ROUND_14 · CIRCUIT DE MADRID
GP de España 2026did we call it right?
Before the race we ran this GP 400 times through our engine, starting from the real qualifying grid, and published the odds. The race has now been run: here is the contrast, misses included.
VERDICT · RACE RUN
What we saidand what happened
Antonelli won, the model's favourite, and a virtual safety car decided it
The model had Kimi Antonelli as the favourite on 21.5% of 400 simulations, ahead of the rest of the grid. He won.
2.43
Avg. position error
across 18 finishers
0.86
Order correlation
1 = exact order
7/8
Front of the race
of the predicted top
9/10
Points scorers
called right
| Actual | Driver | Grid | Predicted | Diff |
|---|---|---|---|---|
| 1 | Kimi AntonelliMercedesStarted 2nd and was the simulation favourite at 21.5%. Pitted under the virtual safety car for Stroll’s stopped Aston and came out ahead of Norris. | 2 | 3.80 | -2.80 |
| 2 | Max VerstappenRed BullStarted 3rd. The model gave him 8.8%, the lowest of the top six. | 3 | 5.20 | -3.20 |
| 3 | Lando NorrisMcLarenHad pole and led the race. He missed the pit entry just as the virtual safety car came out and lost both places there. | 1 | 4.10 | -1.10 |
| 4 | Charles LeclercFerrari | 5 | 4.80 | -0.80 |
| 5 | George RussellMercedes | 6 | 4.00 | +1.00 |
| 6 | Liam LawsonRed Bull | 8 | 8.30 | -2.30 |
| 7 | Franco ColapintoAlpine | 9 | 13.00 | -6.00 |
| 8 | Oscar PiastriMcLaren | 7 | 4.80 | +3.20 |
| 9 | Arvid LindbladRacing Bulls | 10 | 13.10 | -4.10 |
| 10 | Nico HulkenbergAudi | 11 | 15.70 | -5.70 |
| 11 | Esteban OconHaas | 13 | 14.40 | -3.40 |
| 12 | Pierre GaslyAlpine | 14 | 13.20 | -1.20 |
| 13 | Gabriel BortoletoAudi | 12 | 16.80 | -3.80 |
| 14 | Yuki TsunodaRacing Bulls | 15 | 13.60 | +0.40 |
| 15 | Alexander AlbonWilliams | 16 | 15.20 | -0.20 |
| 16 | Oliver BearmanHaasHe set no qualifying time (Haas could not repair his car after his practice crash) and started 22nd. ⚠️ Our preview had him 21st and Stroll 22nd: the two sources we used agreed on that order and the official grid was the other way round. A preview is never re run, so we say it here: the other twenty slots were right. | 21 | 15.70 | +0.30 |
| 17 | Fernando AlonsoAston Martin | 17 | 13.10 | +3.90 |
| 18 | Valtteri BottasCadillac | 19 | 18.30 | -0.30 |
| DNF | Carlos SainzWilliamsStarted 20th after his three place penalty. The model flagged him as the biggest gainer from his grid slot and he did not finish. | 20 | 13.90 | — |
| DNF | Sergio PerezCadillac | 18 | 17.50 | — |
| DNF | Lance StrollAston MartinHis stopped car brought out the virtual safety car that decided the race. | 22 | 18.70 | — |
| DNF | Lewis HamiltonFerrariStarted 4th and retired early with a brake failure, his first retirement of the season. | 4 | 4.30 | — |
Where the model got it wrong
- Franco Colapinto predicted around P13.00, finished 7th (6.00 places off).
- Nico Hulkenberg predicted around P15.70, finished 10th (5.70 places off).
- Arvid Lindblad predicted around P13.10, finished 9th (4.10 places off).
- Fernando Alonso predicted around P13.10, finished 17th (3.90 places off).
- Did not see Nico Hulkenberg scoring.
The pattern is clear: the model orders the front of the race well and underrates the midfield, because each car's performance is set per season and does not capture a given weekend's form. That is the biggest known limitation, and the next one to attack.
Official result: Formula1.com. The probabilities were published BEFORE the race and have not been touched since.
The preview, exactly as published
Favourite
Kimi Antonelli 21.5%
Wins 21.5% of the simulations and finishes on the podium in 54%. Starts P2.
Driver by driver
| Grid | Driver | Win | Podium | Avg. finish |
|---|---|---|---|---|
| P1 | Lando Norris | 18% | 52% | P4.1 |
| P2 | Kimi Antonelli | 21.5% | 54% | P3.8 |
| P3 | Max Verstappen | 8.8% | 28.5% | P5.2 |
| P4 | Lewis Hamilton | 10.5% | 38.5% | P4.3 |
| P5 | Charles Leclerc | 10.5% | 38% | P4.8 |
| P6 | George Russell | 18.3% | 49% | P4 |
| P7 | Oscar Piastri | 12.5% | 39% | P4.8 |
| P8 | Liam Lawson | 0% | 1% | P8.3 |
| P9 | Franco Colapinto | 0% | 0% | P13 |
| P10 | Arvid Lindblad | 0% | 0% | P13.1 |
| P11 | Nico Hulkenberg | 0% | 0% | P15.7 |
| P12 | Gabriel Bortoleto | 0% | 0% | P16.8 |
| P13 | Esteban Ocon | 0% | 0% | P14.4 |
| P14 | Pierre Gasly | 0% | 0% | P13.2 |
| P15 | Yuki Tsunoda | 0% | 0% | P13.6 |
| P16 | Alexander Albon | 0% | 0% | P15.2 |
| P17 | Fernando Alonso | 0% | 0% | P13.1 |
| P18 | Sergio Perez | 0% | 0% | P17.5 |
| P19 | Valtteri Bottas | 0% | 0% | P18.3 |
| P20 | Carlos Sainz | 0% | 0% | P13.9 |
| P21 | Oliver Bearman | 0% | 0% | P15.7 |
| P22 | Lance Stroll | 0% | 0% | P18.7 |
Opinion · our reading
Norris has pole and is not the favourite, and Sainz starts three places behind where he qualified
Two things in the table above need explaining before you read the percentages. First: the starting column is the GRID, not the qualifying order, and it already has the penalties applied. Carlos Sainz qualified 17th and starts 20th after a three place penalty for impeding Valtteri Bottas in Q1. Fernando Alonso moves up to 17th, Sergio Perez to 18th and Bottas himself to 19th. Lance Stroll has a forty place penalty for a new power unit, but he was already last, so it does not move him.
Second: Lando Norris has pole and our model does not make him the favourite. It gives him 18% and Kimi Antonelli, starting second, 21.5%. The simulator has nothing against Norris. It weights the car heavily, and at this point in the season Mercedes leads the constructors with 468 points to McLaren’s 287. Eleven thousandths on Saturday do not cancel that out on Sunday, which is why George Russell starts sixth and still comes out at 18.3%, ahead of the man on pole.
Our favourite number in the table is at the very bottom. Sainz starts 20th because of the penalty and finishes 13.9th on average, a gain of almost seven places and the biggest on the grid. He is the driver the model sees as most undervalued by where he starts, and he is the one racing at home.
One warning of the kind we like to make: this preview was run with the 52 laps our engine had recorded for the Madring, when the race is 57. The figure is now fixed for future previews, but a published preview is never re run. Recalculating it afterwards would make the comparison with the real result meaningless.
How the model works
Track position is weighted by how hard it is to overtake at each circuit, so pole is worth a lot at the Hungaroring and very little at Monza. And even so, the favourite is not the man on pole: Lando Norris stays at 18% and is edged out by Kimi Antonelli on 21.5%, starting 2th. Not even at the circuit where the grid matters most does one starting slot cover the gap between the cars.
A favourite who only wins one race in four is not a broken model: that is what a real Grand Prix looks like. A retirement, a safety car at the wrong moment or a slow stop flips the result, and the engine simulates all of them. That is why 7 drivers have a genuine shot rather than a single name circled in red.
The most interesting column is the average finish: Carlos Sainz starts P20 and averages P13.9, the biggest gain on the grid: the driver the model thinks is most underrated by where he qualified.
The numbers come from 400 full race simulations (57 laps) with car and driver ratings updated to this point of the season. This is not betting advice and not an official prediction: it is what our model said. The odds were published before the race and have not been touched since: the contrast with the real result is above.