Cadillac Bets on Baku: One Retirement, One Promise and a Test of Reliability

On the Madrid circuit, roughly a third of the way into the opening stint, Sergio Pérez's car began moving against the tide. From a modest starting position, the Mexican driver cut through the gaps, picked off a few rivals, and at some point found himself sitting in the mirrors of both Williams and Racing Bulls. This is the kind of sequence every chief engineer at a brand-new team hopes for: evidence that their car is not out of step with the midfield.

Then the electrics arrived.

No contact, no driver error, no controversial call from race control. Simply an electrical system failure, and the car had to be parked. Pérez left Madrid without points. Cadillac left Madrid without points. Last place in the constructors' championship remained exactly where it was.

The interesting detail lies elsewhere. During that brief opening stint, the car was running level with teams that have accumulated years of experience. At the same time, the promise of an upgrade package for Baku was issued before anyone had confirmed whether the car could be trusted at all.

The structure of this story is familiar enough to be predictable: poor results, good messaging, and a fixed date chosen so that hope has somewhere to land. The problem is that, for a new team, the ordering of two variables — pace and reliability — is what decides a season, not the ordering of press releases.

Context: a new team, two veteran drivers and a milestone not yet earned

In Formula 1 history, new teams arriving on the grid are nothing rare. In 2026, three teams joined at once — Virgin, Lotus and HRT — and all three began with seasons of almost no points, no pace, and no prospects. It took many races, sometimes many years, before any of them touched the top ten. In 2026, Haas did the opposite: it scored on debut in Australia, thanks to an entirely different model — buying the maximum permitted components from a larger team, outsourcing the chassis, and pouring internal resources into trackside operations rather than building everything from scratch.

The distance between those two models is not money. It is which team understood more clearly that, in a debut season, reliability is the asset that pays, while raw pace is only an asset awaiting confirmation.

Cadillac chose a third path: building almost from the ground up, but equipping itself with a driver pairing that has won races — Sergio Pérez and Valtteri Bottas. This is a highly symbolic staffing decision. Pérez has won multiple grands prix, spent several seasons as the number two at a leading team, and brings with him an enormous fan market in Mexico and across Latin America. Bottas was once teammate to a multiple champion, has won races, and is one of the most technically grounded drivers of the modern era.

Putting those two into a rookie car carries two implications. First, the team gains a far faster car-development baseline than it would with two young drivers. Second, the team has raised expectations to a level that actual results will struggle to match.

Marcin Budkowski, the team principal, appears in this story as the sole spokesperson. Every quoted statement comes from him. He speaks of Checo's great start, of the team making progress, of having some upgrades for Baku, and of the possibility that Baku becomes a turning point.

Those four phrases, placed side by side, form a complete messaging structure: acknowledge the bright spot, confirm the trajectory, promise something material, and anchor it all to a specific date. This is the kind of communication any team at the bottom of the standings would use. The issue is that the date, once spoken aloud, automatically converts an ordinary race weekend into a referendum on belief.

Baku is the next race. Two weeks. And in the constructors' championship, Cadillac still has not scored a single point.

Core: re-reading the data before re-reading the promise

In performance analysis, there is a boundary that viewers tend to collapse but engineers never do: the boundary between pace and reliability. A slow car is a slow car — to go faster you must change the aerodynamic concept, the suspension, the engine, or the entire understanding of the tyres. A fast car that breaks is an entirely different problem in kind: it is an industrial-maturity problem, and it can be fixed through process, through inspection, through time.

The electrical failure in Madrid belongs to the second category.

This matters for two reasons. First, electrical faults are the kind that can recur unpredictably, because they usually live at the interfaces between systems rather than in a single component. Second, precisely because they are unpredictable, they are the most expensive kind of fault for a new team, because they cannot be solved by buying a new part — only by understanding the car's entire electrical architecture.

When data enters the dressing room, emotion must leave through the window. And the data here says something fairly clear: in the Madrid race, the variable that cost the team points was not pace.

The quote deserves a careful re-reading. The team principal says Pérez gained a couple of positions, then went racing with Williams and Racing Bulls, then hit an electrical problem and had to retire the car. Three pieces of information sit in one sentence. Their order is also information: the bright spot comes first, the failure comes second.

If Pérez really was racing with Williams and Racing Bulls in the opening stint, then the Cadillac's pace under heavy fuel load and before tyre degradation was broadly equivalent to two teams that have been stable in the midfield for years. For a new team, that signal is worth more than any upgrade promise.

That inference needs a limit. One stint, one race, one source — and the source is the team itself. There are no lap times, no recorded gaps, no sector data. So the most reasonable conclusion is this: there is a positive pace signal, and it is not yet strong enough to call evidence. In my line of observation work, we call that a second-tier signal. Enough to monitor, not enough to conclude.

What is interesting is that this second-tier signal raises a question that runs opposite to the one the team is asking itself. The team is asking: how do we go faster in Baku. The better question may be: how do we get the car to the finish in Baku.

Baku is a street circuit. Its layout consists of very long straights — where engine power and aerodynamic efficiency decide almost the entire gap — interwoven with slow, narrow corners walled in on both sides. This is the kind of circuit where retirement rates tend to run above the average, because a single small error at one of those slow corners is enough to put the car into the wall.

For a team with a reliability problem, this is an adverse combination.

Two risks need separating. The first is driver risk: on a street circuit a veteran driver carries a large advantage, because they know how to keep the car away from the walls while still maintaining pace. That is a plus for Cadillac with Pérez and Bottas. The second is technical risk: street circuits place heavy load on cooling systems, braking systems, and the car's entire electronic architecture. And an upgrade package, by definition, tends to target pure performance rather than reliability.

This is precisely the sequencing trap.

For a new team, there are two routes to better results. Route one is to make the car faster. Route two is to make the car break less. Both need resources, and resources are finite.

When a team is stuck at the back without points, the natural pressure is to choose route one, because it yields immediately visible results — half a second per lap is something you can announce, while reducing the probability of an electrical fault is something nobody wants to hear. But across a season, the correct strategy is often the reverse. A team without points needs points first, and pace second. And the cheapest way to a first point is to make sure the car finishes.

The history of new teams makes this fairly clear. Many have had race weekends where the pace was enough for the top ten, only to lose the result to a small failure. Conversely, teams that score their first points often score them in races that were not fast at all — they were simply still there when others were not. In a race with a high retirement rate like Baku, the value of simply being present at the finish line rises considerably.

One detail needs stating plainly: the information about the Baku upgrade package comes from the team itself, confirmed by no third party, and carries no technical detail whatsoever. Nothing about which components are changing, which area of the car they affect, or which performance axis they target.

That places the claim in the marketing-information category rather than the technical-information category. It may be entirely true — or it may simply be a way to hold sponsor confidence for two weeks.

The most valuable contract is usually the one never announced. Likewise, the most valuable upgrade package is usually the one that needs no advance promise.

On the financial side, this story provides no data. No budget, no costs, no cost-cap compliance information. So any speculation about money would be invention, and I do not invent. But there is an economic logic that applies without figures: a racing team's commercial value lies in how often it appears on screen. A team running at the back and retiring early appears on television less often than a team running mid-pack and finishing. In a cost-constrained environment, the effectiveness of every invested unit is measured in seconds of television exposure. At that point, reliability becomes a commercial variable, not merely a technical one.

The Pérez–Bottas pairing is the team's greatest asset right now, and also its greatest pressure. Both are in the late stage of their careers. Both came from teams where the target every weekend was victory. Both understand the gap between a car that can win and a car that runs mid-pack. That is an advantage for car development: they can describe precisely what the car feels like and where it loses time.

But it is also a motivation risk. A veteran driver can endure a pointless season if the progress curve is visible. They endure it less easily if the curve goes flat. The absence of any direct quote from either driver in this story is a small but notable detail. It may be a deliberate communications strategy: during a difficult phase, the team allows only one voice to speak, avoiding any chance of a misread sentence. Or it may be a sign that the two drivers have nothing yet to say.

At the bottom of the constructors' championship, Cadillac's realistic target is not a high position — it is one point. The difference between zero points and one point, in standings terms, is small. In narrative terms, it is everything. A team with a point is a team that has proved its car can finish and can compete. A team without a point is a team still waiting for evidence. And in the second half of a season, when the big teams begin redirecting resources to the following year, the gap through which a small team can slip into the top ten tends to widen. That is why the timing carries not just psychological weight, but strategic weight.

A counter-intuitive angle: those two weeks were created by the team itself

The two-week deadline to Baku was created by the team, not by the calendar. The calendar simply orders the races; attaching the meaning of a turning point to one of them is an act of interpretation.

That interpretation can help: it creates internal drive, gives sponsors a date to hold onto, and gives the media a story to tell. It can also hurt: if Baku ends in another pointless weekend, the story reverses very quickly — and this time not because of a bad result, but because of a promise that did not land. In sport, people forgive a team that loses more readily than a team that promises and fails to deliver.

Here I want to circle back to something I have written many times. I was once a VAR sceptic, and that is why I understand those who hate it. That scepticism did not come from believing technology is bad, but from believing that hastily deployed technology tends to create a new class of error while resolving an old one. The same applies to the Baku upgrade package. A camera finds the fault, but only a human finds the cause. And here, the cause is not in the aerodynamics.

The counter-intuitive reading is this: it is possible that announcing the Baku upgrade was a mistake in expectation management, while still being the correct technical decision. The reason: inside a new team, components are not developed sequentially but in parallel. The Baku package may have been in the plan for months, and announcing it simply reports a pre-determined event. If so, the error lies in calling it a turning point — not in bringing it to Baku.

A second counter-intuitive point: this team may be in a better place than the standings show. If the Madrid pace was real, and if the electrical problem is an isolable fault that can be fixed, then Cadillac's true gap to the midfield is far smaller than the gap between zero points and whatever the direct rivals currently hold.

But this is where I must apply a mandatory-verdict principle. I cannot sit in the middle forever. The available evidence: one good stint, one electrical failure, one promise without detail. The most reasonable verdict: this is a team whose foundation is better than its position suggests, but which carries a reliability risk and has placed itself in a public test within two weeks.

One more thing needs saying about how to read the electrical failure. Inside a new team, every system is new at the same time: the electrical system, the hydraulics, the cooling, the software, the assembly process, the inspection process. A new team is not only developing a car; it is developing an organisation. An electrical failure may originate in a component, but it may equally originate in a process. And if it originates in a process, it will not disappear just because a new component has been fitted.

That is why I rate reliability risk above pace risk in the short term. Pace can be measured, benchmarked, and improved methodically over a few races. Reliability requires a chain of clean weekends to prove itself — and each clean weekend can only exist after the previous one was clean.

There is another dimension worth discussing: how the team manages its own expectations. Announcing an upgrade two weeks out creates an unrecoverable kind of pressure. From that moment, every Baku result will be read through the lens of the promise, not through the lens of the race. A finish in eleventh, under normal conditions, is progress. Under the conditions of a promised turning point, eleventh is a failure.

This is a very concrete form of opportunity cost: the team traded its freedom to interpret results for the creation of a hope milestone. In the short term, that trade may be right, because it sustains attention and resources. In the medium term, if results do not arrive, the price will be paid elsewhere — in the level of trust the team holds in itself.

And internal trust, for a new team, is a scarcer resource than wind-tunnel time.

Conclusion: what to watch in Baku is not the grid position

What needs watching in Baku is whether the car finishes. If the car finishes in the top ten, Cadillac's story shifts from struggling newcomer to improving newcomer, and the upgrade promise will be read as an honest one. If the car retires again for a technical reason, the problem will no longer lie in the upgrade package. It will lie in a different question: whether a new team is prioritising pace and reliability in the right order.

The authority of a referee does not come from the whistle, but from the ability to read the situation. A racing team is no different. Position is not made by pace alone, but by the ability to read the correct order of the variables.

In Madrid, that order was read wrongly once. Baku will reveal whether it was a one-off slip, or a habit.