Picture the scene at the 1989 Brazilian Grand Prix in Rio. Nigel Mansell is sitting in a Ferrari that most of the paddock has quietly decided is either visionary or delusional, depending on which engineer you ask.

The car, designated the 640, was a high-wire engineering hubris. People must remember that the thing was notoriously fragile; it was a miracle it didn’t just disintegrate into a cloud of expensive Italian shrapnel. However, it led a teleological shift in motorsports by having no gear stick! In its place, behind the steering wheel, sit two small paddles—pull one to upshift, pull the other to go down. The idea of shifting gears from the steering wheel was not particularly new. A French inventor named Amédé Bollée had built a purely mechanical version of the concept in 1901, and his Bollée Type F Torpédo of 1912 still sits in a museum at Le Mans, largely forgotten.

While definitely inspired by the Type F, what Ferrari's chief designer John Barnard had built was something categorically different: the first electro-hydraulic sequential paddle-shift system in Formula One history. He achieved this by borrowing actuator technology from the aircraft industry and pairing it with electronics sophisticated enough to complete a gear change without the driver touching the clutch mid-race. This meant no hand leaving the wheel and no missed shift at twelve thousand RPM. A mechanical convenience that saves a tenth of a second, which compounds across several laps into a lost position.

Mansell wins the race on debut. Williams followed with their own system two years later, as their FW14B achieved a total hegemony by sweeping both championships in 1992. The argument was over. By 1996 every team on the grid had abandoned the manual gearbox entirely. Eight years after that first race in Brazil, Ferrari put it in a road car—the 355 F1, introduced in 1997, the first production car in history to offer a paddle-shift transmission. This was the F1 ethos: a laboratory powering whatever goes on behind a checkered flag.

Now jump forward to September 2021, to a hotel conference room near the Monza circuit. With wrinkled foreheads and coffee breaths cutting through the air-conditioned hush, the bosses of the world's most glamorous motorsport are assembled to decide the shape of Formula One's engines from 2026 onwards.

Toto Wolff, the head of the Mercedes racing operation, is talking to reporters. He is the steward of a team running the most technologically sophisticated engine program in the history of the sport. His power unit has just won its eighth consecutive Constructors' Championship. The engine at the heart of it achieves thermal efficiency numbers that would make an aerospace engineer clap with excitement.

Specifically, we're talking about the Motor Generator Unit-Heat, popularly known as the MGU-H. This component took twelve years and hundreds of millions of euros to master, achieving something no engine in the history of competitive motorsport had managed before. It converted more than fifty-two percent of every drop of fuel into actual forward motion. Your road car converts roughly thirty percent, and the most advanced diesel locomotive manages perhaps forty-five. In a device roughly the size of a suitcase, Formula One closed a gap that the rest of the engineering world spent decades failing to bridge.

And yet, Wolff is explaining, with the measured equanimity of a man making a business decision he has already fully processed, that Mercedes is prepared to sacrifice the MGU-H—the most exquisite component of that engine—just to persuade the Volkswagen Group to show up.

The aha is that the sport has undergone a powerful downgrade. It has moved from an era of innovation-as-dominance to an era of mercantile concession. The 52% efficiency milestone is being retired because explaining that level of complexity on a billboard is essentially impossible. We have transitioned from the raw, unpredictable laboratory of the 640 to a boardroom-driven realpolitik, where the most advanced engineering is traded away to ensure the grid stays crowded.

Before we get to the funeral, we should spend a moment at the birth, because the story of Formula One as a technology engine is genuinely remarkable and earns its mythology. Well at least, up to a point.

In the 1950s and 1960s, what actually happened in Formula One from a technology transfer perspective was less a series of dramatic inventions and more a sequence of proof-of-concept moments. Ideas that existed in theory were proved under the specific pressures of championship racing, and then migrated outward into the broader automotive world. Two of them changed the shape of every performance car built since.

The first was the mid-engine revolution, and to understand it properly you need to understand how completely everyone agreed it was a terrible idea. In 1950, every car on the Formula One grid had its engine at the front. The driver sat behind it while peering down a long bonnet resembling a girthy pinocchio’s nose, with most of the car's weight pushing on the front axle. This was how physics, at least how the racing establishment understood it. Enzo Ferrari, the most powerful man in the paddock and a figure who treated dissenting opinions roughly the way a thunderstorm treats a picnic, expressed the consensus view with characteristic economy: "The horse shouldn't push the cart with its nose," he said. "It pulls it." It’s safe to say that his authoritative opinion carrying an imperial-isque significance simply mandated that the engine goes at the front

Into this settled world arrived Charles Cooper and his son John, who had started making racing cars from a garage in Surbiton, Surrey. In the context of 1950s Formula One, this sentence would have struck most paddock insiders as self-evidently unpromising. The Coopers had been putting engines behind the driver in smaller racing categories since the early 1950s, mostly because it was cheaper and simpler to package. The logic that emerged from necessity turned out to be correct: move the weight to the middle of the car to lower the centre of gravity, which in turn would improve traction. In 1958, Stirling Moss drove a Cooper to victory in Argentina, the first world championship race won by a rear-engined car. In 1959, Jack Brabham drove the Cooper T51 to become the first driver to win the World Championship with an engine mounted behind him.

Ferrari, the last front-engined holdout, began experimenting with a rear-mounted engine in 1960. By 1961 the entire Formula One field had to undergo rechristening. In a humbling twist of fate, the garage owners had beaten the horse man, leading him to quietly build a new cart.

He was not done with the analogy, though. When Lamborghini unveiled the Miura in 1966 — the first high-performance production road car with a mid-engine layout, designed by engineers who had watched Formula One's rear-engine revolution with close attention — Ferrari aimed the same horse-and-cart line at Ferruccio Lamborghini. 

He was wrong again, on a grander stage, in front of a larger audience. The Miura became one of the most influential road cars of the twentieth century. Every mid-engine supercar built since, after Enzo eventually capitulated to physics, carries the DNA of what Cooper proved at championship level in 1959. A layout vindicated and adopted by the sport's establishment, and absorbed by the road car industry in the years that followed.

The second turning point came in 1962, and it belonged to Colin Chapman. His approach to engineering was so aggressive, that it occasionally terrified his own team. Chapman looked at the tubular steel spaceframe chassis that every Formula One car used and decided it was fundamentally wrong. He had been thinking about the stressed-skin structures used in aircraft, where the outer shell itself carries the structural load. His inspiration, by his own account, came from the steel backbone frame of the Lotus Elan road car, which he wondered might work if widened to allow the driver to sit inside it.

The Lotus 25, which debuted at the 1962 Dutch Grand Prix, was the first fully stressed monocoque chassis in Formula One history. In the hands of Jim Clark it took fourteen World Championship Grand Prix wins and propelled him to the 1963 title. The monocoque weighed just thirty kilograms and was three times stiffer than the spaceframe car it replaced. The opposition, who had recently purchased the previous Lotus model on the understanding it was mechanically identical to the works car, discovered at Zandvoort that Chapman had a different understanding of the word identical. Within a few years every team in Formula One was building monocoques.

The third act in Formula One's technology transfer story began in a small room at a British Aerospace facility in Weybridge, Surrey, where a single man sat alone cutting pieces of carbon fibre with a bandsaw. John Barnard, a quietly obsessive British engineer who had spent years designing racing cars in America before returning to Europe, was at the time working for Ron Dennis, a former racing mechanic turned team boss. Dennis had recently merged his outfit, Project 4, with the struggling McLaren team, and had handed Barnard the job of designing their first car together. Barnard had been invited to look around the British Aerospace facility by a contact, and what he found there would change everything.

They were manufacturing the outer cowling for the Rolls-Royce RB211 jet engine from a carbon fibre and aluminium honeycomb sandwich. Barnard stared at a man there working with a bandsaw for a moment. His own words, recalled forty years later to MotorSport Magazine: "I thought, 'Crikey! If this is how British Aerospace are making their carbon inserts, I'm bloody well sure I can do the same on an F1 car.'"

So here was the problem. Barnard wanted to make the monocoque dramatically narrower than anything that existed at the time. A narrower monocoque meant wider aerodynamic tunnels underneath the car, which meant more downforce, ultimately leading to faster lap times. Aluminium, the material every team on the grid used, could not be made narrow and stiff at the same time. But carbon fibre could! British Aerospace told him they did not have the capacity to build a full racing chassis from it. Every other British company he approached said exactly the same thing. Barnard was sitting on an idea he was certain would work, with absolutely nowhere to build it.

The way out arrived probably over a beer. A newly hired McLaren engineer named Steve Nichols had previously worked as an apprentice at a company in Salt Lake City, Utah called Hercules Aerospace. Hercules’ primary business had been building rocket components for NASA. Nichols had a hunch they might be interested. Barnard boarded a flight to Utah with a quarter-scale model of the chassis in the seat beside him, walked into a meeting with people who had never built a Formula One component in their lives, and proposed turning them into a manufacturing partner. Hercules said yes.

The McLaren MP4/1 debuted at the 1981 Argentine Grand Prix as the first Formula One car to race with a fully carbon composite chassis. Where a conventional aluminium monocoque of the time required around fifty separate components mechanically fixed together, the MP4/1's was assembled from five. The proof of what this meant arrived at Monza later that season, when John Watson hit the barriers at 140 miles per hour and the engine and gearbox tore clean off the back of the car. Not so miraculously, the monocoque stayed intact around him, and Watson walked away unscathed. Hercules Aerospace took the wrecked chassis back to Salt Lake City and used it as a sales pitch — showing visitors the video of the crash, then pulling back a curtain to reveal the bare monocoque sitting intact behind it. Watson later told Motor Sport Magazine that it probably more than quadrupled any investment Hercules had made in the project.

The rest of the paddock, which had spent months confidently predicting that the car would disintegrate into black dust in a crash, went very quiet and then went away to build their own carbon monocoques. By the mid-1980s, carbon fibre was the only serious material for a Formula One chassis. By 1992, legendary engineer Gordon Murray had taken everything McLaren had learned building race cars and applied it to the McLaren F1 road car, the first production vehicle with a full carbon fibre chassis, and for a long time the fastest road car ever built. The material that began its automotive life in a conversation over a beer and a flight to Utah is now in the Lamborghini Huracán, the Ferrari 488, and the structural members of the Airbus A380. Formula One took carbon fibre from aerospace and sent it back to the world transformed.

The turbocharging story sits alongside this one, though it is messier and the direction of transfer is harder to trace cleanly. In 1977, Renault appeared at the British Grand Prix with an engine nobody had seen before in Formula One: a 1.5-litre turbocharged V6 that produced around 500 horsepower but suffered from turbo lag so severe that Derek Warwick, who drove turbocharged cars throughout the 1980s, once described his early experience as pressing the throttle at one end of a chicane and feeling the power arrive fifty yards after the corner (deja vu for 2026). The car was unreliable and frequently laughed at. Two years later it won a race. Three years after that the entire grid was chasing the concept!

By the peak of the turbo era in 1986, Formula One engines were producing over 1,300 horsepower from 1.5 litres of displacement in qualifying trim. The engineering demands of achieving this — intercooler efficiency, wastegate control, turbo materials capable of surviving turbine speeds of 180,000 rpm, engine management software sophisticated enough to modulate boost pressure mid-corner — pushed manufacturers into territory that no road car program had the budget or the urgency to explore alone. The hot hatches and performance cars of the following decade were better because of it, though the precise line from racing development to road car application runs through the R&D departments of Renault, BMW, Honda, and Porsche in ways that are rarely fully disclosed. The broader claim holds: the 1980s gave the road car industry a decade's worth of compressed learning about what turbocharging could do when the consequences of getting it wrong were sufficiently spectacular.

By the 1970s the principle had reached road car manufacturing. By 1981, when John Barnard built the McLaren MP4/1 from carbon fibre rather than aluminium, the monocoque's journey from Chapman's napkin sketch to the structural philosophy of the modern automotive industry was essentially complete.

It is worth noting some honorable mentions, where transfers were not always about components. In 1999, two doctors at Great Ormond Street Hospital in London, Martin Elliott and Allan Goldman, visited Ferrari's headquarters in Maranello after noticing that the handover of a critically ill child from the operating theatre to intensive care shared an uncomfortable structural resemblance to a pit stop gone wrong. Too many people, unclear roles, no rehearsed sequence, and catastrophic consequences for error. Ferrari engineers subsequently visited the hospital, and watched an actual patient transfer, to help redesign the process. The resulting protocol, documented in a 2007 paper in the journal Paediatric Anaesthesia, reduced handover errors by 42%.

These were some prominent turning points: engine behind the driver, shell carrying the loads, an aerospace composite, and an engine’s performance tweaked by a mini fan, among others. Formula One at its best created conditions in which ideas too radical for the road were forced to either work or fail in front of a watching world. And most failed! However, the ones that worked changed everything. And the man who had been loudest in his certainty that they would fail ended up building both of them himself.

By any serious technical measure, the 2014 Formula One Power Unit is one of the most extraordinary machines ever built by human beings. That is a claim this article is prepared to defend. The fact that most of the public experienced it primarily as a disappointment because the cars sounded like a high-powered vacuum cleaner compared to the V10-symphonies from their predecessors, is one of the more magnificent ironies in the sport's history.

The regulations that came into force that year mandated a 1.6-litre turbocharged V6 engine, which sounds modest until you understand what was attached to it. Two electric motor-generators, a battery, a control electronics unit, and a piece of engineering so advanced it has never appeared in a road car before or since:  the Motor Generator Unit, Heat, better known as the MGU-H. A device that sat on the shaft between the turbocharger's compressor wheel and its turbine, spinning at up to 125,000 revolutions per minute.

For context on that number: an ordinary road car engine typically redlines below 6,000 rpm, and even a high-performance sports car rarely exceeds 9,000. A Formula One engine in the 2014 era was capped at 15,000 rpm, which already puts it in a category most drivers will never experience. The MGU-H ran at 125,000. The only everyday object most people have encountered that operates at remotely comparable speeds is a dental drill, which spins at around 400,000 rpm but weighs a few hundred grams, runs on compressed air, and is expected to survive nothing more demanding than a squeamish patient.

So what did the MGU-H actually do? Two things. First, it harvested energy from exhaust gases that would otherwise have been expelled as heat and noise — gases that passed through the turbine, spun it, and in doing so also spun the MGU-H shaft, generating electricity that went into the battery. Something that was pure waste became a power resource! Second, and more cleverly, the MGU-H could run in reverse. It could use electricity from the battery to spin the turbocharger's compressor, building boost pressure without waiting for sufficient exhaust gas flow. This eliminated turbo lag — the frustrating delay between pressing the accelerator and feeling the turbo's effect — almost entirely. The engine responded instantly because the turbo was already spinning, kept at speed by its little electric motor.

The combined result was a power unit that achieved thermal efficiency — the percentage of fuel energy converted into useful mechanical work at the wheels — of fifty-two percent. That number is the headline of this article, and is also something that nobody outside the paddock ever really noticed.  To understand why this matters, consider that the average naturally aspirated petrol engine in a road car converts roughly thirty percent of fuel energy into motion. The rest escapes as heat through the exhaust pipe and the radiator. Formula One had effectively closed half that gap, through twelve years of relentless engineering under the pressure of a World Championship.

Renault's engine technical director Remi Taffin, speaking in 2018 when the MGU-H was first threatened with deletion for what was then planned as a 2021 rule change, said that all four manufacturers had initially proposed keeping it in any future regulations. He described it as irreplaceable for achieving the kind of energy efficiency that Formula One could claim to the world. Mercedes's engine boss Andy Cowell called it a "fantastic, efficient component." Ferrari's Mattia Binotto, who would later agree to drop it, said at the same time that removing it was "not the best choice" and would represent losing "a very high-efficiency technology."

These were the people who had actually built the thing, describing what they had built with complete accuracy. The MGU-H was the most impressive thermal engineering achievement in the history of automotive competition. Fifty-two percent, in a device that fit between a compressor wheel and a turbine shaft, lasting a race weekend.  

Mercedes tried to prove otherwise. The AMG One, announced in 2017 as a road car built around a Formula One power unit including a functioning MGU-H, finally reached its two hundred and seventy-five customers in 2022 and 2023. Five years late, at a cost of roughly two and a half million pounds per car. Mercedes' own CEO, Ola Källenius, later joked that the ones in charge had been drunk when the company agreed to build it. The MGU-H in the AMG One operated at a reduced specification compared to its racing equivalent, and even so the thermal efficiency of the road-going unit dropped from fifty-two percent to just over forty. That is what it costs to put the MGU-H in something a civilian can drive. But Formula One does it ten times a season and calls it routine. 

Then, sometime in late 2020 and into 2021, the meetings began. The Volkswagen Group sent representatives and so did Ford. Red Bull, freshly abandoned by Honda and facing the prospect of building an engine from scratch, needed regulations that did not require a decade of MGU-H expertise to be competitive. The FIA needed new manufacturers badly enough that it was willing to negotiate on almost anything. Even Mercedes and Ferrari, the two teams that had spent the better part of a decade mastering the MGU-H and whose competitive advantage over every rival was built substantially on that mastery, eventually came around. 

Mercedes had extracted eight consecutive Constructors' Championships from their understanding of the component. Ferrari had spent years closing the gap, and agreeing to delete it meant conceding to handing back an advantage that had cost hundreds of millions of euros to build. However, they agreed anyway. 

A power unit cost cap was offered in exchange, removing the threat of a fresh spending war with well-funded new entrants. Honda had already walked out, leaving the sport with three engine suppliers for ten teams, and Honda's own explanation was that the MGU-H era had consumed resources they could no longer justify committing to a single racing series. The slow realisation that the MGU-H had made Formula One an increasingly exclusive club helped most of all. Volkswagen walked into a room where almost everyone present had already quietly decided they could live without the MGU-H. 

The difference between a murder and a consensus is sometimes just a matter of who gets the blame.

Every car on the grid between 2014 and 2025 carried an MGU-H, but only four organisations in the world actually knew how to build one: Mercedes, Ferrari, Renault, and Honda. Everyone else — Williams, McLaren, Haas, Force India, and the rest of the customer field — simply bought a complete power unit from one of those four manufacturers and bolted it in. The buying price for a full power unit, MGU-H included, ran to over 10 million dollars per season per car. The customer teams paid that figure for hardware they had no hand in designing. And neither could they freely modify, due to an inherent lack of understanding it at the component level. They were, in the most literal sense, passengers inside someone else's most guarded engineering secret.

The Volkswagen Group, the German industrial giant whose brands include Audi, Porsche, Lamborghini, and Bentley, had been circling Formula One for years. Formula One is watched by roughly five hundred million people, and its audience skews toward exactly the demographic that buys German premium cars. Bernie Ecclestone, who ran the sport for decades, had dangled the prospect of a VW group entry at various points, and various VW brands had always found a reason to demur.

The reason, when it finally crystallised into a specific technical objection, was the MGU-H. Audi's engineers studied the 2014 power unit regulations and concluded that developing a competitive MGU-H from scratch meant starting with a decade-long disadvantage against Mercedes and Ferrari, who had been refining theirs since before 2010. Audi's road-car electrification strategy had no commercial use for a turbo-shaft motor-generator spinning at 125,000 rpm, and the investment required to develop one competitively would have returned almost nothing to their road car division.

Audi was not alone in that calculation. Red Bull had watched Honda formally withdraw from the sport at the end of 2021, leaving the team facing the prospect of building a power unit entirely from scratch for the first time in their twenty-year existence. Their new technical director, Ben Hodgkinson, had spent two decades at Mercedes-AMG High Performance Powertrains in Brixworth, most recently as head of mechanical engineering, before Red Bull persuaded him to cross the county and start again. He brought five named senior colleagues with him, all from the same Mercedes facility, covering engine electronics, energy recovery, combustion design, and production. A new Red Bull Powertrains facility was being built at their Milton Keynes base. Even with that talent, developing a competitive MGU-H in time for 2026 was, by any reasonable assessment, an unrealistic ambition. Red Bull pushed for its removal with as much force as Audi.

Porsche, meanwhile, had been pursuing its own entry through a proposed fifty-fifty partnership with Red Bull. Most people associate Porsche with Le Mans and the 911, but the company has a longer and more complicated Formula One history than that. They won their only race as a constructor at the 1962 French Grand Prix with Dan Gurney, quietly withdrew at the end of that season citing costs, returned in 1983 to supply McLaren with turbocharged V6 engines badged under the name TAG, won two constructors' championships and three drivers' titles with Alain Prost and Niki Lauda, and then disappeared again. By 2022, when they announced plans to re-enter as Volkswagen Group accelerated its F1 ambitions, the paddock treated a Red Bull partnership as a formality. Senior figures throughout Formula One spoke for months as though the deal was essentially signed. Unfortunately, it was not. Porsche wanted equal ownership of the team as the price of its involvement, and Red Bull refused. Horner described the breakdown as a matter of "strategic non-alignment," and Porsche's own statement confirmed that the partnership it sought, based on an equal footing across both the power unit and the team, could not be achieved. Their motorsport boss Oliver Laudenbach subsequently told reporters that Formula One was "not a topic for us" and that they were "not spending any energy on that." Ford, watching the wreckage of that negotiation, saw an opening. Mark Rushbrook, Ford's performance director, sent Horner an email asking if Red Bull wanted to talk. They did, and a partnership was agreed on Red Bull's terms, with Ford contributing expertise in battery cell technology, electric motor systems, and combustion engine development rather than demanding a stake in the team. 

The Honda reversal is perhaps the most telling data point of all, though the full story resists a clean single explanation. When Honda announced their withdrawal from Formula One in October 2020, they cited a corporate push toward carbon neutrality by 2050 and the need to redirect engineering resources into EV and hydrogen fuel cell development. Formula One's then-CEO Chase Carey told Wall Street analysts something rather more direct: that from his perspective, the decision was largely driven by economic challenges at Honda as a company. The MGU-H's complexity and cost were a contributing pressure, but Honda's own public statements pointed elsewhere. What brought them back was equally layered! Honda Racing Corporation president Koji Watanabe named the 2026 regulations' direction toward carbon neutrality as the decisive key factor in the return, saying it matched Honda's own corporate goals. The sustainable fuel mandate and the tripling of electrical power in the 2026 architecture, which aligned with Honda's electrification research, were the specific regulatory changes that made re-entry commercially justifiable. In their defense, the MGU-H's removal was part of that picture.

General Motors then entered the picture, committing to a power unit programme through GM Performance Power Units LLC, with FIA approval to supply their Cadillac team from 2029. Cadillac joined the grid in 2026 using Ferrari engines in the meantime. GM's engine commitment was a condition of Cadillac's grid entry rather than a cause of any regulatory change, but their presence mattered symbolically. Formula One now had American manufacturers on both sides of the Atlantic involved in engine programmes for the first time in decades, and the sport could argue, with some justification, that the simplified 2026 regulations had achieved something the MGU-H era never could: making engine development accessible enough that a manufacturer building a facility in Fishers, Indiana could plausibly plan to join the supply chain.

The teams that had spent a decade mastering the MGU-H, Mercedes and Ferrari, eventually came around too. A dedicated power unit cost cap was offered as part of the deal, capping engine development and supply spending for the first time in the sport's history and removing the threat of a fresh arms race with well-funded new entrants. The slow realisation that the sport had just watched Honda walk away, and that Renault would eventually follow, concentrated minds considerably. Honda's departure had been driven by a combination of corporate sustainability targets and economic pressure at a company navigating a once-in-a-generation shift toward electrification. Renault's exit from engine manufacturing came at the end of 2025, after years of uncompetitive results with a power unit that had never fully caught Mercedes or Ferrari. The sport that had once drawn four manufacturers into the MGU-H era was now in genuine danger of sustaining only two.

Then, Ferrari's Binotto confirmed the direction of travel as early as September 2021, telling reporters at Monza: "So far all the discussions we have had have been positive, and most or all of the manufacturers agree to remove the MGU-H." By July 2022, with the regulations still being finalised and Volkswagen's ultimate commitment still unconfirmed, Binotto was considerably more candid about what had actually been surrendered: "We removed the H, and we did it only to try to help them join F1. For us, removing the H is something which is not the best choice. It's a compromise because it's a technology we know pretty well. It's a very high-efficiency technology, which is great for F1."

Read that quote slowly. The team that had spent eight years mastering the most efficient thermal energy recovery device ever fitted to a racing car described removing it as a compromise made for the benefit of manufacturers that had not yet officially committed to the sport. 

To rub salt on the wound, Porsche never came at all! Audi arrived only by acquiring an established team rather than building one, and Ford came in through an email on Red Bull's terms. Honda came back after leaving specifically because the architecture had changed to match their corporate direction. Later on, General Motors signed up to build an engine from a facility in Indiana that had never produced a Formula One component in its life. Volkswagen walked into a room where almost everyone present had already quietly decided they could live without the MGU-H. 

The difference between a murder and a consensus is sometimes just a matter of who gets the blame. 

PART FOUR: WHAT THEY BUILT INSTEAD
Welcome to the 2026 Formula One Car. Please Mind the Gap in the Straight.

So what does Formula One actually have in 2026? The short answer is one thousand horsepower, split almost exactly in half between a petrol engine and an electric motor. The longer answer is considerably more complicated, but you should know about it.

The internal combustion engine, the bit that burns fuel, produces around 400 kilowatts, or roughly 540 horsepower. Alongside it sits the MGU-K, which stands for Motor Generator Unit Kinetic. Where the MGU-H sat on the turbocharger shaft and harvested heat energy from exhaust gases, the MGU-K sits on the drivetrain itself and harvests kinetic energy under braking, storing it in a battery and deploying it as additional thrust when the driver accelerates. It has been part of Formula One since 2014, running quietly in the shadow of its more glamorous turbo-mounted sibling. In 2026, with the MGU-H gone, the MGU-K inherits the entire electrical workload and has been upgraded accordingly, contributing 350 kilowatts of pure electric power, about 470 horsepower. Add the two systems together and you get approximately one thousand horsepower in total, the whole thing drawing on a battery that stores four megajoules of deployable energy, double what the previous generation used. All of it burning 100% sustainable fuel, with no crude oil components whatsoever. 

The fuel flow into the engine is capped at 3,000 megajoules per hour, which is the 2026 regulations' way of measuring fuel consumption by energy content rather than mass. The switch matters because sustainable fuels have different energy densities than fossil petrol, so measuring by kilogram would inadvertently favour whichever fuel happened to be denser. Pat Symonds, Formula One's then-chief technical officer (2017-2024), put it in terms anyone will recognise: the FIA is now telling teams how many kilowatt-hours of fuel energy they can use in an hour, rather than how many kilograms they can burn. 

The MGU-K is not an entirely new idea, though calling it a simple evolution of what came before would undersell how dramatically the concept has been stretched. Its ancestor appeared in Formula One in 2009 under the name KERS, which stood for Kinetic Energy Recovery System, a device that harvested energy from braking and deployed it as a modest 80 horsepower boost for a handful of seconds per lap. When the hybrid era began in 2014 the FIA formally renamed KERS as the MGU-K and integrated it into a far more ambitious power unit architecture, doubling its output to around 160 horsepower and connecting it to a wider energy management system that also included the MGU-H. The underlying principle of harvesting braking energy and redeploying it as thrust remained unchanged, but the engineering surrounding it had grown so substantially that the shared name was somewhat misleading. In 2026, with the MGU-H deleted and the entire electrical burden now falling on the crankshaft motor alone, the MGU-K has been upgraded to 350 kilowatts, nearly three times what it produced in the previous era and more than four times what KERS managed in 2009. 

The car is physically smaller than its predecessor, which matters because the predecessor was, by any standard, somewhat vast. The wheelbase — the distance between the front and rear axles — has been shortened to 3,400 millimetres. The car is 1,900 millimetres wide, and the whole thing weighs a minimum of 768 kilograms, thirty kilograms lighter than the 2025 machine. The wings are simpler. The aerodynamic floor has changed entirely. And there is a new active aerodynamics system, in which both front and rear wings physically move between two configurations: one that reduces drag on straights, and one that generates more downforce through corners.

On paper, this sounds like progress on every front. In practice, 2026 has introduced a problem that the MGU-H was specifically designed not to exist.

Adrian Newey, the most celebrated chassis designer in the sport's history, had joined Aston Martin the previous year and immediately asked Honda whether they could reposition the MGU-K, moving it ahead of the engine rather than behind and double-stacking the battery electronics in the process, creating the aerodynamic space that his designs have always demanded. Honda agreed! The result, as Newey told Watanabe at the Australian Grand Prix with the directness of a man who had concluded that diplomatic language was no longer serving anyone, was a power unit transmitting vibrations severe enough to cause nerve damage to his drivers. Fernando Alonso and Lance Stroll failed to finish two of the first three races of the season, with Alonso describing numbness in his hands and feet from the reverberations coming through the steering wheel. The MGU-H had been removed to simplify the power unit, and Newey's first act upon encountering that simplified power unit was to make it considerably more complicated, in pursuit of the aerodynamic advantage that has defined every car he has ever drawn. Formula One, as it turns out, cannot help itself. 

Here is what happens on a long straight in a 2026 Formula One car. The driver exits the corner and plants the throttle, both the engine and the electric motor firing simultaneously to deliver that full thousand horsepower. The car accelerates at a rate that would be frankly terrifying to experience from the inside. Then, somewhere in the middle of the straight, the battery runs out! In a jiffy, the 350 kilowatts of electric power simply disappear, and the driver is left with only the combustion engine pulling the car toward the next braking zone. Lando Norris, the reigning world champion who tested the 2026 McLaren in January, described the sensation with the kind of diplomatic understatement that professional racing drivers deploy when they mean something considerably more alarming: the car, he said, feels considerably less rapid. 

This is called clipping, and it is not a subtle effect. Norris put it starkly in his first public assessment of the new car after the Barcelona shakedown. The biggest challenge, he said, was battery management: "The biggest challenge is how you can recover the batteries as well as possible, and that's when it comes down to using the gears, hitting the right revs. Obviously, you've got some turbo lag now, which we've never really had before. All of these little things have crept back in."

The turbo lag he mentions is the other consequence of removing the MGU-H. The MGU-H had been used to spin up the turbocharger on the starting grid, giving drivers an instant response when the lights went out. Without it, drivers must now rev the engine considerably higher and for longer before engaging the clutch, essentially recreating the standing-start procedures of the 1980s turbo era. Gabriel Bortoleto, the Audi youngster, described his first practice start in Bahrain with considerable candour: "Oh man, it's complicated. The ten second thing and then after five seconds I already lost the count and then engines revving up, gear in and out, and you need to release the clutch. It's quite a mess. It was much easier last year."

Norris, to his credit, maintained public equanimity about all of this. "In a perfect world, I probably wouldn't have all that in a race car, but it's just F1. Sometimes you have these different challenges." That is either the considered acceptance of a professional making the best of a sporting regulation, or one of the great understatements in the history of motorsport commentary. Possibly both.

What is certain is that turbo lag and battery clipping represent a step backwards from the 2025 car. Formula One spent twelve years engineering both problems out of its power units, and the 2026 regulations have engineered them back in, because the solution to both required a component the regulations specifically forbid.

PART FIVE: THE ROAD RELEVANCE ARGUMENT
What the Industry Actually Needed

The official justification for all of this — the MGU-H removal, the simplified architecture, the pivot toward a fifty-fifty power split — is captured in a phrase that appears in every regulatory document, and every corporate announcement associated with the 2026 regulations: road relevance. The argument runs as follows. Formula One's power unit should resemble the technology in road cars, so that what is learned in the cockpit can migrate to the garage. The 2026 power unit, with its crankshaft-mounted electric motor alongside a petrol engine, resembles the mild-hybrid and plug-in hybrid architecture being pursued by mainstream car manufacturers. Therefore, 2026 is road relevant (and to an extent, progress).

There is something in this. The basic topology of the 2026 power unit — one combustion engine, one electric motor, and a battery sandwiched between them — does resemble the architecture of hybrid road cars from the VW Group, from Toyota, from Ford, from virtually every mainstream manufacturer planning for the next decade of electrification. Engineers working on the 2026 regulations will learn multiple things! This includes battery thermal management collated with electric motor efficiency at extreme power outputs. Additionally, there will be gaps filled around control software that manages energy flow between the two systems. Factually, much of that knowledge will be useful in road cars.

But here is the thing. Bosch had volume production of its 48-volt mild hybrid system running in factories by late 2018, years before the 2026 regulations were finalised. Toyota had been selling the Prius, the world's first mass-produced hybrid car, since 1997 in Japan and globally since 2000. The knowledge of how to make this topology work already existed, in vast quantities, in the road car industry long before Formula One decided it needed to learn the same lesson, through 1000 horses of combined might. The 48-volt mild-hybrid architecture that the 2026 power unit resembles at its most basic structural level is being manufactured in millions of units per year by suppliers whose names would be uncommon to the average Formula One fan.

When Cooper proved the rear-engine layout worked at championship level in 1959, road car manufacturers had not yet committed to it at scale, and the knowledge that flowed outward from the paddock gave the industry something it genuinely needed. When Barnard built the carbon fibre monocoque in 1981, he was industrialising a material that no road car manufacturer had yet attempted to use structurally, and the knowledge that came from doing it under racing conditions accelerated a transformation that the broader industry would eventually follow.

The common thread is that Formula One was ahead. It was doing things the road car world had not yet committed to, while generating knowledge that moved in one direction because there was nowhere else for it to go. But in 2026, that thread runs the other way! Formula One's own official documentation acknowledges the gap with a candour that is almost refreshing: in the eighteen years since the original hybrid power units were designed, the MGU-H technology was never adopted by any road car manufacturer, and the MGU-K, the component the sport is now tripling in power and celebrating as its centrepiece, was already in volume production on roads before the regulations were written.

PART SEVEN: THE LAB IS STILL OPEN
What Actually Transfers, and What That Means

It would be dishonest to leave the impression that 2026 represents a complete surrender of Formula One's engineering identity. As always, the picture is more complicated than that.

A component called the Turbulent Jet Ignition system was developed under the competitive pressures of the 2014 to 2025 engine era. It was associated particularly with Ferrari's power unit program, enabling petrol engines to burn their fuel more completely by using a small pre-chamber to create multiple ignition points simultaneously. The German engineering company Mahle, which commercialised the technology, now licenses it to road-car manufacturers. The result is better combustion efficiency from a given fuel charge, which translates directly into better fuel economy and lower emissions from any petrol engine that uses it. Unlike most of what Formula One claims as technology transfer, this one has a paper trail that leads all the way to a production line and a showroom. 

The one piece of 2026 technology that genuinely does push into unknown territory — the sustainable fuel program — is also the one that has received the least attention. From 2026, Formula One runs exclusively on Advanced Sustainable Fuel, with no crude oil components, produced from biological feedstocks or captured carbon, and certified to reduce lifecycle carbon emissions significantly compared to conventional petrol. The fuel must be drop-in — chemically compatible with existing engine hardware, without modification. The same molecules that go into a Formula One engine must, in theory, be able to go into the petrol tank of any existing car on the road.

This matters enormously if it works. There are hundreds of millions of petrol-powered vehicles currently in service around the world that will not be replaced by electric cars in any realistic timeframe. A certified, scalable, drop-in sustainable fuel that reduces their emissions without requiring any changes to the vehicle is potentially one of the most impactful climate interventions available. Formula One is providing the most demanding test environment imaginable: if Shell, BP, Castrol or Aramco produces a certified drop-in sustainable fuel that survives  the thermal conditions inside a race engine operating at over ten thousand revolutions per minute for ninety minutes, it can survive anything a road car will ever ask of it. This is definitely the Mobile Lab argument in its most compelling 2026 form, which should not be considered unreasonable.

Two other areas of 2026 development carry a strong transfer potential, though both require an honest qualifier before the claim can be made. One is battery thermal management. The 2026 MGU-K pushes the car's battery through a violent charge-discharge cycle that depletes a full charge in approximately eleven seconds at maximum deployment, while simultaneously generating between thirty and forty kilowatts of continuous thermal load that must be rejected without compromising aerodynamic efficiency. Road car engineers working on EV fast-charging infrastructure face a structurally similar problem at lower power densities: how to push large amounts of energy through a battery quickly without destroying it through heat. The solutions developed under racing pressure may well inform that work, though no manufacturer has yet confirmed a specific transfer pathway. 

The second is energy management software. Ford, in its public explanation of why the company entered Formula One, cited the sophistication of managing a 50/50 power split in real-time race conditions as directly relevant to their road car electrification programs. Whether that could lead into the creation of specific software tools finding their way into production vehicles remains, for now, a question rather than a fact. 

These examples, among many many more, matter, because they establish that the transfer mechanism still works when the conditions are right. However, the question hanging over 2026 is whether those conditions currently exist in the right combination.

The honest question is a simpler one: whether stress-testing someone else's idea at extreme conditions belongs in the same conversation as proving that an idea was right in the first place. John Barnard industrialised carbon fibre for automotive use when nobody else would go near it. The engineers who built the MGU-H produced the most thermally efficient internal combustion cycle in the history of motor sport for a petrol engine, and most people missed it entirely because the cars did not ‘sound right’. 

The paperwork from Wolfsburg changed what Formula One is. Mattia Binotto said so himself, in language so direct it almost constitutes a confession: we removed the H to help them come, and it was not the best choice. Somewhere in a facility in Brixworth or Maranello or Viry-Châtillon, the people who built that component understand exactly what was lost, because they built it and they know what it could do.

The rest of us are left watching a sport that held a meeting to eviscerate its greatest invention, and called the result progress. Well, when you think about it, this is exactly what the car industry has been doing for decades. Perhaps the transfer was more complete than anyone intended.

That is either poetic or damning, don’t you think?