I was wrong about the Prius

It was around 2014/2015 when I noticed that rather than being picked up by a leggy Prius with some concerning interior trim stains, London’s Uber drivers had migrated seemingly overnight to an almost two-tonne SUV, without doubt the least suitable vehicle for short trips around the centre of town this side of a pogo stick. But the Mitsubishi Outlander had arrived and quickly became the UK’s bestselling plug-in hybrid (commanding more than 60 per cent of all PHEV sales) due to some low company car tax rates and government plug-in car grants.

To me at least, this initially made a lot of sense. Lots of short journeys and off-street parking? Plug in every night and no local tailpipe emissions. Need to do 400 miles in one go? Fill it up, and off you go. A PHEV covers a lot of bases, right? Well, in theory, as long as you can plug it in every night. Spoiler: most people didn’t.

But PHEVs have got fat. And I don’t mean ‘I probably didn’t need to visit the buffet for a third time’ fat; I mean ‘I bought a pie shop, ate the profits and my trousers no longer fit’ fat. The original Outlander PHEV had a 12kWh battery. The current Range Rover Sport PHEV has a 38kWh battery, more than 20 times the size of the original Prius. The Lotus Eletre X goes even further with a 70kWh battery, although it is really a range-extender: its electric motors propel the car while the 2.0-litre combustion engine primarily operates a 150kW generator. The original Nissan Leaf managed to be an entire electric car with a mere 24kWh battery; we have somehow managed to invent the electric car that brings its own petrol station.

Either way, these massive batteries have taken us back to the ‘two complete powertrains’ argument. But does it matter? Does the weight of a few extra kWh actually make any difference?

Let’s do some simple maths here. Everybody knows heavier cars are less fuel-efficient, but using a 2026 Range Rover Sport P460e or P550e as an example, we can estimate that the battery pack and motor set-up adds about 350kg to the weight of the car, which is roughly four reasonably committed adults, or in the universally acknowledged petrolhead measurement of weight, more than three-quarters of a Caterham Seven.

Our cubic-inch enthusiast friends across the Atlantic have crunched the numbers on the impact of weight on fuel consumption. Modelling by the US government’s NHTSA and EPA found that with the same engine (i.e without downsizing the powertrain), a 10 per cent mass reduction produces about a 3.5 per cent reduction in fuel consumption.

If you also resize the engine so the lighter car retains the same performance (you don’t need as much power to shift it along) the saving rises to roughly 6.5 per cent per 10 per cent mass reduction. As a cross-check, the US Department of Energy gives the broader industry rule of thumb of approximately 6-8 per cent better fuel economy for a 10 per cent vehicle-weight reduction where the whole vehicle is optimised.

Why isn’t a 10 per cent heavier car simply 10 per cent thirstier? Because air doesn’t care what your car weighs. At motorway speed, aerodynamic drag increasingly dominates; around town, mass matters much more. Which is yet another argument against Chelsea tractors blocking London’s arteries.

With our Range Rover, the shift from a mild hybrid with a small 0.2kWh/48V system to a 38.2kWh/400V/350kg PHEV system brings a 15 per cent weight gain, so perhaps 5 per cent worse fuel consumption on a weight-alone basis, or 10 per cent if you were to upsize the engine as well to compensate. But when you slow down in a non-hybrid car, you can’t harvest that kinetic energy to make more petrol. With modern EV powertrains recapturing much of the energy that would otherwise be lost as braking heat, a hybrid can more than offset the extra weight of its batteries in stop-start driving. Though, as we’ll see, you don’t need anything like 38kWh to do it.

So why did PHEVs get so fat? In Europe, at least, much of the blame lies at the door of the official testing protocols, currently called WLTP, and its even less accurate ancestor, NEDC. Here I do have some sympathy for the ‘powers that be’ because they need something to judge the cars on, but just like any form of motorsport, give engineers a test and they will optimise for it. This is not dishonesty; this is essentially the job description. Give them a loophole and they’ll have three CAD models of it before lunch. Or, in the case of VW a decade ago… cheat.

We can’t blame the manufacturers here; they were simply optimising for the rules they were given and us punters were voting with our wallets for those cars which cost least to tax. Like the rest of us, the European Commission knew full well that in the real world, PHEVs in particular weren’t getting anywhere close to the CO₂ figures that their tests reported. Data from the cars’ own instruments led to the mildly awkward discovery that 2021 PHEVs were emitting, in real use, about three-and-a-half times their official CO₂ figure. In exam terms, this is less ‘margin of error’ and more ‘we appear to have revised for geography, rather than chemistry’.

Adjustments were made at the beginning of 2025, with more to come in 2027. However, like it or not, regulatory incentive has helped push hybridisation towards energy capacity, rather than necessarily towards maximum whole-car efficiency. This has led us towards ‘energy hybrids’: cars that store as much electrical energy as possible by stuffing themselves full of batteries.

And that has taken us down a vicious circle which would have Colin Chapman up to the rev limiter in his grave. More kWh means a heavier battery, which means more strength needed in the chassis, bigger brakes, fatter tyres and so on… which requires more batteries to travel the same distance. So, there must eventually be a point where adding another kWh to an ICE car produces diminishing or negative whole-vehicle returns. Surely?

This sets up the fundamental question: how much electrical energy does a hybrid actually need to store? To answer this, I’d like to talk about bicycles. Yes, bicycles.

A few years ago, in the pub, I came up with an idea for a new type of bicycle wheel. Rather than thinking about range, I focused on wasted energy. My logic was that electric bikes are all well and good, but they’re heavy and focused on range rather than efficiency. So I designed a replacement front wheel which stored just the kinetic energy you would otherwise lose during braking. The idea was that, as you came to a stop, rather than burn that energy off as heat in the brakes, you stored it. Then, when you got going again, you hit a button and, hey presto, you were back up to speed.

After several pints and rather more engineering enthusiasm than the situation warranted, I discovered that an 80kg rider on a 20kg bicycle at 20mph carries about one watt-hour of kinetic energy. One. To put that into context, a single AA battery has three to four times that energy stored in it.

Back to four wheels, let’s suppose we’re not asking the car to drive 50 miles electrically. We only want to capture energy that would otherwise disappear as brake heat and then redeploy it. The crucial question is, how large does the storage device then need to be?

Let’s take a 1500kg sports car like a 992-series Porsche 911 Carrera. At 70mph it has 0.20kWh of kinetic energy; at 100mph, 0.42kWh; and at 180mph, 1.35kWh. So if we’re just looking to recapture and redeploy otherwise wasted kinetic energy then we don’t need a big battery at all, do we?

At this point, dear reader, you will be rolling your eyes and pointing out that Porsche already does this with the new 911 T-Hybrid. It has a lightweight 400V, 27kg battery with just 1.9kWh of capacity, while the entire hybrid system adds only around 50kg to the car. ‘Stick to the little cars, Hedley, you’re just embarrassing yourself.’ And you’d have a point. Porsche clearly arrived at much the same conclusion: a hybrid doesn’t need much stored energy, it just needs to move that energy around intelligently.

The gearbox-mounted electric motor can add 40kW (54bhp) and 150Nm (110lb ft), while the electrically assisted turbo can actually recover otherwise-wasted exhaust energy and feed up to 11kW back into the electrical system. In other words, Porsche isn’t trying to make the 911 drive very far on electricity. It is using a tiny battery to stop wasting energy in the first place.

There are some other examples in recent times of these hybrids with relatively small battery packs, or ‘power hybrids’. The McLaren W1 has a 1.4kWh battery and gives a 255kW (342bhp) boost; the Ferrari F80 has a 2.28kWh battery that can charge or discharge at up to 242kW (325bhp). Chevrolet, Honda, Lamborghini and Koenigsegg have all been rummaging around in the same cupboard and coming to approximately the same conclusion.

We all know EVs can give massive power without breaking a sweat, so why are these outputs relatively small in the grand scheme of things? Why aren’t they braking from 150mph, harvesting all the kinetic energy then spitting it out in thousand-horsepower lumps the other side of a corner and showing these newfangled EVs who’s boss?

The catch is that batteries are rather like guests at a wedding with a free bar. How much they can hold is one thing. How fast you can pour it in is another. And how fast it all comes back out is a third – best not discussed at a wedding, but the entire point of a battery. It’s all down to the ‘C’-rate, which is the rate at which it can be discharged, relative to its maximum capacity. A 1kWh battery at 1C can deliver 1kW for an hour. The same-sized pack at 10C could deliver 10kW, but would be empty in 6 minutes.

A normal energy-dense EV cell might be happiest at only a few C, while purpose-built high-power cells can go far beyond that for short periods. Ferrari’s F80 is the slightly deranged proof: its tiny 2.28kWh battery can accept or release 242kW. That’s over 100C for very short periods. Normal EV cells would treat that request the way you’d treat being asked to run a marathon in flip-flops. Achieving numbers like that means sacrificing energy density and throwing expensive cells, cooling and power electronics at the problem. Solid-state batteries may improve the trade-off, but they don’t magically abolish it.

And that’s the snag with making the battery smaller. For otherwise identical cells operating at the same C-rate, halve the battery and you halve the power it can accept or deliver.

There is also the question of whether the EV horsepower wars might be reaching a plateau. Outside of specialist motorsport cells, traditional energy-dense lithium-ion batteries – the type you need for lots of kWh and as little weight as possible – were never designed to be sprinters.

And yet the Rimac Nevera R already manages 1550kW (2080bhp), around 14C. BYD claims a 30C peak discharge rate for the U9 Xtreme and around 2200kW (2950bhp) of output. Chemistry will doubtless keep moving, but at some point tyres, cooling and the basic question of what on earth you’re going to do with another thousand horsepower become more pressing than the battery. So it’s unlikely we’ll be discussing the 5000bhp XPeng Gigantor in the pub any time soon. But perhaps not impossible.

And this leads us to Formula 1 and its relatively small batteries. An F1 battery isn’t really being used like a fuel tank. It’s being used like a golden bucket in a very energetic fire brigade: fill it, empty it, fill it again, preferably several times before anyone has time to ask what it cost.

This week’s F1 ruleset (correct at time of writing, possibly not at time of reading) says that the cars’ batteries can only change their state of charge by 1.1kWh and deliver a maximum of 350kW (470bhp) for very short periods (All 1.1kWh would be used in just 11 seconds at that power). That’s the equivalent of more than 320C – three times the F80’s already deranged figure.

This is achieved with vast budgets, huge brains and no doubt a decent level of skullduggery, but the principal materials used are unobtanium and unicorn horn, so we’re not going to be able to drop one into the back of a 911 any time soon, not least because it would cost many times more than the car itself.

So we’ve come full circle. Smaller batteries mean lighter cars and all the benefits they bring. But at sensible road car cost, a very small battery struggles to swallow hundreds of kilowatts. That 1500kg Porsche 911 stopping from 100mph in, say, 4.5 seconds needs to be able to harvest an average of around 250kW (335bhp) once system inefficiencies are taken into account.

The problem is not finding somewhere to put that 0.3 kWh. That’s almost comically little energy, about 80 AA batteries; the problem is trying to ram it into a battery at 250kW. And a further issue: every stop is another charge cycle, thousands a year, and lithium-ion cells are rated for somewhere between 300 and 5,000 depending on chemistry and quality – ask anyone with a six-year-old iPhone.

Which is why I would like to propose a solution, and for that I need to take you back to my bicycle wheel. What made it different was that it didn’t use batteries at all. To store energy it used something called a supercapacitor.

Supercapacitors are terrible batteries. This is important. If you wanted one to power your EV to Scotland, you’d need one approximately the size of Scotland. But ask one to swallow a ridiculous amount of electricity for two seconds and immediately dump it back out, and suddenly it becomes extremely interesting. They store electrical energy using physical electrostatic fields rather than slow chemical reactions. The upside is they can take power in and fire it back out very quickly and can do it for millions of cycles with very little degradation. The downside is that they are less energy-dense than lithium-ion and more expensive for the same storage capacity.

But here is why I think supercapacitors could be the next big thing in hybrid systems: we’re agreed that we don’t need much capacity if we’re only storing 0.20kWh of kinetic energy. That would be 0.8kg of lithium-ion cells or 25kg of supercapacitors. You’d also need power electronics gubbins to go with it, but they’re getting cheaper and more efficient all the time. And the cost? About £100 for the lithium-ion cells, and about £2500 for the supercapacitors. So expensive, but not compared to the cost of a new 911.

But this is where it gets silly. Those 0.8kg of ordinary lithium-ion cells at 10C could output 2kW (2.7bhp) of power for six minutes. Off-the-shelf supercapacitors could put out as much as 450kW (600bhp) for around a second and a half. And what a second and a half that would be.

Combine that with very power-dense electric motors from companies like YASA, whose 12.7kg prototype has produced 750kW (1006bhp), and the overall weight gain starts to look acceptable for a performance car. For reference, 12.7kg is about the weight of a spaniel. A thousand horsepower spaniel.

At which point I assumed I had invented something terribly clever. Which is when Google informs you Toyota built it back in 2012. It dabbled with supercapacitors in its 2012-13 TS030 Hybrid Le Mans car which stored regenerative braking energy and provided 300bhp of electric boost. In 2019 Lamborghini put the idea into the Sián and didn’t use a battery for its hybrid system at all. A supercapacitor caught energy under braking and immediately returned it through a 25kW (34bhp) motor mounted in the gearbox. Lamborghini didn’t even make much of its energy capacity. Instead, it talked about current, power and weight, so it was no doubt pretty tiny.

Of course, I know this rabbit hole well because we nearly did something very ambitious / insane (delete as appropriate) with one. We looked at fitting a supercapacitor system to our Ferrari Testa Rossa J a few years back. It has three 1.8kWh battery packs which can output up to 5 kW (7bhp) each. Our idea was to swap out one of those packs for a supercapacitor pack to give short-term monstrous boosts to power. We estimated that we would only get 0.1kWh of storage, but at the press of a button you could in theory draw 100kW (134bhp) of power for 3.6 seconds.

To put that in context, it would give our 75 per cent scale Ferrari the power-to-weight ratio of almost 600bhp/tonne, or about the same as a LaFerrari or SF90. When we weren’t drawing full power we’d also trickle-charge that supercapacitor pack from the other batteries so it was ready for an on-demand boost at any moment. Fun? Definitely. Entirely safe and usable in a 75 per cent scale Ferrari? Probably not.

But this is where I see the opportunity. I suspect the best future solution wouldn’t be supercapacitors on their own. It would combine them with a small high-power battery, with the supercapacitors taking the shock loads and big braking inputs and being able to spit out great gobs of power very quickly. It could even be topped up when not all the ICE power is needed so that you have that explosive ‘push to pass’ system ready to go at a moment’s notice.

For urban transport, pure EV probably remains extraordinarily difficult to beat: low speed, lots of regeneration, short journeys, no local tailpipe emissions. For sustained high-speed motorway use, liquid fuel’s energy density remains enormously attractive. For an enthusiast/performance car, perhaps the sweet spot is therefore: efficient ICE plus 1-2kWh of extremely high-power hybrid storage, aggressive regeneration and, perhaps eventually, renewable e-fuel. It’s not an electric car with an engine attached. It’s not a petrol car with a token 48V starter-generator. It is a combustion car that simply refuses to waste energy unnecessarily.

And you know what? I’m pretty sure this new genus of performance car will have about the same energy capacity as the original Prius’s battery back in 1997 – and, give or take, the same working window as a 2026 F1 car. Maybe Toyota wasn’t carrying too little battery. Maybe everybody else started carrying too much.

And what happened to the bicycle wheel? Well, we built a prototype at Cambridge University, had some great conversations with the team at Prodrive, but then I made a dog’s breakfast of applying for a patent myself rather than using an agent and never got it granted. Shortly afterwards I started building Bugattis for a living, which was arguably a much more fun use of my time, and the bicycle wheel quietly disappeared into the shed of good ideas, badly patented.

Ben has chosen to donate his fee for this article to The Forces’ motorsport charity, Mission Motorsport. If you’d like to donate too, click here