racing Bulls, finished 14th on 38 points, and ended up almost exactly where the model expected: -0.09 places a race. Teammates out-qualified Lawson 18 times to 6.

This season the qualifying record against teammates is level, 10 to 10, and the average finish is up from 12.79 to 9.

Season Team Races Championship Points Average finish vs expected, per race Qualifying vs teammates
2023 AlphaTauri 5 20th 2 12.2 +2.91 1–4
2024 Racing Bulls 6 21st 4 13.5 +0.51 0–6
2025 Red Bull / Racing Bulls 24 14th 38 12.79 -0.09 6–18
2026 Racing Bulls / Red Bull 20 9th 89 9 +2.54 10–10

Places a race vs model expectation, over all races including retirements; this season's expectations come from the model's held-out backtest. Short seasons say little: a single race's number is noise.

Race day, not qualifying

The gains don't come on Saturday. In qualifying Lawson sits close to the model's expectation, +0.19 places a session, with an average qualifying position of 10.85. In the races it's +2.54, with an average finish of 9. Starts, strategy and tyres all win places on race day; our data shows where the places come, not which of those it is.

It happened most weekends. The standout was Monaco: 5th, where the model expected 12.7. In Mexico City, Lawson started 22nd and finished 7th.

And in Brazil, the latest race, Lawson started 3rd and was up to second early on. From the race's key moments: Norris passed Lawson for P2 on lap 6. Lawson finished 7th.

Not every weekend: in Miami, the model expected 12.8 and the result was a DNF.

Chart: Liam Lawson's finish against the model's expected finish, race by race

Round Grand prix Team Finish Expected vs expected
4 Miami Grand Prix Racing Bulls DNF 12.8 -7.19
6 Monaco Grand Prix Racing Bulls 5th 12.7 +7.68
19 Mexico City Grand Prix Racing Bulls 7th 12.6 +5.57
20 Brazilian Grand Prix Racing Bulls 7th 10 +3.03

Expected finish: our model's expectation before each race, from its held-out backtest. Positive = finished better than expected. Every race is on the driver page.

Same car: Lawson and Lindblad

The fairest yardstick is the other driver in the same car. In the 17 races they shared a car, Lawson finished ahead of Arvid Lindblad 13 times to 4 and outscored Lindblad 75 points to 32. Qualifying was closer: 10 to 7.

Bars: Liam Lawson and Arvid Lindblad head-to-head as teammates

As teammates Lawson Lindblad
Ahead in qualifying 10 7
Ahead at the finish 13 4
Ahead in sprints 4 1
Points 75 32
Retirements 1 2

17 races in the same car, 2026.

On raw pace they are hard to tell apart: the gap in their race ratings, measured in the same car, is -0.09 ± 0.103 % of lap time. The difference shows up on Sunday.

On Reference Lap: Compare Lawson and Lindblad

Rounds 12 to 14: Red Bull

Mid-season, from the Dutch Grand Prix, Lawson moved to Red Bull beside Max Verstappen for 3 races (the Netherlands 7th, Italy 14th, Spain 6th), and 14 points to Verstappen's 36. Then back to Racing Bulls from round 15, and the results kept coming:

  • Bahrain Grand Prix in Malaysia: 11th on the grid, 7th at the flag
  • Singapore: 8th on the grid, 7th at the flag
  • Austin: 9th on the grid, 9th at the flag
  • Mexico City: 22nd on the grid, 7th at the flag
  • Brazil: 3rd on the grid, 7th at the flag

The pace rating

Our driver ranking rates race pace apart from the car, and it is slower to move than a results sheet. Before the Brazilian Grand Prix, it had Lawson 15th, somewhere from 10th to 19th of 21.

But it has been moving the right way: Lawson's race rating edged toward faster at every snapshot from round 16 to round 20, from +0.09 ± 0.113 to +0.014 ± 0.113 % of lap time against the mean driver (lower is faster; still within the uncertainty).

Chart: race pace ratings with their uncertainty, Driver ranking (model)

Rank Range Driver Team Race rating ± sd
13 7–19 Lewis Hamilton Ferrari -0.037 ± 0.146
15 7–20 Oscar Piastri McLaren -0.004 ± 0.17
15 9–20 Franco Colapinto Alpine +0.013 ± 0.127
15 10–19 Liam Lawson Racing Bulls +0.014 ± 0.113
16 9–20 Gabriel Bortoleto Audi +0.022 ± 0.135
16 8–21 Lance Stroll Aston Martin +0.037 ± 0.172
18 13–21 Arvid Lindblad Racing Bulls +0.104 ± 0.118

Driver ranking (model), race pace, before the Brazilian Grand Prix: the drivers around Lawson. Rank is the median over 10000 draws from the model; the range is where most of them land. Rating: % of lap time against the mean driver, lower = faster.

Whatever the rating says next, the season's results sheet already reads: 89 points, 9th in the championship, and more places gained on our model's expectation (from its backtest) than any other driver.

Watch this spotlight on YouTube, or as a Short.

Explore: Liam Lawson · Arvid Lindblad · Compare Lawson and Lindblad · Driver ranking (model)


Data: Jolpica F1 API (CC BY-NC-SA 4.0, https://jolpi.ca); OpenF1 (CC BY-NC-SA 4.0, https://openf1.org).

This post is derived from that data and is offered under the same licence, CC BY-NC-SA 4.0, for non-commercial use only: no ads, sponsorships or affiliate links.

This site is unofficial and not associated with Formula 1 companies. F1 and related marks are trademarks of Formula One Licensing B.V. This post is likewise unofficial and not affiliated with them.

Data version 20261117T082657Z.

The companion video's narration is a synthetic voice (text-to-speech).