Updated at 11:05pm ET on 27 April 2026
To understand the importance of someone running a marathon in less than two hours, you also need to understand that, until recently, the concept of this is actually happening, completely meaningless. Indeed, a physiologist named Michael Joyner had suggested that such a thing could happen in humans. magazine paper back in 1991. But his colleagues laughed at the idea, and it hasn’t changed much in the decades since. In Athlete’s World in 2014, I predicted that it would happen in 2075. To be honest, even that prediction seemed overly optimistic to me, but I thought I would be dead by then, so no one could call me.
Well, I was wrong. Yesterday morning, the two-hour marathon barrier finally subsided. A 31-year-old Kenyan named Sabastian Sawe who has never been announced won the London Marathon with a time of 1:59:30. That is, for reference, 26.2 miles run at an average of 4:34 miles—or, to put it another way, a pace that most recreational runners would struggle to sustain for more than a few seconds, if they could hit it at all. Perhaps most surprising is that the runner-up, Yomif Kejelcha of Ethiopia, too ran under two hours, finishing just 11 seconds behind Sawe.
The performance was the culmination of a shift—or, perhaps more appropriately, a complete disruption—in marathon racing over the past few years, where finally breaking the mythical two-hour mark went from impossible to guaranteed. When sports are young, they develop to a great extent and limits. The first marathon over the current distance of 26 miles, 385 yards, was contested at the 1908 London Olympics, won in 2:55:19. Development over the following decades was rapid, but by 1991 the sport was mature, professional, and lucrative. When Joyner made his prediction, the world record was 2:06:50 and they had advanced by less than two minutes since the 1960s. Logic dictated that the coming decades would see even slower progress, as runners approached insurmountable limits in things like how much training they could handle and how much fat their muscles could store.
A change came in 2016, when Nike announced its Breaking2 project. The famous Kenyan athlete Eliud Kipchoge and two others were chosen as the main pillars of the multi-million dollar attempt to create every detail of the two-hour race: nutrition, water, training, shoes, weather, drafting, speed, and so on. At the Formula 1 track in Monza, Italy, in May 2017, Kipchoge ended up running 2:00:25, surprisingly and unexpectedly close to the barrier. He ran almost the entire race behind six arrows that held the wind for him; the pacers alternated in and out of the race, deliberately breaking the rule that all competitors must start at the same time, which meant it did not count as a world record. But at that time, the conversation changed from when for if.
What remained unknown after Breaking2 was how Kipchoge had run very fast. Was he just the genius of a generation? It was a preparation, which aerodynamics experts argued could shave several minutes off his time alone? Or was it the shoes? Nike had launched an all-new Breaking2 design, featuring a curved carbon fiber plate in a thick midsole spring foam wedge, which external lab data suggested would make runners several percent faster. Two years later, when Kipchoge ran 1:59:41 under similar conditions and not eligible for the record in Ineos’s 1:59 Challenge in Vienna, those questions still persisted. But it was clear that the shoes really worked. National and international records for every distance were falling, and every major shoe company had come up with its own version of Nike’s plate-and-foam upper.
Now that everyone has big shoes, you might think the playing field is level. In fact, the innovation arms race is on. The full functionality of plate-and-foam architecture is still not fully understood, so shoe companies continue to tinker and develop better shoes. For yesterday’s record-setting marathon, Adidas unveiled a new shoe with an ultralight foam midsole that reduced the overall weight of the shoe to just 3.4 ounces. Sawe was wearing a shoe; four of the top five men’s winners, including Sawe, are sponsored by Adidas.
It’s hard to overstate how incredible this is for the running world, which used to pride itself on being a simple, gear-agnostic game. Every year since ancient times, shoe companies have launched new shoes with the promise that they will be game changers. Until 2017, this was never true. But now the record books are being rewritten. Kipchoge lowered the official record to 2:01:39 in 2018, then 2:01:09 in 2022. The following year, another Kenyan, Kelvin Kiptum, ran 2:00:35 in the Chicago Marathon. This was proof that Kipchoge wasn’t an irreplaceable freak of nature—and it invited more questions about shoes and the meaning of comparing runners year after year. Could Sawe break two hours in different shoes? Could he even do it in last year’s shoes? The head-to-head comparison is difficult: Kipchoge, now 41, is past his competitive prime, and Kiptum was killed in a car accident at age 24, just months after setting his world record.
Then there is the question of drugs. If sprinters on steroids were the epitome of the 1980s, endurance athletes on blood transfusions have become an equally popular breed. Kenya, in particular, has been cited as a serial offender: More than 140 runners from the country He is currently serving a doping suspensionincluding the women’s marathon world record holder, Ruth Chepngetich, who tested positive for drugs in the summer of 2025. (Kipchoge and Kiptum have not faced any official doping charges.) In this regard, Sawe and Adidas have been aware. In the two months leading up to last season’s Berlin Marathon, Adidas rode a information $50,000 to test the World Athletics Integrity Unit in Sawe 25 times. Berlin turned out to be too hot in a hurry, but Adidas and Sawe continued the plan this year. “I wanted people to know that whatever happened in the race, I was not to be doubted,” Sawe he told it the website that runs LetsRun.
Sawe’s extraordinary performance justified the extraordinary precautions. In London, a group of six runners broke away early, trailing three pacemakers by the halfway mark, clocked in at 1:00:29—which, you’ll remember, is the slowest of the two-hour course. Sawe seemed to be conscious, conserving his energy, his eyes locked on the back of the pacemaker in front of him. One of the pacemakers continued until after the 25km mark, by which time the pack had been reduced to three. As soon as the last microphone came out, Sawe came to life and started turning the screws.
If the magic of Kipchoge’s unofficial sub-two-hour race was in the drafting, then Sawe having to lead for more than 10 miles would have done him a disservice. Instead, he got faster. It was only in the last few miles that the BBC race commentators suddenly realized that history could be interesting. You can’t blame them: No one could have foreseen how much Sawe would accelerate. He ran the second half in 59:01—a time that, by itself, would be a national record in all but a few countries. And clinging to Sawe’s shoulder until the last mile was Kejelcha, an Ethiopian runner, waiting for him to jump. Sometimes the best runners prefer to slow down the competition when chasing world records so they don’t have to worry about being passed if they misjudge the pace. But in this case, it seems that Sawe’s speed was motivated at least in part by a desperate desire to get rid of his lingering shadow.
All told, Sawe’s accomplishments—the face-off, the drug-testing program, the dramatic finish—were exactly how you’d write a full-time performance. He did everything right—which is why I feel bad about the lingering anticlimax I feel in myself and the feelings in my running friends. The truth is, Sawe’s performance was the second surprising result of the weekend marathon. In the marathon in Toledo, Ohio, an unknown 25-year-old local named Vincent Mauri won in 2:05:55, beating the previous course record by more than 13 minutes. This makes him the fourth fastest American in history. These are both, in their own way, performances for the ages, unless next year’s shoes turn out to be even better.
This story misstated Sawe’s time in the half of his record-setting race as 1:00:26.
