Honda Spent Its Second ADUO Token on a Turbo and Singapore Still Lagged — The Deficit Is Calendar, Not Hardware

The throttle pedal was already flat as the car exited Turn 3 at Marina Bay, and the response arrived late. Fernando Alonso used a technical word afterwards — driveability. The paddock filed it as routine noise. For me it was a ledger entry. Earlier this year in Madrid, Honda spent its second ADUO token of the season on exactly this problem: a new turbocharger whose only job was to kill turbo lag. In Singapore the lag came back. The token was gone and the deficit stayed on the books.

The Sakura teardown made the picture sharper. Honda found no specific hardware fault. The turbocharger was not broken, not degraded, not built to the wrong specification. What they found was the old illness itself — turbo lag, the time it takes the turbo to spool, and therefore the delay between pressing the accelerator and feeling the boost. When a manufacturer spends its second and last major upgrade token of the year on a component and the post-mortem clears that component, the problem stops being engineering. It becomes calibration — and calibration is paid for in weeks, not pounds.

Context: the 2026 rules that brought lag back

To read this story properly you have to understand why turbo lag exists again in 2026. The previous rulebook carried the MGU-H — an electric machine on the turbo shaft that harvested waste exhaust energy and, crucially, kept the turbo spinning when the throttle was closed. That machine absorbed most of the lag. The 2026 power unit regulations deleted the MGU-H in the name of cost reduction, simplicity and attracting new manufacturers. At the same time the internal combustion and electric split moved to roughly 50:50 and the MGU-K was strengthened to around 350kW.

This is where my first note belongs. The rulemakers removed the anti-lag component and booked a saving, but the cost did not disappear. It migrated out of the parts bin and onto the driver's right foot. What Honda is fighting today is the outstanding balance on the 2026 rule package.

Shintaro Orihara, Honda Racing Corporation's trackside general manager and chief engineer, admitted as much to Motorsport.com. Even with the new turbocharger fitted to manage lag, Singapore made it hard to control, because of the unique track layout and the energy-rich situation. The result was unstable acceleration for the driver.

Mike Krack, Aston Martin's chief trackside officer, described the weekend in management language rather than engineering language — emotional dynamics. Sepang had surprised them on outcome, which lifted expectations for Singapore beyond what was realistic, and when delivery fell short everybody became disappointed and the weekend got heavier. Krack also conceded that simulation simply cannot render driveability in full.

Core: a token is a currency with its own exchange rate

I built a token ledger across the four manufacturers and Singapore turned from a headline into a payment schedule. ADUO — Additional Development and Upgrade Opportunities — is the mechanism that lets a manufacturer running behind on performance take a limited number of upgrades after homologation. Treat it as the cost cap's coinage. There are only a few coins a year, each buys exactly one upgrade, and once an upgrade is sealed the homologation door shuts.

Honda spent that coin in Madrid on a turbocharger. The logic was sound: the new power unit specification introduced at Zandvoort was losing driveability to lag, and the turbo was the address. The coin does not come back. That was capital expenditure — one-off, irreversible, non-interest-bearing.

Singapore reported that the capital had not gone to work as intended, and Orihara explained why. I will write it in budget language. The MGU-K is a fixed energy budget. On a normal track, part of that budget is spent spooling the turbo — in the off-throttle sections the engine runs at a high operation point and the lag falls away behind you. In Orihara's words, if you have enough MGU-K harvesting power this helps. But on an energy-rich circuit that same budget gets pulled into deployment demand, the harvesting margin thins, and the line item that was supposed to cover lag sits empty.

| Race | Specification / step | Outcome | |---|---|---| | Zandvoort | New power unit specification | Lag costs driveability | | Madrid | Second ADUO token, new turbocharger | Hardware changed, optimism built | | Sepang | Same specification | Result better than expected | | Singapore | Same specification, energy-rich layout | Lag returned, no hardware fault | | Austin (upcoming) | Software and settings countermeasures | First real test |

There is a subtlety here that routine reporting misses. Honda said MGU-K deployment can help against turbo lag but not enough to compensate for it. And the irony is that at Singapore the MGU-K became part of the problem, because its energy was spent exactly where the track demanded it. Orihara said that in such cases the engine's operation mode is not enough to control the lag.

Put differently: Honda's engine is not weak; its control layer is lagging the driver's demand. That distinction matters enormously, because a weak engine is cured with money and a control deficit is cured with time.

Now the question nobody asked loudly after Sakura. Is the absence of a hardware fault bad news or good? Bad, because no single component can be grabbed and fixed, which means the same condition can return at the next energy-rich circuit. Good, because something that can be corrected with settings rather than parts is generally the cheaper repair. Orihara pointed to that route — the new measures are not hardware but how Honda operates the power unit through software and settings. Lessons have been learned from Singapore, countermeasures are being prepared, but they still need to be checked on the actual power unit.

Here I repeat an old habit: every upgrade is a token schedule before it is a headline. In 2026 power unit development has its own financial regulation, teams sit under a separate cost cap, and aerodynamic testing carries its own allocation. All of it means one thing. Honda may well have the money to fix Singapore, but it does not have the token, and the token is the real currency. A fault you can cure with a cheque is cheap; a fault you can only cure with a calendar is expensive. Honda is in the second category right now.

The real constraint sitting in Sakura is dyno hours. Replicating a layout like Singapore on a dyno is not simple — nineteen corners, repeated braking, short acceleration zones, constantly shifting energy flow. Which is why Krack's point carries weight: simulation cannot give you that picture clearly, because this goes beyond simple sensitivity simulations. You can break aerodynamics down in CFD. You cannot break down that hesitation of a tenth or two under the driver's right foot.

I have to add one human line here, because a pure ledger never shows the whole picture. At Marina Bay, Alonso was covering with his foot a deficit the regulations had removed from the system. At a stage of a career where every tenth is priced, he was stretching braking zones and altering exit lines while waiting for a push that arrived late. That is not free. A driver's trust is spent the way tokens are spent, and it does not refill.

Contrarian: Singapore was a valuation loss, not a performance loss

The convenient paddock story is that Honda is behind and Aston Martin's 2026 is quietly unravelling. I read it the other way. Sepang delivered better than expected — that was good news, but booking it on the balance sheet was the error. The team treated a temporary performance as permanent capability and carried that valuation into Marina Bay. What Krack calls emotional dynamics is a failure of the risk register. The failure was not that the car was slow in Singapore; the failure was that the team bought Singapore at Sepang's price.

Second unpopular point. Spending the token on the turbocharger was not a wrong decision — the symptom pointed there. The mistake was treating the upgrade as a component when the problem was control. Fitting a new turbo while leaving the control logic as it was meant the hardware benefit existed mostly on paper at a layout like Singapore. It is the same error companies make when they buy software: the licence is purchased, the configuration never happens.

Takeaway

Austin will be the first genuine audit report. COTA is a circuit of long straights and heavy braking — similar energy demand, different corner character. If the new calibration tames the lag there, Honda will have converted an engineering problem into a process asset and the rest of the season becomes survivable. If it does not, next year's third token becomes the most expensive line in the budget. The question is not whether Honda has the money. The question is how many weeks Sakura has left — and weeks cannot be bought with a token.