Both Cars Retired in Five Laps in Mexico: The Real Story Behind Honda's ERS Cooling and MGU-K Failures

I watched the Mexico City data log three times. The first time, only one number stood out — five laps. The second time, I saw the mechanics in both Aston Martin garages emptying at the same moment. The third time, I saw what the broadcast never showed: abnormal pressure readings in both power units. This was not a race result. It was a public demonstration of an engineering problem the team had already anticipated.

Context: When Altitude Becomes an Engine Test

The Mexico City circuit sits 2,286 metres above sea level — the highest point on the calendar. Thin air means the turbo must work harder to build the same pressure, and the cooling system must extract more heat from air that is less available. That is physics, not setup.

Honda had already acknowledged reliability pressure before the race. Aston Martin's chief trackside engineer Krack later said the team knew before the race that neither car would finish. That sentence is easy to overlook, but it is the heaviest fact in this entire episode. When a team accepts defeat before the race, the question is not why the car stopped — the question is what options the team tried and why they were not enough.

Honda said abnormal pressure levels were recorded in both power units, and that the failure chain involved ERS cooling stress and MGU-K failures under special circumstances. Alonso and Stroll both retired within five laps. Alonso's qualifying pace was described as promising, but he was told to park — meaning the problem was not in the car's speed, but in the system that sustains it.

Core Analysis: The Arithmetic of Five Laps

Both cars retiring at the same time is not an accident — it is a signal of a system-level vulnerability. If this were two separate driver errors, the failure pattern would differ. The same pressure-related anomaly appearing in both cars suggests the problem lay in the operating condition — the thin air of Mexico and the cooling demand it creates.

But there is an important distinction the media merged. Honda cited ERS cooling and MGU-K, but did not disclose the full engineering root cause. Abnormal pressure, cooling stress, and MGU-K failure are part of one chain, but which link broke first was not established publicly. This distinction matters because if the root is in cooling design, Brazil's hardware changes can work; if the root is in the durability of a component inside the MGU-K, software tuning will not be enough.

I have worked on several power-unit reliability reports over the past few years, and one pattern recurs: when a team predicts retirement before a race, it has already tried something. According to Krack, Aston Martin and Honda discussed measures to extend running — free air, limited gear use, and turning down power. These are the same tools an engineering team pulls out when it knows the race will not run under normal conditions. But Krack also said these options quickly proved insufficient.

The real question here is: what was the trade-off between reducing power and running more laps? No one answered. But five laps tell you the trade-off was very narrow. If power had to be reduced to a level where the car was no longer competitive, staying in the race makes no sense. That is the arithmetic that explains Krack's "we knew" statement.

For Brazil, Honda planned changes to both hardware and software. That is the right direction, but it is not yet proven. Interlagos sits at a lower altitude than Mexico — around 800 metres — so cooling pressure will be lower. But lower pressure does not mean the problem is gone. If the same failure mode appears in Brazil, the problem will not be Mexico-specific but about the general robustness of the power unit.

One more thing that never appears in the standings: the impact of a double retirement falls directly on the constructors' championship, but the real loss is the data the team did not get. In five laps, the team got no race-pace data, no reliable sample of tyre degradation, and no basis for comparison with rivals. That loss is bigger than the points, especially in a season where every race shapes development direction.

Contrarian Angle: Is This Really Just a Mexico Problem?

Here I have to doubt my own thesis. The mainstream argument is simple: Mexico is at altitude, the air is thin, cooling is reduced, so the failure happened. Brazil is lower, so the problem will not repeat. That argument is convenient, and possibly wrong.

Honda itself said the failure occurred in "special circumstances" — a phrase worth reading carefully. "Special circumstances" may mean more than altitude. It could refer to the race-weekend duty cycle — sustained high-power runs in qualifying, slow running in traffic where cooling is ineffective, and sudden power demand after a pit stop. Honda mentioned "different lap types," which points in this direction.

If the problem is lap-type-dependent, it can appear at any circuit where long periods of low-speed running occur due to traffic or a safety car. That reduces Mexico's uniqueness and makes the problem more general. I do not have proof of this, but Honda's words do not rule it out.

The second angle: Krack's statement that the team knew the cars would not finish creates a communication risk. If readers interpret it as the team knowingly accepting a certain retirement, questions will arise about whether better options existed. Deciding to withdraw before the race and retiring after five laps — the line between them is thin, and the team must clarify what it tried.

Takeaway: Brazil Is a Test, Not an Answer

The real value of Honda's hardware and software changes in Brazil is not whether the cars finish. The real value is whether they hold up under the same operating conditions that caused the Mexico problem. If the cars run the full race at Interlagos, it will be a sign of improvement, but not proof — because the altitude is lower. If abnormal pressure or cooling problems appear there too, the question will move beyond Mexico to the robustness of the entire power-unit programme.

Honda said the experience will feed into its future engine programme, including the 2027 power unit. That is a plan, not a performance. Until a concrete design change and its testing are made public, the claim remains at the level of speculation.

What I will watch is this: after Brazil, does Honda name a specific root cause, or only claim "improvement." A team that understands the problem names it. A team that hides it only says it has been fixed.

One thing is certain — both cars stopping after five laps is not a lost weekend for that team, but a question about the direction of its season. And when a team accepts defeat before the race, the real problem is not the circuit's altitude but the limits of the system it will use again next year.