Cadillac at Baku: Bottas lost to a car that cooked itself, not to his own hands

On the final lap of the Baku race, Valtteri Bottas locked his front wheels at Turn 1 and, seconds later, buried the car in the barriers at Turn 15. The way he described the machine afterwards is the detail worth keeping: "I felt pretty bad in the car." At the same time, he insisted the differential maps and brake maps were "in a good place." If the control systems were not the problem, what made a driver who has won at Baku lose the car at the last corner? I watched the onboard several times, and the telling part sits in the moments before the lock-up, when the Cadillac MAC-26 began to slide out of the driver's hands in the most literal sense. Cadillac entered 2026 as a new team, running a customer Ferrari power unit, and bet on two veteran drivers: Bottas, ex-Mercedes, and Sergio Pérez, ex-Red Bull. The new team principal is Budkowski. At Baku, Pérez shone as usual — he has an affinity with the Azerbaijani street circuit — while Bottas sat at the bottom of the order. That gap is contaminated by a variable rarely mentioned: Pérez ran Ferrari's "ADUO 2" engine, while Bottas used the old-spec unit. The straight-line deficit reached 6-7 km/h. Two cars from the same team, same number, not quite the same machine. Baku is a brutal street circuit: abrasive asphalt, high kerbs, heavy track temperatures, and long acceleration zones that demand rear-axle stability. Pérez briefly touched the cusp of the top 10 after the restart, then fell back. That is the real ceiling of a first-year team. For a team like that, every lap is a test session, and Bottas was cast as the test driver in the harshest conditions. Budkowski later admitted the team had to take "shortcuts" to meet production deadlines for the first car. That is a valuable admission from a new team principal: the MAC-26 is a learning platform, not a finished machine. A stranger needs no ticket; they open the door with their own feet — but those feet need time to learn the floor. The mechanical side is the real story. Bottas said the team changed dampers and floor before qualifying. The ride improved markedly in qualifying. In the race, everything reverted. The team found the right setup direction; what they lacked was durability over a full race distance. Thermal damper degradation is well understood in racing engineering. Fluid inside a damper heats with every compression, viscosity drops, resistance drops, energy absorption drops. The car loses stability as the race goes on. For a first-year team, this usually signals an under-specified damper or an insufficient thermal model. Bottas put it plainly: "The main limitation we have is still mechanically." Then there is a second blind spot: the team lacks full pressure-tapping coverage on the floor. In some areas, they are "blind" on aerodynamics. Their development loop runs partly open-loop — wind-tunnel and CFD data struggle to correlate with the track. Over the long run, this may be more damaging than the raw pace deficit, because it slows how fast an entire car concept can be corrected. Power steering also degraded through the weekend. Three simultaneous degradation modes on one car — dampers, power steering, and an engine fault — point to systemic immaturity. Then came the crash. The power unit "failed to offer the correct level of engine braking" while superclipping, meaning when the MGU-K harvests heavily. The driver lost a rear-axle stabilising tool right at corner entry. The causal weight shifts from driver to machine. On Pérez's side, he "stayed on the overtake mode for the entire race" to hang with rivals. A driver spending a whole race in overtake mode shows the car cannot run competitively in clean-air conditions. The limit here is structural. I once believed in the timing sheet, until the timing sheet was torn apart by a counter-attack. The Baku classification tells the story of Bottas at the bottom. It does not tell the story of damper fluid losing viscosity, of missing pressure sensors, of an old-spec engine. Framing this as a "Bottas horror show" is a misread, though the opposite should not be absolutised either. Bottas still chose to attack to the end in a car he could not feel. For a team with no points, attacking is defensible for data collection, but it raises incident probability. A survival approach — keep the car, gather data — would have been lower-variance. I learned to bet on the stranger, and lost to understand that I had won; but betting on an unfinished machine is a different wager. On Pérez, the line "I made the difference" needs discounting. He himself warns the next round will be "back to our reality." He was strong at Baku because Baku suits him, and because he had the newer engine. That is circuit affinity plus a hardware advantage, not a step change in form. A power-unit spec asymmetry inside one team is a sporting-fairness issue. If Pérez keeps receiving newer hardware, that is a seed of internal friction, and a question Cadillac needs to answer publicly. Next round, when the track is no longer a street circuit, the real gap will show. Watch two signals: whether Bottas receives a newer engine, and whether Cadillac adds floor pressure-tapping coverage. The next car is expected to change suspension architecture — the team has quietly conceded the current concept is a dead end. With no crowd, I hear the ball breathe; at Baku, that breath was damper fluid boiling in every corner.