Cadillac's Perez-Bottas Deal: What a Rookie Team Actually Bought

16 Grand Prix victories. More than 500 race starts. And a team that has never completed a single official racing lap.

On 26 August 2026, Cadillac confirmed Sergio Perez and Valtteri Bottas as its driver pairing for the team's debut season in 2026. Within four hours, the F1 paddock split in two: one half called it a prudent move by a newcomer, the other called it a mobile museum. I opened a spreadsheet before I opened the comment section. That habit has followed me for 44 years.

What deserves analysis here is not the age of the two drivers. It is the question of what Cadillac actually bought. If the team from Michigan paid for raw speed, it made a mistake. If it paid for data, it may have got it right, and that is an entirely different calculation, one the F1 transfer market rarely bothers to solve until the results arrive.

At 60, I no longer believe in luck, only in numbers that have not yet had time to speak.

The 2026 regulatory cycle is the biggest structural change since 2026. The power unit keeps the 1.6-litre V6 block but loses the MGU-H electric turbocharger entirely, while electrical power rises to roughly 350 kW, nearly three times the 120 kW of the previous generation, and fuel moves to a fully sustainable blend. Aerodynamically, the cars become smaller and lighter, with active aerodynamics replacing fixed wings.

For a brand-new team, this is the hardest possible brief. Cadillac enters 2026 without a single line of its own correlation data. No tyre model. No multi-lap circuit map. No degradation history. The team runs a customer power unit in its early phase before General Motors brings its own in 2029, and it must borrow the data others have already accumulated.

What stands out is that the team finishing last in the constructors' standings receives the largest aerodynamic testing allowance under the FIA's sliding scale, while the champion is capped at the lowest level. Cadillac therefore gets more wind tunnel time than anyone. But testing time only has value when you know what to test. That is precisely the gap an experienced driver fills.

In that situation, the most important data source does not sit in the wind tunnel. It sits in the cockpit.

This is where the story leaves sentiment behind. Three historical markers need to be placed side by side. In 2026, three new teams joined the grid: Virgin, HRT and Lotus. All three started with inexperienced drivers, and all three finished the season on zero points. In 2026, Haas debuted with Romain Grosjean, by then approaching 90 starts, and Esteban Gutierrez. Haas scored 29 points in its first season, finished eighth in the constructors' championship, and Grosjean came home sixth in Melbourne. In 2026, Haas switched to a rookie pairing and ended the year on zero points.

The gap between those two data sets is 29 points. That gap only means something if you understand the variable standing behind it.

The real variable is the volume of feedback a driver can transfer to engineers during the first 18 months of a regulatory cycle. Without correlation data, a team depends on subjective description: where the car is unbalanced, how slow the braking into corners is, at which point the energy recovery system intervenes. A driver who has crossed the 2026, 2026 and 2026 regulatory boundaries already carries a vocabulary of comparison. Bottas debuted in 2026 and has crossed three cycles. Perez debuted in 2026 and crossed exactly three. Both have driven cars with an MGU-H, and both know precisely where its absence will feel different.

In the 2026 cycle, the balance between mechanical braking and energy recovery changes completely once the MGU-H disappears. A driver able to say exactly where the car lacks braking, and in which part of the lap, can shorten the development loop from weeks to days. That is a metric that never appears on the timing screens, but it appears in the number of set-up direction changes between FP1 and FP3.

Based on my experience following races since 2026, I have seen four regulation changes, and each time the winning team was the one that misunderstood its own car the least during the first six months.

The transfer market is a contest in which whoever prices correctly wins.

On valuation, both drivers were free agents. Perez left Red Bull after the 2026 season; Bottas left Sauber at the same point. No transfer fee, no release clause, no bidding war. Cadillac paid for an asset whose market valuation had fallen sharply, while its utility value inside a new team was at its peak. That is the logic of a buyer who knows what he needs, not the logic of a buyer who wants new stock.

But correlation is not causation, and this is where I have to argue against myself.

The Haas 2026 case is usually told as proof of the power of experience. Strip that layer away. The Haas VF-16 ran a Ferrari engine, a Ferrari gearbox, Ferrari suspension, and was designed by Dallara. Those 29 points came from an off-the-shelf technical package, not from a veteran driver. Attribute the whole result to Grosjean and you are reading data with your ears rather than your eyes.

And there is a variable Cadillac itself cannot control: neither Perez nor Bottas held a race seat through the whole of 2026. That is 12 months without collisions, without qualifying pressure, without reading tyres in shifting temperatures. History does not support an easy return. Michael Schumacher came back in 2026 after three years away and needed more than two seasons to reach a podium. Fernando Alonso returned in 2026 and adapted faster, but he returned to a midfield team with no championship expectation.

The empty circuits of 2026 exposed one truth: many things we called character were just noise. When the noise vanished, people finally saw what was skill and what was reflex fed by habit. What Cadillac really needs to know is whether it bought skill, or only the memory of a regulatory cycle that no longer exists.

The signal to watch is not the grid position. It sits in the long runs across the first five rounds, and in the tyre degradation of the two Cadillacs on the same compound. If the gap between Perez and Bottas stays under a tenth of a second per lap, the data hypothesis holds. If it exceeds three tenths, the price of experience was set wrongly.

Data is never in a hurry. People always are.