Eliud Kipchoge is the first athlete to run a marathon in less than two hours.
Eliud Kipchoge is the first athlete to run a marathon in less than two hours. (Photo: Jed Leicester/AP)
Sweat Science

What the Future of Fast Marathons Looks Like

Physiologists from around the world share their pet theories and crazy ideas on what it will take to break records

Eliud Kipchoge is the first athlete to run a marathon in less than two hours.
Jed Leicester/AP(Photo)

Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.

In the Journal of Applied Physiology last month, four influential researchers broke down the science of marathoning—where it stands, how we got here, and where we’re headed. As they pointed out, we’re at an interesting moment in the sport’s history. Eliud Kipchoge set the current men’s record of 2:01:39 in 2018, and Brigid Kosgei set the women’s mark of 2:14:04 in 2019, continuing a period of unusually rapid progress. Check out the percentage improvement in world records since 1999 in a range of long-distance running events:

(Journal of Applied Physiology)

The scientists—Michael Joyner of the Mayo Clinic, Sandra Hunter of Marquette University, Alejandro Lucia of Universidad Europea de Madrid, and Andrew Jones of the University of Exeter—outline the basic model of marathon physiology, in which performance depends on three key traits: maximal oxygen uptake (VO2 max); sustainable intensity (which is closely linked to lactate threshold); and running economy (a measure of efficiency). Then they discuss which factors may have affected these three traits to enable the recent boom in fast marathon running, including genetics, body shapes, training, drugs, and of course shoes.

It’s interesting stuff—but what really caught my attention was the 17 pages of responses from 35 different groups of other researchers that the journal also published. If you really want to get a sense of the full diversity of what endurance researchers are interested in these days, scanning these responses is a great start. Most of the ideas are at least somewhat familiar, but a few are unexpected. Here’s a sampling of some of the themes that cropped up in these responses:

Synchronized Strides

This is the one that came from farthest out of left field, at least for me. A trio of researchers in Germany led by Laura Hottenrott of Ruhr University (herself a 2:33 marathoner) suggest that matching strides with the runners around you enables you to run more efficiently, burning less energy and thus finishing faster.

Is this real? It’s certainly true that we tend to instinctively sync up with runners around us, and it also appears that various other patterns among cadence, heart rate, and breathing seem to emerge spontaneously. But following the references from Hottenrott’s letter doesn’t reveal any smoking-gun evidence that it significantly improves running economy. Classify it as highly speculative for now—but, for fun, imagine a future where the pacemakers at major marathons are selected based on having similar leg length and cadence to the top contenders, in order to maximize the probability of synching strides.

Brain Oxygen

At least three different responses, from French, Danish, and Brazilian researchers, focused on the role of brain oxygen. Running a fast marathon takes a lot of brain power, particularly in the prefrontal cortex, to integrate information, make decisions, and resist the urge to slow down—and that brain power requires a lot of oxygen. There’s some evidence that oxygen levels in the brain begin to drop during all-out exercise, and intriguingly, it appears that top Kenyan runners are unusually good at maintaining sufficient brain oxygen.

So how do you improve “neural efficiency”? It may be that some runners are able to basically turn their minds off and run on autopilot during races, reducing their brains’ oxygen needs. They may also be some external circumstances that make it easier to keep thinking to a minimum, like reliable pacemakers and a familiar and well-signed course—think, for example, of the set-up for Kipchoge’s sub-two attempts. This, too, is pretty speculative, but what’s neat is that brain-imaging studies are starting to put some of these ideas to the test.

Improving Running Economy

In the main paper by Joyner and his colleagues, they present some evidence that VO2 max and lactate threshold haven’t changed dramatically compared to previous generations of runners. So the big changes seem to be happening to running economy. Indeed, according to a response from University of Colorado physiologist Rodger Kram, measurements of economy in elite runners show an improvement of roughly 26 percent over the past 90 years.

No one is entirely sure how running economy has improved or whether it can improve further, though. One response suggests strength training; another suggests plyometrics; another focuses on mitochondrial efficiency and other microscopic properties of muscle fibers. Maybe it’s biomechanics, or high mileage, or being born Kenyan. Or maybe, several commenters pointedly suggest, it’s all about shoe technology. This is a longstanding debate, but is far from settled.

Measuring Running Economy

If we really want to figure out how to improve running economy, we first have to ensure we’re actually measuring it correctly. A comment from University of British Columbia researcher (and Olympic steeplechaser) Shalaya Kipp highlights some methodological issues like the appropriate speed and slope of the treadmill. The most intriguing point: how stiff is the treadmill deck? Good research-grade treadmills are firm like the roads; typical health-club treadmills, as Dutch researcher Bas Van Hooren recently pointed out, are soft and bouncy. So what happens when a study measures running economy in a shoe with a carefully tuned amount of bounce, like the Vaporfly, using a treadmill with additional bounce? The results are meaningless. This is something to watch out for when you’re reading research about the latest shoes.

Fatigue-Related Changes

Take two video clips of a runner at mile 1 and mile 25 of a marathon. Do you think you’d have any trouble guessing which was which? Fatigue takes a visible toll on you, and it probably takes an invisible toll too. The key point raised by a pair of Slovenian researchers is that any calculations based on VO2 max, lactate threshold, and running economy are based on values measured when the runner is fresh. But all three likely change over the course of a marathon, so understanding the fatigue-induced dynamics of a runner’s physiology becomes crucial. Maybe Kipchoge is so great not because he has amazing running economy, but because he has an amazing ability to sustain his running economy even when fatigued.

A more subtle point is that these variables also change from day to day based on training and recovery status. The lactate threshold you measure a few weeks before your race will not be identical to the lactate threshold you have on the starting line. What factors determine whether the latter value is better or worse than the former?

Marathon Pacing

A comment from a group in Spain led by Jordan Santos-Concejero explores the eternal puzzle of optimal pacing. I’ve written about some of the research from this group before, which reveals an interesting contrast between the pacing of pre-1988 marathon world records, which tended to start quickly and then tail off, and post-1999 records, which tended to start a little more conservatively and then speed up after 25K. The more recent records also tended to have a more even overall pace, with fewer unusually fast or slow 5K segments. That points to a future of hyper-optimized pacing—but does that necessarily mean even pacing, or does the recent trend of speeding up in the second half make more sense? The answer to that question will depend in part on the previous point about how and why things like lactate threshold and running economy change with fatigue.

Optimal Age

Your VO2 max typically starts dropping by about 1 percent per year after the age of 25. Elite athletes can fight this curve for a while, but the clock is ticking. On the other hand, long periods of intense training (perhaps along with other factors like strength training and plyometrics) can gradually increase running economy. Paula Radcliffe, for example, managed to increase her economy by 15 percent between 18 and 29 years of age. So the sweet spot, according to Mayo Clinic researcher Jonathon Senefeld, is dictated by the compromise between those two factors.

There are other factors affecting the age of marathon champions, like the steady increase in the popularity of marathons. As a group from Pepperdine University points out, the resulting boom in prize money has lured more top runners to the distance, and kept them in the sport for longer. It’s pretty hard to draw conclusions from the latest records: Kipchoge was 33.8, while Kosgei was just 25.6. On paper, you’d have to guess that Kosgei is far more likely to have more records up her sleeve—especially since no one knows how long it will be before professional marathoning makes its return. But if there’s one thing I’ve learned from the past few years, it’s to never bet against Kipchoge.

For more Sweat Science, join me on Twitter and Facebook, sign up for the email newsletter, and check out my book Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance.

Lead Photo: Jed Leicester/AP

When you buy something using the retail links in our stories, we may earn a small commission. We do not accept money for editorial gear reviews. Read more about our policy.