The Latest Research on Super Shoes and Fast Marathon Times
World Athletics looked into eight years of performance improvements, and some crazy statisticians from Cornell reviewed thousands of web photos — and they have numbers.
Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
Two new research articles on Nike Vaporfly and similar super shoes indicate that interest remains as high among sports scientists, runners and running fans as when I wrote a New York Times article on the topic in October, 2019 and it was briefly the most-accessed story on the entire NYT website. Who woulda thunk it?
The two new papers reinforce prior research indicating that super shoes have improved marathon performances by 1 to 3 minutes. In a strange and rare coincidence, both papers come up with the exact same figure for the effect on elite women’s marathoning: 2 minutes and 10 seconds.
The new papers, one from World Athletics and one from a little-known but obsessive statistics group at Cornell University, differ slightly in other ways. One finds that women have gained more than men; the other reaches the opposite conclusion.
At the end of 2016, the men’s and women’s marathon world records stood at 2:03:13 and 2:15:25. Three years later, they had dropped to 2:01:39 and 2:14:04. Most running experts agree that shoes made the difference.
The World Athletics Report
This paper, published two weeks ago by Frontiers in Sports and Active Living, is available in free, full text from the linked website. The first author, Stephane Bermon, is head of the Health and Science department at World Athletics.
The WA research focused on the arc of global performances at three key distances (10,000 meters, half-marathon, and marathon) by the world’s fastest runners from 2012 through 2019. It concluded that Advanced Footwear Technology, which is essentially code for Nike Vaporflys and their kin from other brands with high rebound foam and rigid, curved plates, has significantly improved half-marathon and marathon race times, with the biggest impacts being for female runners and the marathon distance.
In the marathon, women improved by an average of 2:10 and men by 1:03. In the half marathon, the equivalent time improvements were 1:01 and 0:15. The improvements were equal among non-African and African runners. The latter group made up 87% of top performances for men in the three distances, and 72% for women.
World Athletics was able to identify “small numbers of athletes” (about a half dozen men and women) who ran the same marathon course pre-super shoe and then a year or two later in super shoes. Among these individual runners, the women improved by 2:01, and the men by 1:42.
Average of 20 Best World Marathon Times, by year:
After 2012, marathon times slowed until the super shoes were widely available in 2018. Adapted from World Athletics.
The Cornell Report
The Cornell paper analyzed results from 22 North American marathons from 2015 through 2019. Its runners, with bests of 2:24 or faster (men) or 2:45 or faster (women), are much slower than the top-20-global marathoners in the WA paper. Overall, the Cornell researchers found that male times dropped by an average of 2:57 and female times by 2:10.
The Cornell results have actually been available for the last 15 months, though the paper has gone virtually unnoticed in open view. This happened because the authors chose to skip the traditional-but-often-lengthy process of peer-reviewed journal publication. Their paper is easily and freely available here at a site named arXiv, and all their files can be found on github.
“We wanted an outlet that anyone could access easily,” says Cornell statistics professor and 1:10 half marathoner, Joe Guinness. “We didn’t want to wait for a stamp of approval. People can trust our data because we put it out there for everyone, and if someone finds a mistake, we’ll make a correction.”
The Cornell process was the most interesting part of the research. Guinness and four grad students didn’t just compile easy-to-find marathon stats. In addition, they searched tens of thousands of web photos to identify both athletes and the shoes they wore.
Impressively and eye-wateringly, they were able to identify the runners’ shoes in over 95% of 1690 total performances. “The photo searching was way more difficult than the math modeling part of the paper,” Guinness says. “It took five of us most of a semester to complete.”
The payoff? The Cornell paper tells us much more about individual improvements in super shoe runners than World Athletics can. Among men, 74.5% of marathoners who switched to super shoes raced faster marathons. Among women, 71.4%. They improved, on average, by 2:06 (men) and 1:48 (women).
Marathon Improvement, Men and Women, with super shoes: