Two Denials, One Silence

At the Sunday press conference after the United States Grand Prix in Austin, two men were asked about the same subject and both answers stepped around the same thing. Laurent Mekies was asked whether Red Bull would supply Aston Martin with a power unit in 2027. The answer: "no serious conversations have taken place." Mike Krack was asked whether Aston Martin was looking beyond Honda. The answer: Honda is our long-term partner, and this is their 500th race.

Both denied. Neither denied the one premise the whole story rests on — that Honda's 2026 power unit is running behind expectation, and Red Bull's is running ahead of it. Mekies denied supply; he did not say the rumour's root was false. Krack denied leaving; he did not say the performance criticism was wrong.

Let me rewind to the frame everyone skipped. The nouns missing from the denials are the actual story.

---

First, what it actually means to be a power unit manufacturer in 2026

2026 is not a fuel change. It changes how a power unit works. The MGU-H is gone — the component that made turbo lag an engineering question from 2026 onward and kept Mercedes, Honda and Renault apart for years. Now the internal combustion engine produces roughly 400kW and the electrical side roughly 350kW: near enough to half and half. Fuel is 100% sustainable. On top of that sits active aero, which is directly coupled to cooling and packaging.

There is one consequence. A power unit is no longer a thing you sell separately from a chassis. When half the force comes from a battery and deployment, then where the battery sits, where the cooling duct draws from, how energy recovery is managed when the active aero zone flips — all of these are power unit decisions that are also aerodynamic decisions. In the old era a supplier handed over a black box. Now it hands over a package, and if the package sits wrong, the engine's power stays on paper.

This is where Red Bull's context matters. Red Bull Powertrains is not a supplier that pulled an engine off a catalogue. It is a manufacturer built after the Honda relationship ended, in partnership with Ford, which had to put its first power unit into a race car in 2026 — the same year it had to run two teams, Red Bull and Racing Bulls. A startup, a supplier and a brand operation, all in one building.

So this is not merely a story about who supplies whom. It is the first-season capacity audit of a new manufacturer, and the outcome of that audit will set the price in the 2027 supply market.

Add one more layer: power unit manufacturers sit under a separate cost cap — broadly in the region of $130 million a year, with additional allowance for new entrants in their early seasons. (That figure comes from my notebook; I do not have the final FIA confirmation in front of me. Confidence: medium.) The point is that Red Bull does not have unlimited engineering hours. Every hour that goes into a customer team's integration file comes out of its own calibration. That is the invisible cost sitting behind every supply deal, and it never appears in a press conference.

Now put Aston Martin in the frame. The Silverstone team is Honda's works partner from 2026. Works means the power unit is built under their name, the cooling and packaging design is set around their chassis, and the contract term is such that an exit route exists on paper but is very expensive in practice. That is the trap teams fall into: works status confers prestige and removes flexibility.

---

"Benchmark" is a label, not a measurement

The heaviest word in the report is the one attributed to the FIA — that Red Bull's combustion engine is regarded as the "benchmark." I read that sentence three times and hit the same problem each time: not one number came with the claim. No lap time, no GPS speed trace, no energy deployment log, no dyno curve. Just an adjective.

And the body it is attributed to does not have that record. The FIA does not normally publish comparative power unit rankings. It monitors performance, runs homologation, requests data under convergence provisions — it does not issue certificates saying "this is the best engine." So the claim is either a mis-framed reference to a monitoring report or a paddock rumour that acquired an official stamp. (Confidence: medium. If that word genuinely appears in an FIA document, this paragraph will be the first thing I retract.)

There is a basic rule in technical analysis: a claim without data is a direction, not evidence. So I hold "benchmark" as a direction. Red Bull's combustion concept may well be strong. Beyond that I cannot go.

Two wins and a "benchmark" do not sit together

By Austin, Red Bull has two wins. That is a solid number and nothing more. Two wins means the car could win in at least two different circuit archetypes or conditions — and for a podium-tier team at this stage of the season, that is roughly par.

The real collision is here: if the power unit is genuinely class-leading, then two wins by this point is very few for that power unit. I write it plainly because this is the conclusion the paddock grapevine did not draw. If the strength is in the engine and the results stall at two wins, the constraint is elsewhere — chassis, aero, tyre management, operational execution, or the sum of all of it.

That is a materially different story from the one Red Bull's communications is selling. In one, the engine is the hero. In the other, the engine is fine and everything else is in question.

Let me put my conditions on the table, because I prefer the difference to be visible: I have not seen the circuit archetypes of those two wins. If both came at low-drag, power-sensitive circuits, the benchmark claim gains indirect support. If they came at high-downforce circuits, the claim weakens — aero and mechanical grip dominate there. The report withheld that. Withholding is itself information: PR is choosing what to disclose.

The sentence PR left out, and the one that mattered most

Rewind to the skipped frame. The report contains a line from Mekies that breaks the "sorted engine" narrative entirely. He said — and I will not soften it in translation — "there are a thousand driveability considerations that we know we can do better."

That is the most honest admission a young manufacturer can make, and it carries more weight than any report calling the unit a benchmark. Driveability is where the gap between peak power and race pace is settled.

Think about what a race weekend actually is. The driver brakes and engine braking torque tells him how quickly the car will rotate. On corner exit, the moment electric deployment expires determines how hard the next straight begins. At the end of a lap, how much battery is left shows up in the first sector of your qualifying lap. These are a thousand small decisions with no single dyno test. They surface only in a race, as tyres degrade, as fuel burns off, as track temperature moves.

When a team principal says himself that a thousand of those details can be improved, he is also saying that the engine's operating window is currently narrower than its peak output. That is the classic signature of a new manufacturer: strong combustion concept, incomplete calibration.

For me that is the biggest technical disclosure in this story, and it arrived inside a denial.

Integration: the reason not to supply, or the excuse not to supply?

Mekies gave the real reason for building their own power unit: integration. It sounds like corporate language, but under the 2026 rules it is a hard technical argument.

When half the force comes from a battery, battery mass and location set your weight distribution. When the MGU-H disappears, the energy that used to spool the turbo from the exhaust now has to come from the electric motor — and where that energy comes from decides your recovery strategy. When the active aero zone flips, drag drops suddenly and cooling demand changes; the power unit has to be mapped for that instant.

So on engineering grounds Mekies is right. A chassis manufacturer that builds its own power unit can close its design loop. A customer team cannot — it waits on the supplier's calendar.

And here is where I get suspicious, and I will not hide it. The technical case for not supplying and the commercial case for not supplying produce the same answer — which means the answer cannot tell you which one is real. If RBPT genuinely lacks the bandwidth to run two teams in year one, then the integration argument can be both true and convenient. There is exactly one way to separate them: look at RBPT's customer register for 2027. (Confidence: low to medium.)

Two teams, one year: the capacity arithmetic

One comparison is enough to place Red Bull's powertrain division. Mercedes has supplied four teams for decades; Ferrari has run multiple customers for years; both have a customer-support structure accumulated season over season. RBPT carried two teams in its first year — its own and Racing Bulls.

That is not small. Every customer team means a different chassis, a different cooling layout, a different driver's driving style, and therefore a different calibration. Under a cost cap, that work is done with engineer hours, not money, because the money is capped.

So when Mekies says "there is enough work on our side," that is not modesty. It is cost accounting. And if you take that accounting seriously, adding a third team stays a large question in 2027 no matter how attractive the rumour looks.

Krack's defence, which was not a denial

Here I have to be blunt: Mike Krack did not rebut the performance criticism.

Look at what his answer was made of — "long-term partner," "500th race," "when we win our first race together." That is the language of the future. That is the language of a relationship. There is not one sentence in it saying: our power unit is competitive this season and we can prove it with data.

In defensive communications this is a recognisable structure. When you cannot answer the present criticism, you change the time horizon of the conversation. Drop today's question into a ten-year relationship and the question starts to look small.

The Honda 500th-race reference did exactly that work. It is a genuine milestone, but in the context of that press conference it was a relationship-repair device — a way to re-anchor the partnership in public. And the thing that needs public repair is, internally, a little bit broken. (Confidence: medium.)

---

The denial is really a denial about price

Now the part I found most interesting, and the part no headline will carry.

Mekies did not say "we will never supply." He said "no serious conversations." Elsewhere: "we don't feel we are ready."

Read both again. Neither is a permanent no. Both contain a condition — readiness. And readiness is a thing that changes with time and money.

In my experience, denial language in the paddock comes in three grades, and each means something different. One: "no, absolutely not, never." That closes the door. Two: "no plans at this time." That leaves the door open. Three: "we are not ready." That hangs a price board on the door.

Red Bull chose the third sentence. That is not technical clarification; it is market signalling. The translation is: we can supply, but the price we would ask right now is above what you expect — and that price will be denominated not in dollars but in engineering hours and homologation calendar.

On the other side, the thing nobody denied is that Red Bull is uncomfortable with the praise. The report calls it "unwanted positive publicity." That reads oddly — why would a team be troubled by compliments? The answer sits in the rulebook. If a power unit is publicly labelled front-running under the 2026 governance framework, rivals will immediately ask the FIA for convergence measures to close the gap, or will question the allowance given to new manufacturers. Praise does not arrive as a compliment. It arrives as an audit.

So Red Bull's denial is working two fronts at once. One side says we will not be a supplier yet. The other says do not call the engine this good, because the scrutiny follows. In one breath it steps back and pushes forward.

And Aston Martin has no parachute in this story. Changing a works partner is a contract change for a customer team; for a works team it is a change of entire design language. If Honda's 2026 power unit is genuinely behind, Silverstone has one route into 2027 — patience, and a bet on the promised improvement. Nobody calls that option exciting, but it is the only one that works inside the cost cap and the homologation window.

That asymmetry is what the whole story rests on: Red Bull has the freedom to choose. Aston Martin has only the freedom to wait.

---

What to watch next

I built all of this from a Pune flat, logging sector times off the timing screen, and I am not presenting it as settled fact. I am calling it a reconstruction, with its assumptions and its error bars attached.

Three things will decide this story, and all three will show up on track, not in a press conference.

First, cooling-sensitive circuits. The rounds with high temperatures and low grip are where deployment management and cooling margin get genuinely tested. If Red Bull looks inconsistent there, Mekies' "thousand details" will outlast this season.

Second, Honda's improvement curve. The first year of any regulation set is always incomplete. The question is not who is ahead in October; it is whose curve is steeper. If Aston Martin's qualifying gap halves over the next six rounds, the rumour dies on its own by next season.

Third, the 2027 supply register. That is the document the truth gets printed on. If RBPT adds a third team, "not ready" was a price lever. If it does not, the capacity argument was real.

And if neither happens — no supply, no improvement — the question stops being whose engine is fastest. It becomes whether the 2026 power unit rule set, written in the name of closer competition, actually did that job, or whether it simply put one manufacturer early in front and everyone else in a long wait.