Eight years into the supershoes era and, from a position of relative anonymity in the marathon world, Puma completely shocked everyone — including its own team.
“We were all like, ‘holy s***, we have something unbelievable,” says Erin Longin. “The advancements were incredible, so extreme, we were in the lab seeing results that we had never seen before. We knew we really had something special.”
Longin, Puma’s vice president of running and training, is speaking to The Athletic from their offices in Massachusetts. That “something special” was the third iteration of the company’s Fast-R shoe, which launched publicly in April for the Boston and London marathons.
This Sunday, like for those marathon majors, 100 sub-elite athletes have been recruited for the New York marathon. They are targeting sub-three-hour performances, and Puma, as part of what it calls ‘Project3’, gave those athletes Fast-R 3 shoes to train and race in.
“We weren’t forcing anyone to wear it,” Longin says of Boston and London, where Project3 athletes had to sign agreements not to post pictures on social media because the Fast-R 3 was not yet public knowledge.
“It was (a case of), ‘If you love it, and it works for you, you can wear it on race day’. We ended up with such a big presence of runners running in it. That helped with visibility,” Longin adds.
Puma says 69 (just over one-third) of the runners bettered their personal bests. Thirty-eight Project3 athletes improved their PBs by at least three minutes, a target for which Puma paid $3,000 (about £2,250) to each athlete — the sportswear company gave out a six-figure total in prize money.
Those are numbers they hope to increase again in New York this weekend.
Runners pictured crossing the Verrazzano Bridge at the start of the 2019 New York marathon (Emilee Chinn/Getty Images)
Proof that the shoe was industry-leading came quietly at first, when a pre-print of a study by the University of Massachusetts was circulating in the media in April.
UMass — which is partnered with Puma but conducts research independently — tested the Fast-R 3 against the previous iteration (Fast-R 2), plus Nike’s Alphafly 3 and Adidas’ Adios Pro Evo, regarded as their leading marathon shoes.
On average, athletes had running economy improvements of more than three percent in the Fast-R 3, according to the study. Running economy is the amount of oxygen needed to run at a given pace, like a car’s miles per gallon, the gold-standard metric because it robustly captures efficiency.
Puma’s Fast-R returned results about as superior to other current super shoe models as what Nike’s first-ever Vaporfly was versus traditional racing flats in 2016.
“The first time we tested it (in-house), we thought maybe the machine that we use to measure running economy was broken,” says Laura Healey, who is the senior research and sports science manager on Puma’s innovation team.
Wouter Hoogkamer, an assistant professor of Kinesiology at UMass, was on the research team that tested the Fast-R 3. Previously, he worked at the University of Colorado, where he was the lead author on a study testing the first Nike Vaporfly in 2017, which gave it the famed “4%” tagline.
The most striking finding with the Fast-R 3, for Hoogkamer, was not the three percent average improvement in running economy.
“For a range of different people, everybody we tested — on top of all the people that Puma tested in their lab — they all do better,” he says. “Some do a lot better, some do a little bit better. It shows that the shoe is very robust, but it’s also evidence of the magnitude of the effect. That’s why this is the next big thing.”
Responder effects are a real thing with super shoes. Some athletes see a much bigger improvement than others, and there are occasions where certain runners perform worse. This uniform improvement is uncommon — and actually quite unexplainable, says Hoogkamer.
When we spoke in early August, Hoogkamer was undergoing the peer-review process to have the Fast-R 3 paper published in a journal. “One of the reviewers asked: ‘Can you speculate more about what’s going on?’. The answer that we’re writing: we’re looking at cadence (step frequency) and contact time (how long the foot is on the floor) and those don’t change.”
Hoogkamer says that to understand the improvements, analysis needs to go deeper than biomechanics. “If you start looking more closely, how quickly are the joints moving? How quickly are the muscle fibers working? We probably will find the rest of those energy savings there. It’s a subtle under-the-skin phenomenon.”
The foam is one reason the Fast-R 3 returns such good numbers. The next iteration of a shoe tends to be roughly one per cent better than the last, according to an Offenburg University study. Puma’s version three is triple that compared to the Fast-R 2.
The Fast-R 2 was 249g (8.8oz), heavy by super shoe standards — almost double the extremely light Adidas Adios Pro Evo 1, and another 50g more than Nike’s Alphafly 3. Incredibly, the Fast-R 3 is one-third lighter than ‘version two’.
“The journey for this shoe was all about keeping that maximum propulsion, but reducing the weight,” Longin explains. “A big part of that was because we had a step forward in our ‘Nitro foam’ compound — the foam became lighter but even bouncier than before. That helped.”
Companies do not make their own foams but source and buy them. “We’re only limited by the materials that we’re given, and that is always advancing,” Healey says. She describes her job on the research team as “to figure out the best way we can put these new little puzzle pieces together”.
Puma’s engineers used computer software to understand how, using analysis from the Fast-R 2, performance would change with different versions and combinations of foams and plates.
Puma’s Fast-R 3, which improves average running economy by more than three per cent, according to an independent study (Puma)
Healey explains the process: “We used athletes’ data — biomechanical data, plus pressure maps and the forces under their feet while they’re running — so we can model them. We know all of the components and properties of the shoe, and we could really understand how they’re using it.”
They made sure to run the simulations with different footstrikes, to see if and how things changed, and then set out to test “in the wild”, as Healey calls it. Getting athletes to run in the shoe — where they can evaluate it for fit and durability — is better done the old-fashioned way.
Hoogkamer tried on the shoes before he tested them. “They’re actually not great for standing around in,” he laughs. “Once you start running, they have the foam where you need it. This is a shoe for running at a decent pace and not anything else.”
The beauty, though, of modelling was being able to objectively answer questions that engineers ask when designing.
“We could really get down so minuscule,” Healey adds. “Which parts of foam don’t we need? How do we reinforce the plate so that it’s going to be the proper ride and the proper propulsion? We went to that deeper level and could get the answer so quickly.”
This can be seen in the design, as Puma chopped out a load of unnecessary foam, creating what Hoogkamer calls a “decoupling” between the forefoot and the heel, where the plate is exposed.
It saved time and money, with fewer prototypes, and Puma could even bring the planned launch forward from this fall to the spring — for the London and Boston marathons.
“We found that, because the modelling was so good, the result was pretty much a one-to-one from that digitalisation,” Longin says. “We ended up not having to change it.”
That is why Healey, with some excitement, thinks “iterating digitally is going to be a big thing moving forward”.
Digital modelling is not unique to Puma or the Fast-R 3. That technology exists with other companies. Designs are hard to patent and companies tend to shop at the same sellers for their shoe materials. This makes the battle one of ideas.
Healey says there was a lot of in-house analysis on the carbon plate in the first two versions of the Fast-R, because the design had it exposed out the front of the toe box. “We had to make sure that the plate was strong enough, and that we could get it as thin as possible while still being strong.”
Puma is a leading track and field brand — its sponsored athletes won 20 medals at the Paris Olympics last summer.
Karsten Warholm competing for Norway at September’s World Athletics Championships in Japan (Christian Petersen/Getty Images)
That included sweeping the men’s triple jump podium, 100-metre gold from Julien Alfred, silver for Matthew Hudson-Smith over 400m and another by Karsten Warholm in the 400m hurdles. Plus, all four Olympic champions from the men’s and women’s high jump and pole vault finals are Puma athletes.
In Tokyo, at the World Championships this September, Puma was third on the shoe sponsor table (excluding relays), with 25 individual medals and six world champions, behind Nike (42 medals, 17 gold) and Adidas (33 and nine).
At its ‘house’ in Minato, Tokyo, Puma showcased five different ‘concept cars.’
These are road shoes (not used in races) where its engineers have taken design elements to extremes to better understand performance.
The shoe on the second row is inspired by the pole vault spike of 14-time world-record holder Mondo Duplantis. Another, on the top row, was crafted using the principles of the running blades used by Paralympic sprinters.
Puma’s five ‘concept cars’ on display in Tokyo (Liam Tharme/The Athletic)
The Athletic asks Longin what, from its track success, has been applied to the roads.
“The learnings from testing with Karsten (for a bespoke spike) absolutely crossed over, when it comes to plates and some of the geometry,” she answers. “But the learnings have really gone in both directions.
“We developed Nitro foam really thinking about road running and marathons. It is such a superior foam compound that, as we learned to innovate within that space, we also found amazing applications for track and field.”
In 2016, Puma had no representation among the 150 athletes at the United States Olympic marathon trials.
Fast forward to 2024, and Fiona O’Keeffe, racing in Puma’s not-yet-released Deviate Nitro Elite 3, won the women’s marathon in an Olympic trials record of two hours, 22 minutes and 10 seconds.
With Dakotah Lindwurm in third (2:25:31), two-thirds of the U.S. women’s marathon team for Paris were Puma, and it had other athletes in 10th, 18th and 28th.
The Fast-R 3 was still under construction last summer, and, as Longin points out, even if Puma has made its breakout in the marathon over the past 12 months, success never happens overnight.
“You say, ‘Why now?’, but it really isn’t just now. (It’s been happening) for a while,” she says. “It was 2018 and 2019 when we decided that the timing was right to get back into road running, based on some of the growth and momentum we’d had.”
Puma has recruited athletes to run in the Fast-R 3 at the New York marathon (Puma)
On relaunching, Puma ripped up its pre-existing setup and built a lab in Boston. Advice, support and inspiration have come from partnerships with UMass, the University of Loughborough in England, and the Mercedes Formula One team, whose engineers helped with Puma’s spike design.
Reaching out for Hoogkamer to test the Fast-R 3 was, he says, the same premise that Nike wanted in 2016 after promising in-house results with the Vaporfly — external validation.
While the 2010s were all about Nike and Adidas locked in a foot race pursuing world records and a sub-two-hour marathon, Puma has tapped into an undervalued part of the market — the competitive three-hour marathon runner, while creating a shoe that makes it eligible to be a major player.
This is not to discredit Puma’s elite presence, which also includes Rory Linkletter, who ran the second-fastest marathon by a Canadian man in Chicago last month (2:06:49).
“It just shows how far we’ve come in a short time,” Healey concludes.
Puma has started writing the next chapter in the super shoe story, which will play out on the streets of New York on Sunday.
