How ‘shoe doping’ changed marathon times forever – in ways we still don’t fully understand

Five years ago today, in a park in Vienna, Eliud Kipchoge ran the most significant marathon ever.

The clock stopped at 1:59:40. Kipchoge was already a legend. He had won 10 consecutive marathons between 2014 and 2019, including Olympic gold in Rio de Janeiro in 2016.

This, though, was something else. A sub-two-hour marathon was the sport’s Holy Grail. File it next to Roger Bannister’s four-minute mile in 1954.

The course design (start and finish location, and elevation changes) were perfectly legitimate but Kipchoge’s time wasn’t ratifiable, under World Athletics rulings, to be a world record — because of hydration delivered by bicycle, a rotating group of 42 pacemakers and a car that projected lasers to aid pacing and formation.

Still, Kipchoge was a superhero in a supershoe. He’d been agonisingly close to this two years earlier when he ran 2:00:25 at Nike’s Breaking2 in Monza (also not under record-legal conditions) in 2017.

The Vienna run was in Nike’s Alphaflys, the brand’s latest technology born out of their industry-breaking Vaporflys in Monza — the ones with the big midsole and the curved carbon plate.

This is the story of how the marathon has changed, told by some of the people behind the technological revolution.

Prepare yourself for an avalanche of statistics that all tell the same story — the marathon has become astonishingly fast. The top 14 men’s and top five women’s marathon times have all come since 2016.

Three men’s world records have dropped the time by 144 seconds, a bigger improvement than between 2003 and 2014 (118 seconds across eight world records). This week was the anniversary of Kelvin Kiptum’s 2:00:35 world record in Chicago last year.

GO DEEPER

Kelvin Kiptum: The tragic loss of the marathon runner who was set to redefine the sport

Paula Radcliffe’s mixed (2003) and women’s only race (2005) world records lasted over a decade, but have both been beaten twice since 2016, improved by 212 and 86 seconds: Tigst Assefa ran 2:11:53 in Berlin in 2023 and Peres Jepchirchir won London, this April, in 2:16:16.

Former GB runner and current athletics TV commentator Tim Hutchings describes an “orgy of records. They are dropping every week and they’re too common”.

It is much easier to measure the impact of supershoes in races, however the benefit of training — reducing loading, enabling athletes to run farther, faster and harder — in them is likely even more significant.

Likewise, iron sharpens iron. Marathon running is done individually but succeeds best when athletes work in packs, either with official pacemakers or using each other to share workloads and draft. One person being faster helps another.

Across the board, the frequency of top performances per calendar year now reflects what used to happen in a whole decade.

There were more than 100 sub-2:24 women’s marathons and sub-2:07 men’s marathons in 2022 and 2023. That threshold should be exceeded again this year.

Between 2001 and 2010, there were a combined 105 instances of sub-2:24 women’s marathons, and 138 sub-2:07 men’s marathons.

The average winning time of a marathon major (Tokyo, Boston, London, Berlin, Chicago, New York) in 2023, compared to the 2013-16 average, was four and a half minutes faster for women and three and a half minutes quicker for men.

This summer’s Olympic Games was the best demonstration of the benefits of supershoes. Paris boasted the toughest Olympic marathon course: three huge, punishing hills in the middle, and warm, humid August conditions.

Kipchoge was a DNF for the first time. Kenenisa Bekele, another great, was down in 39th and ran his second-slowest marathon.

Those two might be in their twilight years, but this was the first time the Olympic marathon record was broken in the men’s and women’s race at the same Games.

Kipchoge with INEOS CEO Sir Jim Ratcliffe after the sub-two-hour run in Vienna (Alex Halada/AFP via Getty Images)

So, what do the shoes actually do? The answer is long and complicated.

“You have four joints that are heavily used when you’re running,” says Maxwell Muttai, a gait analyst with an MSc in biomechanics. “That’s your hip, knee, ankle and the metatarsals — the MTP joint. It represents the heads of the toes.” That joint is commonly referred to as the MTPJ (metatarsophalangeal joint).

“Research was able to see that the energy put in on the MTPJ is not returned when pushing off,” explains Muttai.

“The carbon plate was to try and increase the stiffness and keep the MTPJ from bending at all — make the shoe really, really stiff, make sure that there’s no bending and therefore no energy loss.”

Initial research, from the University of Calgary in the early and mid-2000s, analysed jumping and sprinting performance in shoes with straight carbon plates. They put one into a marathon shoe in a 2006 study, got sub-elite athletes to run on treadmills, and found a 1.3 per cent benefit in terms of energy reduction.

“Then nobody did anything for about 15 years. That was kind of interesting,” says Benno Nigg, the founder of the Human Performance Laboratory at the University of Calgary. It was his PhD students who completed that study.

“We expected it would be: the stiffer, the better,” he says of the plates. “The results were not that, but somewhere that has an optimal (stiffness).

“So, Nike, people from our lab that went there to work there, came up with that Vaporfly. But they had a lot of things in their shoe. One thing was the plate — that was not straight, it was bent. Now is that important?”

The short answer: yes.

“If you just use a flat plate, it makes it actually harder for the calves to push up. That’s where this curve came in,” says Wouter Hoogkamer, lead author on the 2018-published paper of the University of Colorado study behind Nike’s ‘4%’ Vaporfly.

“Nike provided us with these new shoes that, at their lab, looked really good — they wanted to see if they were really that good,” explains Hoogkamer.

“We set up at that point (2016), one of the most in-depth and best controlled studies on running shoes.

“We were tasked to find 18 really good runners that would all fit that shoe. We were in Boulder, Colorado — it’s very athletic, so we managed to do so.

“We could test them at speed — not at two-hour marathon pace, but pretty close. We did the study over a range of speeds (from 3:00 to 2:20 marathon pace)”.

Athletes repeated five-minute trials in two control shoes (the fastest marathon flats at the time, which had less foam and no plates) and the Vaporflys’ then-prototype.

“Every single runner used less energy while running in those new prototype (Vaporfly) shoes. On average, that was four per cent. Nike was super excited, they named the shoe after our results.”

Critically, supershoes do not make people faster. They return energy better, which means athletes can run at the same speeds with reduced metabolic demands. Athletes are the ones running faster — a fussy but important distinction.

More on the sporting summer in Paris…

Comparing the same two control shoes, and with some of the same athletes, the University of Colorado did a follow-up study on the Vaporflys in 2018. It focused on biomechanics.

The Vaporflys did not change how athletes used their hips and knees, though there was reduced ankle movement. That was not enough to explain the four per cent.

“It’s no question that the shoes affect performance,” says Nigg. “Question is: how much is this part? How much is the material? That is not solved.”

Nigg says he is “lonely” in a belief that the shoe design and curved plate create a ‘teeter-totter’ effect, a beneficial rocker motion that propels runners forward.

Hoogkamer isn’t convinced. He says he’s “pretty confident that’s not the idea that went into the shoes. It might be happening, but it wasn’t the idea, and it’s also pretty hard to disprove. So there is no strong evidence for it”.

The clearest answer is that the obsession with the plates should be on the mid-sole foam. It is not just having more of it than traditional racing flats and creating longer legs, but also better, modern material(s) and its interaction with the plate.

“Foams are under shoes to cushion, (for a) softer landing,” says Hoogkamer. “Traditionally, they have been a sort of wet sand or a pillow (material). It’s soft, but you’re not getting anything out of it. You’re losing energy.”

Eliud Kipchoge’s shoes for his first sub-two-hour attempt in 2017 (Pier Marco Tacca/Getty Images)

Muttai explains the benefits of modern midsoles. They’re “very cushioned, which are very compliant, the way that you go down and they release the loading on your joints, and they’re also resilient”.

Compliance and resilience are keywords. Compliance: how much the shoe’s sole deforms when it hits the floor. Resilience: stored and returned energy.

“They’re still predominantly losing energy, not generating energy. You lose less (in super shoes) because foams are better,” says Hoogkamer.

“So rather than every landing, losing a lot of energy and having to regenerate them with their muscles, now you can store some of that energy in a shoe and get that back.”

As part of the 2018 University of Colorado study, they put the Vaporfly into a machine actuator. It looks like a hydraulic press, allowing a set force to be applied and energy return/loss calculated. The two racing flats returned 75.9 per cent and 65 per cent energy. The Nike Vaporfly gave back 87 per cent.

Individually, the foam and plate have marginal benefits, but collectively are worth more than the sum of their parts. Hoogkamer says, “They work together. If you land on your heel on a foam without a plate, the foam just deforms. If you have a plate under it, the load gets spread out over more foam.”

This helps athletes maintain rhythm, balance and stride consistency. Muttai discussed that as an under-appreciated part of super shoes when The Athletic had their gait analysed at Adidas’ Flagship Store in London.

(Liam Tharme/The Athletic)

Academics are trying shoe combinations with wacky variations of plates and foams, even ones that would not be run in, hoping to explain the numbers. Often, they don’t add up.

“We did a study on the four per cent, took a table saw and sliced it up so that the plate was no longer providing any bending stiffness,” explains Hoogkamer.

“Cutting down the plate didn’t change energetics much, about half a per cent,” he adds. That finding goes against initial explanations for its benefit.

Research is limited by current methods and medical equipment, which makes it hard to understand energy transfer/loss at muscular and tendon levels. Hoogkamer uses MRI scanners at the University of Massachusetts, which “you can’t really run in.” Some opt for ultrasound devices and others for muscle oxygenation sensors.

“It’s not easy to do these experiments because these forces are not that big,” says Nigg, “and there are a lot of other forces that act, other components of forces, that you have to have under control.”

Most elite distance runners look similar: lean, short-ish, with long legs and skinny calves. Yet indistinguishable differences in form and body composition mean some improve in super shoes by, according to Hoogkamer, six per cent, whereas others get worse.

Soh Rui Yong, the Singaporean national record holder in the marathon, speaks of his friend, Molly Huddle.

“She identifies as a non-responder. She was the best female athlete in the US, for distance. She ran the (U.S. 10,000m) national record in Rio on the track (30:13, 2016 Olympics). She ran 2:28 at the New York City Marathon, in flats.

“The supershoes came out and she never felt that she responded well.” For every Kipchoge there is a Huddle.

“There’s not going to be a fair (shoe), no one product that affects everyone the same,” says Soh. “Everyone’s biomechanics are different. Unfortunately, this is a part of the sport. If you really wanted to be fair, everyone should be running naked in bare feet. You don’t have any expensive clothes or shoes, but that’s obviously not where we’re at”

Soh Rui Yong points out that no one product affects every runner in the same way (Yong Teck Lim/Getty Images)

Soh points out that birthplace, genetics, finances and role models all significantly impact performance and could be classified as unfair. Track athletes in the 1960s benefited from the introduction of artificial tracks. Sports and technology evolve at unpredictable times and non-linear rates.

For instance, Soh was signed with Asics between 2016 and 2020. “Asics launched a supershoe in 2021, so I never actually got to race in that. My previous national records were in small Asics (racing) flats.” Then he “went to Under Armor (in 2021), who at the time didn’t have a supershoe”.

Brands have caught up to Nike, which explains the second (and bigger) wave of more records and faster times.

It is also essential given the exclusivity of professional running contracts, which prevent athletes from racing in competitor shoes, despite prototypes taking years to reach a final form. Shoes must be on the market for a month before they are raced in under World Athletics rules.

Levelling of the playing field should alleviate a concern shared by Soh and Hutchings. Like Formula One, imbalances of resources and opportunity could create a marathon version of the constructors’ championship — a race winner might not be the best runner but the runner in the best shoes.

World Athletics is making adjustments to track spikes, limiting stacking height (the amount of material between the foot and the floor) from 40mm to 20mm, but race-legal marathon shoes still have the same rules: 40mm stack height limit and one carbon plate.

The critique is that they made the rules according to the first Nike Vaporfly, when it should be the other way around. That shoe had a 39mm stack height and one plate.

Hutchings is prepared to open a can of worms about where fairness lines should be drawn: “Shoe doping, it sounds a bit harsh because it puts a wrong slant — nobody’s cheating. But it is. It’s a similar effect to being on EPO (erythropoietin, the red-blood-cell-creating hormone that the body naturally produces but can be injected artificially to illegally enhance performance). It’s giving people a massive advantage.”

Trying to predict exactly where marathon runners will be in five years is futile. A few things are almost certain: Kipchoge won’t be on the circuit, researchers will have more answers, and the times — from new heroes — will be faster.

Who knows, though, what their shoes will look like?

(Top photo: The record-breaking Nike ZoomX Vaporfly at the “Beazley Designs Of The Year 2020” photocall, by Mike Marsland/WireImage)