Skip to main content
ABC News
Can Science Help Runners Break The Marathon’s 2-Hour Barrier?

When Kenyan Dennis Kimetto set the world record at the Berlin Marathon in 2014, his time, 2:02:57, made him the first runner to complete a marathon in less than two hours, three minutes. His time was 26 seconds faster than the previous record, set by fellow Kenyan Wilson Kipsang at the previous year’s Berlin Marathon. Such has the marathon world record progressed over the past 20 years: in increments measured in tens of seconds.

But now three teams — one sponsored by Adidas, another backed by Nike, and one called SUB2 that’s being led by a team of academic researchers — are aiming to push the record nearly three minutes faster. Their audacious goal: to break the two-hour marathon mark.

When Roger Bannister broke the four-minute mile in 1954, he did so as an amateur athlete, often training during his lunch breaks while attending medical school. By contrast, these two-hour marathon attempts are being made by professional runners handpicked by teams of researchers and bankrolled, in two cases, by corporations eager to show that their products can turn good runners into makers of history.

We gathered a few of our favorite running science geeks to discuss the two-hour record, the current attempts to break it and whether these projects are good for the sport. The transcript has been lightly edited.

Our participants:

Christie Aschwanden is lead science writer at FiveThirtyEight.

Wouter Hoogkamer studies the biomechanics, energetics and neurophysiology of running and other sports at the University of Colorado.

Alex Hutchinson covers the science of endurance sport for Runner’s World and other publications.

Steve Magness is coach to professional and collegiate runners and co-author of the book “Peak Performance.”


christie: I want to start by asking (on behalf of our readers who aren’t marathoners): What is the magnitude of the goal here? Is breaking two hours a gargantuan feat? Or is this a technicality that will happen soon regardless?

alex: Under “normal” circumstances, it’s very big. A few years ago, I predicted it would happen in 2075, which gives a sense of where my thinking was at that point.

wouter: Without targeting some external factors to make it easier, it will be a long time before it happens.

steve: Agreed. It’s a huge task. We’re looking at a several-decades jump in performance, at the minimum.

christie: Is there some equivalent or analogy here that might help nonrunners understand the scope? How much faster is this record-breaker going to have to run?

wouter: 2.5 percent faster.

alex: For comparison, Usain Bolt has improved the 100-meter record by 1.6 percent, if I recall correctly.

christie: Wow, that makes this look like a pipe dream.

alex: And Paula Radcliffe improved the women’s marathon record by about 2.5 percent.

wouter: In two increments, though, Alex.1

steve: I think the important thing is it’s a long, long way from our current reality.

christie: So what makes the folks behind these three projects think they can do it?

alex: People have been aware for a few years that there’s some “low-hanging fruit,” like optimizing the course and pacing, that would bring the goal much closer. It’s been a question of when someone would believe it’s close enough to invest money to make those things happen.

wouter: Exactly.

steve: Personally, I think it’s scientific arrogance and naivete that makes them think they can get it done within months/years. Can they drop the world record? Of course. But to get sub two hours is another question.

alex: In rough terms, I always figured the “easy” stuff could get us halfway there. For Nike, I suspect it was their development of a new shoe that made them believe they could bridge the other half. Not sure about the other two.

christie: Does science have a track record for facilitating faster running times? How much of the previous records are attributable to scientific advances?

steve: As a coach and exercise scientist, I’d say the advancements due to science are minimal, at best. That seems like sacrilege, but if we look at the drops in time, they aren’t athletes training utilizing scientific gadgets. They are East Africans training with coaches doing standard training. The one advancement that has contributed in the marathon is refining of fuel intake.

alex: I think running improvements are less influenced by science than almost any other sport. But there are still effects: equipment, track surfaces, etc.

steve: When we look at running, science only adds the very small finishing touches. We’re talking fractions of a percent here and there.

wouter: Outside running, there are multiple examples of where technology improvements have substantially improved sports performance: pole vault pole, clapskates for speed skating. Kimetto’s world marathon record was set with a shoe with a midsole material that has been shown to save 1 percent of energy.

steve: But does 1 percent energy saving result in 1 percent improvement? I’d highly doubt so. Yet that’s what people immediately jump to. They think, “Oh, I’m 1 percent more efficient, so I’m 1 percent faster.” But it doesn’t work like that.

alex: I think Wouter might disagree that changes in efficiency don’t translate to changes in race speed!

wouter: Correct. We showed that adding 100 grams to shoes costs about 0.8 percent more energy and makes you run about 0.8 percent slower.

alex: Of course, whether that holds true over 26.2 miles is a very big question!

wouter: The relationship might not be perfectly linear at high speeds due to the effect of air resistance.

steve: The marathon is a different ball game. In the shorter events, the physiology is mainly the limiting factor. When we get to the marathon, the causes of fatigue multiply.

christie: So what are the limiting factors in the marathon?

steve: Factors that all could be the weak link in the chain depend on the athlete and the race, things like fuel utilization, muscle damage or cramping, mental fatigue or psychological coping, and on and on.

Basically, in the marathon, there are a lot more pipes that can burst than, say, in a mile or a 5K.

alex: It’s important to note that Nike’s project, at least, isn’t really trying to change any of these traditional limiting factors. (I should add that Nike would probably disagree with that characterization, but it’s my take.) They’re instead trying to optimize some of the well-understood limiting factors like air resistance, as well as course details like the number of hills and turns.

christie: Alex, you’ve reported on the different projects. What distinguishes them?

alex: I know basically nothing about the Adidas project other than that they have a pair of shoes. The Nike and SUB2 projects are somewhat similar in the tactics they’re trying, with the difference that Nike has a lot more money and is staging its own event instead of using an existing race.

wouter: I think that is the most important part of their approach.

alex: Even more important than the shoes? 🙂

wouter: Current races are not optimized to run as fast as possible, coursewise.

christie: So what makes for fast conditions?

steve: Flat course, perfect racing temperature (45 degrees or so), fewer turns. If you want to truly optimize, you’d have a slightly downhill finish.

alex: Wouter’s dream course, outlined in his recent paper, was a flat, sheltered loop for the first half, followed by a gradual downhill (just within the 1 meter-in-1,000 rules for record-eligible courses) for the second half. I pretty much agree, except I might save the downhill portion for the last six miles, when things really get tough.

christie: Track and field’s governing body, IAAF, has rules for what makes a legal course: a maximum overall drop in elevation of no more than 42 meters and a start and finish that are no more than 13.1 miles apart, as the crow flies, to prevent aid from a tailwind.

wouter: We say you need to look for a course that drops exactly 42 meters. Same for the wind, if we can go 13.1 miles in one way, let’s do it and make sure you have a tailwind there and limit the negative effects of wind during the first half.

christie: You also mention drafting.

wouter: Yes, that’s the final strategy. The problem is you would need four guys who run a 2:03 marathon all running the same race and collaborating, and that won’t happen without serious monetary incentives.

christie: How many runners are currently capable of that kind of time?

steve: If they are all on their best at the same time (which never happens), four.

wouter: Right, it’s not very likely. But if you could bring two of them together, and apply the downhill, tailwind and shoe technology, they’d have a fair shot.

christie: Why is it so hard to run a perfect marathon?

steve: There’s a reason that at most major marathons, with all the best guys trying to run fast, you are normally left with one, maybe two survivors — despite almost a dozen of them on the starting line who are able to run in the 2:04 range. We forget that these are humans, not machines. The training it takes to even get in 2:03-04 fitness is crazy. Marathoners live near the edge. They do as much as possible without falling over the cliff of overtraining or injury.

alex: Steve’s point is important. If you compare the start list to the finish results at major marathons, it’s like, “What happened to all those fast guys?”

That’s one reason I was pretty surprised that Nike went with a team of just three runners. Even getting to the start line of a marathon is a low-odds game. On these points, it comes down to “will it happen on any given day?” rather than “can it happen in theory?”

christie: So the proposed approaches include a fast course, drafting and fast shoes. The SUB2 team also has a newfangled sports drink. How likely are any of these to make the difference?

alex: I’d say a fast course, drafting and shoes are the big ones. There are lots of other things people are doing, but they’re not make or break. Nike’s not doing much new with sports drinks — just trying to make sure the athletes execute best practices, which not everyone does even at the elite level. The SUB2 project has an interesting new drink that makes some bold claims but hasn’t published any data to back it up yet.

steve: And we haven’t even touched on the biggest limiting factor: the psychology of it all …

christie: How does psychology play in, Steve?

steve: You’re almost three minutes from the record. These guys are going through the half-marathon at near their half personal record. People think that “elites” are invincible mentally. But they freak out, they panic if they are too fast or slow, even if they don’t show it. The way fatigue works is a comparison to our prior experience and our current context. If your prior experience is nowhere near what you are trying to do, your body’s default setting is to freak out.

alex: There’s that famous Herb Elliott quote about how, to set a world record, you have to have the arrogance to believe that you can be better than anyone else in history and then the humility to actually do it. Everyone who does it starts with an arrogance that is basically irrational, and most people will never do something that justifies that arrogance — but without it, you don’t even come close. Anyway, Nike (and SUB2, etc.) clearly have the arrogance, but the jury is still out about the humility.

christie: I want to switch gears for a moment. Last week, we learned that the women’s gold medalist in the Olympic marathon in Rio, Jemima Sumgong from Kenya, has tested positive for the blood booster EPO. At this year’s Sloan Sports Analytics conference, Peter Weyand, a researcher on the SUB2 team, argued that their project could show athletes that there’s a way to excel with science that doesn’t send the sport down what he called the “pharmaceutical gene-doping freak show.” What do you all think? Could he be right, or is this just naive optimism?

steve: I think it’s naive optimism, and Peter was my adviser during undergrad, so I can say that! The money and fame in running sub two are enormous. We still use Roger Bannister as an example of a barrier breaker nearly 70 years later! With that much to gain, athletes, coaches, agents, even sponsors will take risks. Athletes are already taking risks for much lower payoffs. If the payoff is that large, you bet it will bring doping into it.

wouter: I don’t think the sub-two-hour quest will encourage doping more than gold medals and “regular” prize money.

alex: I’ve been totally baffled by that message from the SUB2 team. Everyone wants to run fast and beat people, so it’s not clear to me why one sub-two project would encourage more doping and the other would encourage less!

steve: We’re talking about all these measures that may improve a half percent here or there. EPO can improve performance by up to 6 percent, according to some research!

christie: This leads me to another question: How important is it, really, to reach this arbitrary number? The marathon distance is an artificial construct, and the two-hour mark is just a function of our love of round numbers, right? I don’t recall anyone getting excited about breaking the 2:03 mark.

alex: Well, I was pretty excited.

christie: Ha!

steve: Ha, so was I. But it’s a round number, like a four-minute mile or a 10-second 100 meter. So, yes, it will capture our imagination.

christie: If it were the women’s record that stood this close to the two-hour mark, would there be as much interest?

alex: You know the answer to that, unfortunately, Christie.

christie: 😞

alex: That said, that’s one area where this sport isn’t too bad. Think of the biggest marathon stars in the American firmament — Deena Kastor, Shalane Flanagan, Kara Goucher, etc. There’s definitely respect for female accomplishments.

christie: Most of the major marathons have equity in prize money, right?

alex: Yes, as far as I know.

steve: Exactly. Compared to most other sports, running is much more progressive. As someone who coaches mostly professional women marathoners, I wish it was more even in popularity. But you are seeing more split coverage and starting times so that the women’s race gets equal coverage. Which is huge.

wouter: There are also scientific papers on the women’s equivalent of the two-hour marathon.

christie: That research says the women’s equivalent of the two-hour marathon “has already been achieved.” 🏆

steve: That’s because Radcliffe’s 2:15 is such a large outlier.

christie: Right. It has stood for 14 years!

alex: Put those worms back in that can!

steve: I didn’t say why! I just said it was far better than anything else we’ve seen!

christie: I want to finish by asking: Is the pursuit of the sub-two-hour marathon good for the sport?

steve: I think it’s bad for the sport. We’re getting further and further away from what makes sport interesting: competition. Our obsessive drive for faster has hurt track and field as it is. We set ourselves up for failure by hyping up world records and then being disappointed when they do not occur. If we ever want to have this sport gain popularity, we need to take it back to its roots. Draw a line on the street and race to the next light pole. That’s the essence of running. These gimmicky approaches using artificial environments just push us further away from competition. Yeah, it’s great to see where limits lie, but I think this will push us more toward doping and a focus on times.

alex: I should start by saying that I recognize many of the concerns that people have. I, too, love good, old-fashioned competition for the sake of head-to-head racing, instead of commercialized mega-events. But even with those caveats, my general sense is that it’s a net positive. I can’t overstate the number of people I’ve heard from whose messages start with some version of “I don’t usually follow running, but I saw your article on the sub-two thing and wanted to ask. …”

wouter: Well, I think it’s definitely good for science. Everything we learn from what’s limiting human (sports) performance will have the potential to be used somewhere down the road in making walking easier for specific patient populations, which has many known benefits.

alex: Ultimately, the sport is driven by what people want to see. Many people within the sport say “the focus on times is bad.” And many others watch the Olympics (where, e.g., the men’s 1,500 was the slowest time since 1932) and say “the focus on competition rather than time is bad.” To me, those are both interesting aspects of the sport, at opposite poles. I don’t want all one or the other, but neither do I want to get rid of one aspect entirely.

christie: Yeah, although I lean toward Steve’s point of view here, the fact that it’s garnering more attention for the sport seems like a good thing.

Wouter, you’re running the Boston Marathon on Monday. Do you have a time goal?

wouter: I aim for sub-2:40.

alex: Good luck, Wouter. Hope you’ve got some cooperative drafting lined up!

wouter: Using the downhill and cooperative drafting. Regular shoes, though.

christie: Best of luck! Only 40 more minutes to shave off … 😜

Thanks for being here, everyone.


  1. Radcliffe broke Catherine Ndereba’s record of 2:18:47 with a time of 2:17:18, then broke that with the current record of 2:15:25.

Christie Aschwanden was a lead science writer for FiveThirtyEight. Her book “Good to Go: What the Athlete in All of Us Can Learn from the Strange Science of Recovery” is available here.