Sex Differences in Running Are More Complicated Than We Thought

This article originally appeared on Outside

A pretty good rule of thumb is that women, on average, tend to be about 10 to 12 percent slower than men across a wide range of running distances. That used to be an uncontroversial observation. But these days, with debates swirling about the participation of transgender athletes and those with differences of sexual development (DSD) in elite sport, such pronouncements are more fraught. Is that gap an ironclad rule of nature, or a sociocultural artifact?

A new study in the Journal of Applied Physiology, from Emily McClelland and Peter Weyand of Southern Methodist University, digs into the nuances on this topic, and finds evidence that different rules apply in short sprints compared to long endurance events. (Ultramarathons, as usual, are yet another story.) This is a particularly interesting finding given the seemingly odd regulations that currently apply to DSD runners such as Caster Semenya, which affect only events between 400 and one mile--the distances at the intersection between sprint and endurance regimes.

McClelland begins by noting that the performance gap between human males and females is somewhat unusual. Among other running mammals, horses seem to have a sex gap of between about 0.5 and 3 percent, while dogs have an even smaller, or perhaps even non-existent, gap. In humans, by contrast, there's almost no overlap between elite male and female running performances. No women's world record is within 4 percent of the minimum men's qualifying standard for the Olympics.

That's largely because of differences in size and body composition. Elite male track athletes are typically about 6 percent taller and 20 percent heavier than elite female track athletes. For a given mass, men have more muscle while women have about 10 percent more fat. Men also have more oxygen-ferrying hemoglobin in their blood. These factors, the authors conclude, are sufficient to explain the consistent performance differences in endurance events.

Sprinting is different, though. Speed, particularly in the shortest distances and at the beginning of races, is dictated not by aerobic energy but by how much force you can apply to the ground relative to your mass. And in this case, being smaller is actually an advantage, relatively speaking. It's why squirrels can scoot up tree trunks while elephants can barely climb a step (and why, if you got shrunk to the size of a nickel, you'd be able to jump out of the blender). Muscle force depends on the cross-sectional area of your muscles, while body weight depends on your volume. Do the math, and it turns out that smaller people (or animals) can produce more force relative to their overall weight.

This relationship shows up in competitive weightlifting: athletes in lighter weight classes can lift more per kilogram of body weight than their heavier peers. Using data from weightlifting along with the typical sizes of male and female sprinters, McClelland predicts that women should have a 10 percent advantage in the force per unit mass that they're able to apply to the ground.

That's not the whole story, though. The 10 percent advantage would apply if women were simply "mini-men" whose size was reduced with a shrinking ray. In reality, as noted above, women also have a lower percentage of muscle, distributed differently on their bodies. The overall effect of body composition, according to the paper, gives male sprinters a 16 percent advantage. Combine those two factors--10 percent in favor of the women thanks to their size, 16 percent in favor of men thanks to their body composition--and you get a net prediction that male sprinters should be about 6 percent faster than female sprinters. That's a long way from the 10-to-12 percent rule of thumb.

So what does the data say? The researchers looked at the top 40 performers each year between 2003 and 2018 at distances between 60 and 10,000 meters. Going all the way down to 60 meters, unlike many previous studies, gives us a better chance of seeing whether the gap in sprints really is different compared to endurance events. Here's what those male-female differences look like:

For events 800 meters and up, the gap is constant, clustering around an average of 12.4 percent. But below 800 meters, the gap shrinks steadily as the distance gets shorter. By the time you get down to 60 meters, it's down to 8.6 percent, well below the rule of thumb.

And you can take a deeper dive within sprint events. The researchers looked at ten-meter splits from the 100-meter finals at four different editions of the World Championships. After ten meters, on average, the women were just 5.6 percent behind the men. But with each successive split, they fell further behind, and their last ten meters was 14.2 percent slower than the men's final ten meters.

There are a bunch of different things going on here, but in broad strokes it's simple: force matters most at the beginning of the race when you're accelerating, so that's when the race is closest. As the race proceeds, the effects of women's shorter steps and metabolic fatigue (yes, you get tired even in a 100-meter sprint!) take over, particularly in longer sprints. That's why, in the first graph, we see the male-female gap gradually widening as the race distance goes from 60 meters to 800 meters, by which time the benefit of greater force at the start of the race is irrelevant.

What, then, do these results tell us about the knotty problem of who should be allowed to compete in the women's category in elite running competition? The current World Athletics rules restrict DSD athletes only in events between 400 and the mile, which looks suspiciously like a targeted attempt to keep Semenya, the 2016 Olympic 800-meter champion, out. After all, what possible rationale could there be for restricting some events but not others?

While McClelland's paper doesn't wade into this debate, it does bolster the case that the male-female gap might depend on the specifics of the event. It's worth recalling that Semenya wasn't the only 800 runner affected by the new rules. All three women's 800-meter medalists at the 2016 Olympics--Semenya, Francine Niyonsaba, and Margaret Wambui--turned out to have DSDs that led to higher-than-permitted testosterone levels under the revised rules.

What's special about the 800? Maybe it's not a coincidence that it's the breakpoint between two regimes in that first graph above. For shorter events, women's greater force-to-mass ratio starts to reduce the male-female gap, so athletes with characteristics that fall between typical female and male values would have less of an edge. For longer events, smaller body sizes tend to be advantageous (and indeed, elite marathoners have been getting shorter and lighter since the 1990s), so the larger body sizes and greater muscle mass associated with higher testosterone levels may become a liability for DSD athletes. That's pure speculation: there may be other biomechanical or physiological factors at play, or the apparent preponderance of intersex 800 runners may simply be a fluke. I suspect that this particular World Athletics rule is unlikely to stand the test of time, given the recent success of DSD athletes at distances like 200 and 5,000 meters. But perhaps the broader point is that the performance gap between men and women has once again been shown to be more nuanced and multifaceted than we suspected, and we should be wary of anyone on either side of this debate who claims that it's simple.

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.

For exclusive access to all of our fitness, gear, adventure, and travel stories, plus discounts on trips, events, and gear, sign up for Outside+ today.