I do some freelance writing for a website called RunnersConnect. Part of my work there is a weekly column that involves writing short reviews of scientific articles on a topic relevant to runners. This weekly column gives me an opportunity to look into what the science says about some of the less-though about topics in running. For example, one of my articles was all about the various skin ailments (blisters and so on) that afflict marathon runners. A few weeks ago, I decided to look into the effects of wind on oxygen expenditure—that is, how much will the wind slow you down? I found some pretty interesting studies on the subject which involved putting a treadmill inside of a wind tunnel, then measuring the oxygen consumption of a runner when subjected to different wind speeds. As expected, wind had a significant effect on a runner’s efficiency: for each incremental increase in wind speed, the impact on running effort goes up by a power of two. So a 10 mph wind is four times as fatiguing as a 5 mph wind. The same effect applies to running speed in calm air. The wind resistance that has to be overcome at 5:40 mile pace is twice that at 8:00 mile pace (root-two slower). This is why wind can be such a nuisance (or aid!) in the sprints. A large percentage of a sprinter’s energy goes towards overcoming air resistance as he or she powers forwards. As a general rule of thumb, a wind equal to your running speed will slow you by about 6 seconds per mile (so running 6:00 mile effort with a 10 mph headwind will result in a 6:06 mile). Running with a tailwind only aids you by about half that amount.
None of this should surprise an experienced runner, however. We’ve all had plenty of stiff headwinds and generous tailwinds. But the researchers investigated something I might not have thought of: the drop in air resistance when you “draft” behind another runner. In a 1971 study by L.G.C.E. Pugh (the initials stand for " Lewis Griffith Cresswell Evans"), oxygen consumption values were first recorded for a single runner on the wind-tunnel treadmill.1 Then another runner was put on the treadmill (presumably this was a fairly large treadmill) and oxygen consumption values were recorded for both. Trailing about one meter behind another runner reduced air resistance by 80%! To confirm this, Pugh took air pressure measurements at various positions behind a single runner on the treadmill. As is clear in the image below, air flow drops precipitously in the wake of another runner.
|Air pressure is significantly lower in the wake of a runner. From Pugh.|
Pugh’s measurements were largely confirmed by a similar investigation nine years later by C.T.M. Davies.2 Drafting on another runner virtually eliminates air resistance. According to Davies, a runner in a middle or long-distance track race in still air can expect to save about 1 second’s worth of effort per lap. That is, if you draft behind a runner clocking 70-second laps, your energy expenditure would correspond to 71-second laps if you were running alone. Given the impact of a headwind, it’s only logical to expect this advantage to be magnified on windy days. Four seconds per mile is a huge advantage from a competitive standpoint, and the physiology behind wind resistance might explain why the runner who takes the lead early on in a race almost never wins unless he or she is significantly more fit than the rest of the field. Many coaches and athletes try to explain this away with psychology, saying that it is much more difficult to lead a race than to follow. And this is entirely true! But the physiology of air resistance is a non-negotiable fact. Mental toughness and a competitive mindset can’t change any of that, which makes front running a tactically poor decision for most high-stakes championships. How did “Pre” do it? Well, he was just that much more fit than everyone else! Keep in mind that runners behind YOU are only getting a competitive advantage if they are within a few meters of you. Having a large lead neutralizes the effect of wind resistance, at least for the next runner back. Of course, to actually establish a large lead might necessitate some poor pacing choices from a physiological perspective, but that’s a topic for another day!
So, if running from the front (or out by yourself) has a consistent and measurable negative effect on performance to the tune of one second per 400m, what are the implications? For one, it emphasizes the importance of getting yourself into highly competitive races. In my opinion, too many colleges (and probably high schools) short-change their runners by going to low-key meets without strong competition. A top-flight 5k could boost your time by 10 or 12 seconds vs. running alone at a dual meet. And that’s in calm air—it could be much more with a wind. Additionally, it helps explain why taking the lead early on in a distance race is usually a big mistake. There are plenty of other tactical considerations to factor in, of course. If you’re a big kicker and want to make sure the race goes out slow, you can get to the front and slow it down for a few laps before tucking back into the pack. Or you could take the lead to help a teammate hit a time standard or run the kick out of a fast finisher. But from a completely selfish perspective, the only time to take the lead is when you are sure you can open up a significant gap (several meters or more).
So, when you are watching the Olympic track action get started at the end of the month, don’t be surprised to see races going out at a woeful pace as the field plays a game of chicken to see who will be the “sacrificial lamb” and take the lead. And you can bet money that the leader at 1/3 or 1/2 of the way though the race won’t be the winner at the end. Not necessarily for a mental reason—just plain old fluid dynamics.
1. Pugh, L. G. C. E., The influence of wind resistance in running and walking and the mechanical efficiency of work against horizontal or vertical forces. Journal of Physiology 1971, (213), 255-276.
2. Davies, C. T. M., Effects of wind assistance and resistance on the forward motion of a runner. Journal of Applied Physiology 1980, 48 (4), 702-709.