Nike Went to Extraordinary Lengths to Run a Marathon in Under 2 Hours. What Was It Really For? (It Came Up 25 Seconds Short)

No question, this was a show of force — but one that was good for the brand and good for science.

At the 30-kilometer mark, Eliud Kipchoge's pace had slipped 1 second behind.

His muscles were starting to loosen, his legs quietly leaking energy, and his mind could already do the arithmetic: he had probably missed his shot at "breaking the 2-hour marathon." Even so, a smile kept flickering across his face.

The 32-year-old Kenyan holds the third-fastest marathon time in history, and since 2013 has won six of the seven biggest marathons in the sport. Kipchoge has a habit of smiling at the hardest moments of a race. He says it is because running is a happy thing.

Through a screen it is hard to feel what it means to hold 5.83 meters per second for two hours straight — roughly your all-out sprint. Still, people watching the livestream on Twitter, Facebook and Sina Weibo kept blurting things out: "Is that a smile or a muscle spasm?" "I swear he smiled. Unbelievable!" "He looks like he's out for a walk in the park!"

Nike's "Breaking2" marathon was nearing its end.

This was not the kind of city marathon people are used to, where a gun goes off and a tide of bodies starts moving. Only 3 athletes ran: Rio Olympic men's marathon champion Kipchoge, half-marathon world record holder Lelisa Desisa, and three-time Boston Marathon winner Zersenay Tadese. Their target was the 2-hour barrier.

The world record stands at 2 hours 2 minutes 57 seconds, set by Kenya's Dennis Kimetto at the 2014 Berlin Marathon. Cutting those three minutes means cutting about 4.2 seconds from every kilometer — the equivalent of finishing while Kimetto still has a kilometer to run. The last time the marathon record fell by three minutes, it took roughly 19 years.

At 5:45 a.m. Italian time on May 6 (11:45 a.m. in Beijing), the race began. Fifteen minutes later the sky was only starting to lighten.

Most marathons start at 9 a.m. or later. Breaking2 went off more than three hours earlier, which cuts against the usual theory that marathons are run more efficiently in the evening. But "in our plan, the environment mattered more," Dr. Philip Skiba, who ran the athletes' physiological health and training program, told QDaily. The early start was the only way to get the temperature and humidity they wanted.

Wind speed 0.0–0.4 kilometers per second, temperature 11.3 degrees Celsius, humidity 79%. In Philip's words, "very good for running."

Alongside the careful planning, poetic and grandiose language turned up in every corner of the event. The livestream cut now and then to footage Nike had prepared: Roger Bannister going after the 4-minute mile on May 6 sixty-three years earlier, with the word "Moonshot" on screen. In a pre-race interview with QDaily, Nike chairman and CEO Mark Parker called breaking 2 hours "one of many holy grail challenges."

The idea did not originate with Nike. In 1991, Michael Joyner, a researcher at the Mayo Clinic, predicted that a human could run a marathon in 1 hour 57 minutes 58 seconds. It became genuinely tempting in 2008, when Ethiopia's Haile Gebrselassie became the first man under 2 hours 4 minutes. In 2011 the Journal of Applied Physiology published a paper titled The Two-Hour Marathon: Who and When? In 2014, shortly after Kimetto set his record, Yannis Pitsiladis, professor of sport and exercise science at the University of Brighton, announced his Sub2Hur project and said he would get there before 2019.

Crowdfunding site, recruitment drive. As the first person to make a serious public run at the limit, Pitsiladis needed $30 million. He tried to win sponsorship from Nike and others and failed. Nike, it turned out, wanted to do it alone.

In December 2016 Nike announced Breaking2. Two months later, word got out that Adidas had a two-hour marathon project of its own, called "Sub2," and had been working on it "for years."

On execution speed, Nike was for the moment winning this moonshot race.

The night before the race, Andy Jones, the applied physiology professor Nike had brought in as an outside consultant, wrote his prediction on a scrap of paper: 2:00:59. The actual result, 2:00:25, was better than that, and better than most sports scientists had forecast.

The only problem is that the time will not replace the marathon world record. The IAAF, which writes the rules of running, considers what Nike did for Kipchoge cheating.

Like Tuning a Race Car

To guarantee a satisfying time on the morning of May 6, Nike controlled almost everything it could.

Start with the venue: the Monza Formula One circuit in Italy. The "Temple of Speed," northeast of Milan, holds 115,000 spectators and has been roaring with engines since 1922. The notorious "death corner," the Parabolica, was the start line.

A track that has killed nearly a hundred people now made an excellent marathon course. The loop has almost no corners, so no energy is wasted turning, and the marked 2.4-kilometer lap gave precise control over pacing, hydration, nutrition and the movement of the support crews.

"What we're doing is unusual: we look for a window of time in which we can pick the single day when humidity, temperature and every other condition is right for the attempt." After analyzing six years of weather records and comparing them with the conditions at the fastest marathons ever run, Monza won: temperatures around 12°C, mostly cloudy skies (which lower a runner's heat load), and steady air thanks to the track's distance from the coast being just right. More important, a circuit far from a city center is far easier to control.

"The ideal course is flat. So far the fastest marathon records have come in Berlin, which is considered one of the flattest courses there is," said Brett Kirby, lead physiologist at the Nike Sport Research Lab. In January the team settled on Monza, "flatter than Berlin," with every environmental factor they needed. By some estimates the circuit and the conditions it delivered were worth 20 seconds.

Berlin's Brandenburg Gate, Boston's Heartbreak Hill, the Thames in London, New York's Verrazzano Bridge, the wide sheet of Lake Michigan — in Andy Jones's words, "no major city marathon course is perfect."

The strange course was not the main reason the IAAF refused to recognize the race. Harder for the federation to swallow were the 30 elite pacemakers wrapped around the 3 runners, and the research teams piled up around them like the crew around an F1 star.

Kipchoge, Desisa and Tadese, meanwhile, were the supercars, picked for the biggest engines (VO2 max) and the best fuel economy (running economy).

Anyone who has run along a seafront or a river will tell you a kilometer costs several seconds more into a headwind than with one. Since the 1970s, marathoners have used drafting: sit on someone else's rhythm early, where the air is easier, and save the legs for the end.

The 30 pacemakers, among them world champion and American middle-distance star Bernard Lagat, would run in groups of 6 as a windbreak, each man covering 2 laps, with 3 of them swapped out each lap.

At the press conference the day before, Nike's Breaking2 scientists unveiled the pacing plan with visible pleasure, and when the animation on the screen showed the pacers snapping into an arrow formation, the room full of reporters drew breath.

"We had plenty of other pacing strategies. Let your imagination run wild," said Matthew Nurse, vice president of the Nike Sport Research Lab.

A Tesla Model S would also drive directly ahead of the runners for the whole race, showing current pace, elapsed time and projected finish, and firing lasers at the ground behind it to mark the formation.

"To get 85% of the drafting benefit you have to sit exactly in the pocket," explained Alex Hutchinson, a columnist for Runner's World and one of the race commentators. "That almost never happens in a normal race. Until last Friday."

Fueling was the other priority. Nike set up a station in the middle of the circuit, where at 2 a.m. the scientists laid out a full row of bottles for each runner, mixed to different ratios of carbohydrate, caffeine and water, some liquid, some the texture of jelly. Every lap a courier on a motorbike handed one over — energy in, and less of the stomach trouble that hard running brings.

"2,300 calories." After the race Andy Jones pulled out his phone and worked out what the runners had burned. "We gave them about... 15 grams of carbohydrate an hour, so 500 calories in all. They can make up the rest with champagne!"

Environmental scientists, physicians, biomechanists, designers, R&D directors. According to Mark Parker, working across that many disciplines at once was new for Nike. The roughly 400 people involved, outside consultants included, had one job: work out what breaking 2 hours requires, and optimize every variable that might be worth a few seconds.

"We've already taken 6 seconds off the mile. To get from 2:00:25 to 1:59:59 we just need to find the next second. There has to be something more we can do," Andy said.

Nike will not say what it paid the three runners. What is known is that all three skipped the recent London and Berlin marathons to train for Breaking2, in Kenya, Eritrea and Ethiopia respectively. Thin air at 5,000 to 8,000 feet raises the red blood cell count, so the blood carries more oxygen to the muscles, and the elevated count lasts up to two weeks after coming down.

"In the end we neither pushed nor pulled these athletes. It was all human power," said Matthew Nurse, who holds doctorates in biomechanics and medicine. "Talk about the science all you like — they still have to run."

"It's Gotta Be the Shoes!"

None of it made two hours easy. It would take 3% off the world record, and no track record has ever fallen that fast — Paula Radcliffe's two women's marathon records improved the mark by 1.4%, and Usain Bolt took less than 1.6% off the 100 meters.

For a company that sells running shoes, the answer was obvious, and it was the old Jordan slogan: "It's gonna be the shoes!"

In June 2013 Nike started a secret project inside the company. Sandy Bodecker, vice president of special projects, was its driving force. The 35-year Nike veteran and self-styled problem solver walked into the CEO's office, told Mark Parker what he wanted to do, and the two agreed Breaking2 was a great idea. "My part was saying yes the whole way and being the cheerleader," Parker said.

Around the same time, Dr. Luo Geng, who grew up in China, joined the Nike Sport Research Lab as a biomechanics specialist, charged with improving a shoe's running economy by 3% — one piece of Breaking2.

The first few hundred test pairs were all built on the same principle: lighter is better. That was where the trouble started.

Most marathon shoes chase minimum weight, on the theory that every 100 grams adds 1% to a runner's energy cost. But once the target became a two-hour marathon, with the athlete holding a very fast, very even pace, Luo Geng's team found that a wafer-thin sole could not take Breaking2's 20,000-plus footstrikes, each landing with three times the runner's body weight.

In 2015 they built a test shoe so plain that the team called it the "Cinderella shoe": two layers of very low-density blown foam rubber, which Nike named ZoomX, and a forefoot 21 millimeters thick, unheard of in a marathon shoe. Luo Geng compared the energy the thick sole returns to a big battery.

Between those two layers sat the thing everyone would later argue about: a spoon-shaped carbon fiber plate weighing under 20 grams, designed to cut the work done by the calf.

In March, Nike showed the Zoom Vaporfly 4%, due to go on sale in June at $250. Breaking2 would use a custom version, the Zoom Vaporfly Elite. The main difference is that its carbon plate is tuned to each athlete's foot and stride.

Yannis Pitsiladis, the man behind Sub2Hur, somehow got hold of a pair of Zoom Vaporfly 4%. The New York Times ran his CT scans of the shoe in a piece headlined "Do Nike's New Shoes Give Runners an Unfair Advantage?" In the images the plate looks like something a security scanner flags at an airport.

The article put the plate in the same category as the J-shaped carbon blades worn by the sprinter Oscar Pistorius. "We're getting into a really interesting area of the ethics of sports technology," South African sports physiologist Ross Tucker told the paper.

Tucker has been among the sharpest critics of Breaking2, and he also writes the blog Science of Sport. As soon as the May 6 race ended, he posted a call to ban any carbon-plated equipment from competition.

"Runners in these shoes have won London, Berlin, New York and Chicago. Kipchoge's 2:03:05 in London, and the gold, silver and bronze in the Rio Olympic marathon — all in these shoes," Luo Geng said in his presentation, barely hiding his delight. Before Nike announced the Zoom Vaporfly 4%, the shoes had been disguised as ordinary models, and the athletes and researchers were sworn to secrecy.

Nike put out a press release for the occasion, quoting a run of athletes who had raced in them. They reported less fatigue over long distances, faster recovery, a three-course meal for the calves.

For a runner like Kara Goucher, who finished fourth, the feeling was the reverse: two rivals in Zoom Vaporfly 4% pushed her out of a place at the Rio Olympics.

Carbon plates in running shoes are not new. Adidas used one in a product called Pro Plate at the turn of the millennium, and the man who helped develop it, Dr. Darren Stefanyshyn, was Luo Geng's doctoral advisor.

Nike sent the Zoom Vaporfly 4% to a lab at the University of Colorado for testing, and the "4%" in the name comes from one number in the results: it saves 4% of the energy of Nike's best previous marathon shoe. Rodger Kram, who runs the lab, told Runner's World: "Almost every modern running shoe has a spring in it. We just call it a foam midsole now."

Tony Bignell, Nike's vice president of footwear innovation, answered the criticism this way:

"First, I don't think it's illegal. It's a bit like tennis — if the ball came off the racket with more energy than you hit it with, that would be cheating. All we're doing is minimizing the energy you waste every time your foot lands."

Who Is the Next Kipchoge?

Breaking2 aside, the Zoom Vaporfly is plainly being sold to people whose marathons take more than 6 hours.

At its peak the Sina Weibo livestream drew 1.28 million viewers. Brandwatch counted more than 584,000 mentions of Nike on Twitter, Facebook and Instagram over seven days; the #Breaking2 hashtag was used 400,000 times on Twitter, 87% of it in positive terms. The hashtag's reach passed 2 trillion, which a senior Brandwatch analyst said was the largest he had seen for a single tag.

Even CCTV, invited to Milan, streamed the whole race through its mobile app.

The brand marketing firm Red Peak said in an interview that Breaking2 will stand as one of the outstanding marketing precedents of the next few years: "They got everyone's attention with something novel and hard to pin down, then used social media to push it further."

"A marathon attempt prepared in secret for years is about to be livestreamed at enormous risk." The event was closed to the public, but anyone with a phone and Wi-Fi could watch every minute of it. The risk, in fact, had been minimized: whether all three broke 2, one or two did, or nobody did, Nike had a plan ready for each outcome.

Everything about Breaking2 suggests a project prepared thoroughly and then pushed out the door as fast as possible.

Fashion has reshaped the sportswear business over the past two years. Nike, the master of the beautiful television commercial and the mass-market sports product, kept losing ground on growth to Adidas, which sold staggering numbers of sneakers on the social media pull of trend-setting celebrities.

Steve Prefontaine broke the American record in Nikes in 1972, back when an athlete could be both the face of a brand and its director of public affairs. That era is gone; most people today register a champion far less than a street shot of an influencer. Which explains why, even as one athlete after another won in the new marathon shoe, Nike kept quiet about it. In Tony Bignell's words, Breaking2 came out of "the running brand team, the commercial team, the PR team, the research team and the innovation team all working together."

"A hugely courageous attempt. Congratulations, Eliud Kipchoge." That was Adidas Running, replying under Nike's Weibo post after the race.

The IAAF tweeted: "Great run by Eliud Kipchoge in Monza."

"Given that a sport under a doping cloud is in the middle of a push to wipe out every world record set before 2005, getting excited about a race that won't count as a world record is a bit strange," BBC sports writer Ed Harry wrote afterwards.

The freelance sportswriter Sarah Barker put it this way: if Nike's Breaking2 was meant to show what humans are capable of, it did the opposite — it showed in the plainest terms how far we still are from breaking two hours.

"Even unrecognized, it will be called the fastest marathon ever run," Martin Polley, professor and director of the International Centre for Sports History and Culture at De Montfort University, told The Guardian. "The real test comes next. If Nike's runner can go under 2 at Monza, can he do it again at the Tokyo Olympics?"

Who will be the next Kipchoge? Everyone who watched the race is asking, and one man may be better placed than most to answer: Kipchoge's agent, Jos Hermens.

Grey-haired, 67, a former Nike employee, Hermens represents many of distance running's best-known names and is also an investor in Pitsiladis's Sub2Hur project. Every year he goes to East Africa looking for new local talent. Like many people attached to this project, he likes to talk about genes, red blood cells and the falling curve of the marathon record. He once said, "African runners are not good at drinking during a race. They don't understand how important fluid is, and they don't practice drinking while they run."

And yet, like Kipchoge, plenty of Kenyan and Ethiopian runners seem to come out of nowhere. He broke through internationally at 16, won the 5,000 meters at the 2003 world championships, and went on winning at every distance from 1,500 meters to 42 kilometers. No hydration schedule, no heart rate monitor, no custom shoes, no army of scientists.

"If I'm not happy, I don't run," he says.

Cover image from Nike