Cadillac, the Madrid Electrical Fault and a Reliability Test Before Baku

The car stopped at the edge of the Madrid circuit, and on the radio one word kept repeating: electrical. Moments earlier, Sergio Pérez had gained a couple of positions and had gone racing with Williams and Racing Bulls — exactly where a brand-new team needs to be: in the middle of the fight, not at the back of the field. Then the electrical system spoke, and his race ended early. The Cadillac team principal called Checo's opening stint “great” and called the ending “an electrical problem, we had to retire the car.” One compliment, one diagnosis. Of the two sentences, the diagnosis is the one I took back to my desk.

Across its debut season, Cadillac has yet to score a point and sits bottom of the constructors' championship. For a new team, that bottom place is no surprise. What draws attention is something else: the way this team narrates its own failures.

Context: a new team, two veteran drivers

Cadillac entered its first season as the eleventh team on the grid, backed by General Motors and an engineering operation built almost from scratch: factory, wind tunnel, supplier relationships, race-weekend procedures. In the cost-cap era, nobody buys speed with a cheque. A new team only gets faster at the rate its engineering organisation matures.

On human capital, Cadillac is no rookie operation. Sergio Pérez has won Grands Prix and raced at the front across many seasons. Valtteri Bottas won constructors' titles with Mercedes and stood on the top step ten times. Putting two race-winning drivers into a start-up is a deliberate move: it buys credibility, buys high-quality technical feedback, and buys expectations. Expectations are double-edged, especially when neither driver has scored.

Based on my experience following race weekends across many seasons, I keep a rather dry habit: open the results sheet, read the retirement column first, and only then read the timing sheet. The retirement column tells me which teams have a machinery problem and which have a people problem. The timing sheet only tells the rest.

The Azerbaijan Grand Prix at Baku comes two weeks later. That is the deadline the team set for itself, and the deadline by which outsiders can test what the promise is worth.

An electrical fault is not a pace deficit

In a race there are two entirely different kinds of deficit. A pace deficit is structural: a chassis short on downforce, aerodynamics not efficient enough, a power unit behind the opposition. Fixing it takes months, sometimes a whole season. A reliability deficit is a timing failure: a component breaking at the exact moment the car is running at 300 km/h.

Where does an electrical fault live in the hybrid era? In the wiring loom running through the car, in the control electronics, in the sensors, in the energy-recovery system, and in hundreds of connectors that absorb constant vibration. The nastiest property of this fault class is intermittency: on the bench, the car can run perfectly. In the garage, engineers may find nothing. Then, out on track, surface shocks from a street circuit, temperature, and vibration at a particular rev band bring it back. This is the fault class engineering departments fear most, because it cannot be verified by feel — only by data.

Files do not lie. People build files to lie on their behalf.

That line came after four months of keeping a draft in a drawer, when I was following test records from a laboratory in Moscow and did not have enough to publish. Since then, every conclusion about a team has to pass through three independent evidence layers before it reaches the page. In Madrid, layer one — the team's own statement — already exists: electrical fault, forced retirement. That layer is the single source, and it comes from the very person who needs to protect the project's credibility. Layer two has to be objective data: Pérez's sector times in that opening stint, top speed on the straight, a like-for-like comparison with Williams and Racing Bulls on the same tyre and the same fuel load. Layer three is independent observation from rival engineering areas — people with no reason to paint Cadillac's story in rosy colours.

The information in my hands has only touched layer one. When a team is losing, “we are making progress” is always the easiest sentence to say. It is not wrong yet. It is also not verified.

The upgrade promise and the sequencing trap

Cadillac says it has upgrades for Baku and believes they could be a turning point. Noted. But I hold no specification at all: no component names, no homologation date, no aerodynamic figures, no downforce or efficiency numbers. An upgrade package exists twice — once on a drawing in the factory, and once on the timing sheet at a circuit. The second time is the real one. A press briefing is only a market stall; the control-box data is where a fault gets verified.

There is a sequencing trap new teams fall into constantly. Under pressure for results, they pour resources into pace items — wings, floor, exhaust — before a reliability baseline is secured. The car gets quicker and still retires. For a team at the bottom, that trade is almost always a loss: a tenth of a second means nothing if the car is parked on the verge on lap 40. A new team's first points usually come not from outright pace, but from staying in the race longer than everyone else.

Baku: street walls and an electrical system under suspicion

Baku is a street circuit with the longest straight in the sport's history, running through the city centre before narrowing abruptly into tight corners hugging the walls of the old town. For a car with a demonstrated electrical fault, that is the worst conceivable combination: it demands straight-line speed and stability when the car climbs kerbs and changes direction violently. Baku's road surface is not flat. Every connector, every loom, every mounting bracket takes exactly the kind of vibration control electronics dislike.

Street circuits carry another property: retirement rates there tend to run higher than at permanent tracks, because a small mistake puts the car in the wall. That cuts both ways. For a team hunting points, rival attrition is opportunity. For a team with a reliability problem, that opportunity sits next to an equally sized risk.

How points are distributed at the back of the field makes the game harsher still. One point for tenth place. To get it, a new team needs almost simultaneously: a qualifying result deep enough not to start too far back, race pace good enough not to be dropped, a clean pit stop, and a finish. Four conditions, none of which substitutes for another. The team managed the hardest part in Madrid — matching midfield pace — and then failed the theoretically easiest part: bringing the car home.

The people in the garage

Behind every retirement is a night shift. An electronics engineer has to strip bodywork, inspect connectors one by one, measure circuits, cross-reference control-box logs against vibration traces, and then decide whether to change an entire loom or just one plug. It is work nobody sees on television, yet it decides whether the team leaves Baku empty-handed. At a new team, these people are simultaneously building procedures for the future and firefighting the present. Meanwhile, in another department, someone still has to write a press release about an upgrade package. One organisation running on two opposing emotional tracks: a promise about speed, and a race against components that can fail at any moment.

The other side of the difficulty

The sceptical reading is real. A statement about an electrical fault, issued by the very person who needs to reassure management, sponsors and staff, can be read as a palatable phrasing for a larger problem. When a team repeatedly says “progress” and “turning point” without handing over a single data file, my professional reflex is to look at the retirement column rather than the interview answers. And there is reason to doubt: one good stint in Madrid is one sample. Madrid's low-speed corners and traction demands can flatter a car that suffers elsewhere.

Alongside that scepticism, I keep the legitimate part of the team's position. The Madrid electrical fault carries meaningful good news: it shows their core deficit is a fixable engineering matter, not a structural one that must wait for next season. A team that retires while racing Williams and Racing Bulls is in a better position than a team finishing cleanly at the back of the field, because the first question has been answered. If the fault turns out to be systemic and recurring, the conclusion changes entirely.

Clean is not the same as transparent. One is a scent; the other is double-entry bookkeeping.

On a race track, the translation is this: a contact-free weekend is not the same thing as an electrical system without a fault. The team is telling the public a story about cleanliness. What I am waiting for is the ledger: a list of retirement causes, race by race, updated publicly.

Signals to track

To turn guesswork into something verifiable, I keep a short list. Retirements from non-driver causes, at Baku and after. Best-lap gaps to Williams and Racing Bulls on the same tyre. Any public quote from Pérez or Bottas, because driver silence in a story sourced solely to management is itself data. And the points gap to the team directly above.

None of those items can be filled in with a promise.

Takeaway

For a new team, the first season is not measured by points collected but by the record it leaves behind: the failure catalogue, the causes, and the speed of repair. A claim about an upgrade package can make a headline, but it only stays with a reader if the circuit confirms it. The worthwhile thing at Baku is not promising to be faster, but getting both cars to the finish line — and then publishing why they failed last time. The audience that follows this sport has been around long enough to tell an opened ledger from a handed-out leaflet.