Carbon plates and super-shoes: what they do, and what they cost

It is a thin, moulded sheet of carbon-fibre composite, curved along its length and buried inside the midsole foam. You never see it and you cannot feel it as an object. What you feel is what it does to the foam around it: the shoe stops bending where your foot bends and starts rolling instead.

The mechanical job is to stiffen the forefoot. In an ordinary shoe your toe joints flex hard at push-off, and the muscles that hold that joint together do work which never becomes forward motion. A stiff plate resists that flex, so less energy leaks out at the toes. At the same time the plate holds the curve of the sole, which turns the shoe into a rocker: instead of bending, it tips you forward onto the next stride.

That is the whole mechanism, and it is worth knowing how narrow it is. A plate does not add energy. It reduces one particular loss and redirects the rest. Everything else people claim for a super-shoe — the bounce, the fresh legs at 35 km, the feeling of being pulled along — comes from the material the plate is sitting in.

Plates in running shoes are not new. Racing flats carried stiffening shanks for decades and nobody wrote articles about them. What changed was the foam: a generation of supercritical polymer midsoles that are simultaneously very light, very soft and very good at giving back the energy you put into them. Put a stiff plate in a firm, dead foam and you get a plank. Put it in a tall stack of the new stuff and you get a shoe that behaves unlike anything before it.

Three things have to line up, and the plate is only one of them.

• The foam: light enough that a 40 mm stack still weighs nothing, and resilient enough to return most of what you compress into it. This is the expensive ingredient and the one that ages fastest.
• The plate: stiff enough to hold the foam in shape under a landing that would otherwise crush it sideways, and shaped to set the rocker.
• The geometry: where the rocker sits, how tall the stack is, how wide the base is. Move the rocker a centimetre and the same plate and foam feel like a different shoe.

You can watch the system working in the catalogue’s energy-return scores. The 58 shoes here with no plate have a median of 58 out of 100. The 6 with a nylon plate — softer, springier, much cheaper to make — jump to 81. The 14 carbon shoes sit at 92.5. That last step, from nylon to carbon, is smaller than the step from nothing to nylon, which tells you where most of the effect lives.

It also explains why nylon-plated shoes are a genuinely good answer for a lot of people. All six here cost between €160 and €180, median €175, and they keep a durability score of 73.5 against carbon’s 48. That is most of the feeling for a fraction of the consequence.

Almost everything you have read about super-shoes traces back to one kind of experiment: a laboratory running-economy test. A small group of well-trained runners run on a treadmill at a fixed speed while their oxygen consumption is measured, first in the new shoe and then in a named control shoe. Economy is the outcome — how much oxygen it costs to hold a given pace. The headline figure the whole category is named after, around four per cent, is a figure of that kind.

Three things about that are usually left out. The first is that running economy is not a race time. It is an input to one, alongside how fast you can run at threshold, how much you can hold, what the course is like and what the day is like. A shoe that improves your oxygen cost by a few per cent on a treadmill is not promising you a few per cent off your marathon.

The second is the spread. The averages get quoted; the individual results rarely do. In the published tests the response varies a great deal between runners, and some runners in those studies got nothing at all or ran slightly worse. Nobody has a way to tell you in advance which one you are, which is the honest reason a shop assistant cannot answer this question.

The third is the comparison. The percentage is always against something — usually a well-known racing flat, sometimes a marathon shoe of the previous generation. It is not a comparison against your daily trainer, and it is certainly not a comparison against the best modern shoe you could buy for €160.

The catalogue answers the money question directly, because it holds all three kinds of shoe side by side.

The median carbon shoe here weighs 214 g against 278.5 g for the median shoe with no plate. That is about 64 g a foot — roughly a quarter of the weight gone — and it costs a median €250 against €150. Say it plainly: a hundred euros for sixty-four grams and a much livelier ride.

Two things in that table are worth staring at. One is that the carbon price range starts at €160, not €250: the cheapest carbon shoe here costs the same as plenty of unplated daily trainers, weighs 215 g, and scores 87 for energy return. So "carbon means €250 or more" is simply false, and if the plate is what you want, the entry price is far lower than the category’s reputation. The catalogue’s dearest shoe is €310, and the gap between those two is bigger than the gap between the cheap carbon shoe and a good trainer.

The other is that the lightest shoe in the whole catalogue, at 170 g, has no plate at all. Light and plated are not the same property. If what you actually want is a shoe that disappears on your foot for a 5K, the one shoe here that undercuts every carbon shoe on the scale is unplated: the 170 g Streakfly, at €160. How much to spend on running shoes works through where the money stops buying anything.

The price tag is the cost people argue about. It is not the one that matters. The catalogue scores durability out of 100, and the three tiers separate more sharply on that than on anything else: 80 for the median unplated shoe, 73.5 for nylon, 48 for carbon.

The spread inside the carbon group is the interesting part. The scores run from 38 to 75 — and not one of the 14 carbon shoes reaches the median of the unplated group. The six worst-scoring shoes in the entire 78 are all carbon racers, and the very lowest score in the catalogue, 38, belongs to the most celebrated race shoe of the lot. Their own caveats say it out loud: fragile, save it for races and key sessions.

Those are relative scores rather than a mileage figure, and no honest catalogue can tell you how many kilometres a shoe will last — it depends on your weight, your gait, your surfaces and the weather. But the ranking is consistent and it is not subtle. The soft, resilient foam that makes the shoe fast is soft, and the outsole rubber is thinned out to save the grams you paid for.

So do the arithmetic the way it should be done: cost per race, not cost per pair. A €250 shoe kept for goal races and a handful of sharpening sessions is a defensible purchase. The same €250 shoe used as a daily trainer is the most expensive way there is to run easy kilometres, and it will be finished long before a €140 trainer bought on the same day. When to replace running shoes covers what "finished" actually looks like.

People who buy a super-shoe on a Friday and run easy in it on a Saturday usually come back confused. It felt tippy. It felt like hard work. The catalogue agrees with them: median stability score 41 out of 100 for the carbon group against 62 for the shoes with no plate.

The reason is in the design. A tall stack of very soft foam is not a stable platform on its own; the plate is part of what holds it together, and the rocker geometry is tuned to a particular speed and a particular way of landing. Run fast, land where the rocker expects, and the whole thing works as a lever. Jog, land further back, drift off the centre line on a cambered pavement, and you are simply balancing on a soft tower with a stiff spine.

The tall part is not an accident. For road racing, World Athletics caps legal shoes at 40 mm of sole thickness and a single rigid plate, and the brands have built right up to the line: 9 of the 14 carbon shoes here are exactly 40 mm at the heel, and 12 of the 14 are 39 mm or more. That is a rule showing up as a shape. Stack height and drop explained covers what those numbers mean and why the box usually prints only one of them.

The practical consequence: a super-shoe is a bad shoe for the run where you are tired, distracted, on uneven ground or going slowly — which is most of your running. It is also why these belong in a rotation rather than on their own. Building a shoe rotation is the companion piece to this one.

This is the part the marketing never writes, so here it is without hedging. A carbon shoe is a poor purchase if any of these describe you.

• You do not race. If there is no day the time matters, you are buying a shoe that is worse at everything you actually do, and buying it at a premium.
• You run three times a week and own one pair. The super-shoe cannot be that pair. You would be spending €250 to get a shoe that wears out fastest and handles easy running worst.
• You are new to running. Your gains for the next two years come from consistent weeks, not from equipment, and a shoe that punishes a wandering stride is a strange first shoe. Start at how to choose running shoes .
• You need a stability shoe and know it. Six of the 14 carbon shoes here are built on a narrow last and none is a stability model; the platform is the opposite of what you are looking for. Stability versus neutral shoes explains the choice.
• Your budget is one pair a year. Two €150 trainers alternated will serve you better across twelve months than one €250 racer and nothing else.

One thing not on that list is injury. It is tempting to say a plate will protect your calves, or wreck them, and neither claim is supported. Shoes have not been shown to prevent running injuries, and the strongest predictors of getting hurt remain training load and injury history rather than what is on your feet. If something hurts, a new shoe is not the answer and a clinician is — shoes and running injuries sets out what the evidence does and does not show.

Wanting a fast shoe for a race you care about is a perfectly good reason to buy one. It just pays to buy it deliberately.

1. 1

   Decide which tier you are in

   Chasing a personal best at a goal race is one case; wanting a shoe that feels quick on Tuesdays is another. The second is a nylon-plated super-trainer at €160–180, or the €160 carbon option, rather than a €280 racer.

2. 2

   Try it on a fast run, not a jog

   These shoes flatter the pace they were built for and feel awkward at any other. Judging one at recovery pace tells you almost nothing about how it will behave at race effort.

3. 3

   Size it with the fit you will race in

   Many of these are cut narrow — 6 of the 14 here — and a foot swells over a marathon. Leave a thumb of room at the front and check width at the widest part, not just length. Getting the fit right has the method.

4. 4

   Run in it before race day, but ration it

   Two or three faster sessions is enough to learn the rocker and find out whether your calves object. More than that and you are spending your race shoe on training.

5. 5

   Keep it out of the rest of the week

   Every easy kilometre in a racer is a kilometre of its short life gone. Put those in a trainer and let the plated shoe do only the job it is good at.