Research by Stephen McGregor, PhD, is showing that the worst way to improve your running stride is to consciously try.
Interview by: Matt Fitzgerald
Until recently we knew very little about the difference between skillful and unskillful running, which is essentially the difference between economical (energy efficient) and uneconomical running. All we really had were crude descriptions of “good form” captured by cameras, joint angle measurements, and so forth.
Within the past few years, exercise scientists have become increasingly able to study the running stride from the inside, as it were, and the emerging result is a radically new conception of what skillful running truly is. Among these scientists is Eastern Michigan University’s Stephen McGregor, coauthor (with me) of The Runner’s Edge, who studies the running stride using accelerometers, the same types of devices used in some speed and distance devices, but his are much fancier. McGregor’s studies indicate that there is no such thing as good running form. Rather, skillful running is the result of an unconscious, evolutionary process wherein each runner’s unique body finds its own best way to run economically, resulting in a form that is slightly different from that of any other runner.
McGregor’s research also suggests that any conscious effort to practice a universal standard of good form is far more likely to do harm than good. If you have gone in for, or considered going in for, the current trend of technique training (natural running, the POSE Method, Chi Running, etc.), you’d better read this interview with Stephen McGregor.
Competitor.com: Tell me about the instruments you use to study running technique.
Stephen McGregor: For the past few years we’ve been using what we call high-resolution accelerometers. They are triaxial, so they measure accelerations in three axes: vertical (up-down), medial-lateral (side to side), and anterior-posterior (fore-aft). We place them around the waist at the middle of the lower back. This approximates a person’s center of mass. When a person is running, there are a lot of things moving in space—the arms and the legs—but really what it boils down to is how you’re moving your center of mass, and what happens with that center of mass.
Different people run in different ways. From these devices we get information about how these different people run. In contrast to other ways of looking at this information, like a traditional gait lab in biomechanics, we get a lot more information. Most importantly, though, we’re not tied to a very constrained area or condition. We can use these devices in a lab, on a treadmill, which we do quite a lot, but we can also do it outside on a track or even an open run around the neighborhood.
The great thing about them is that they’re portable, but they give you a lot of high-quality, high-resolution information across the spectrum of movement patterns.
You’re interested primarily in gathering information that defines what it means to run well, correct?
Yeah, we look at the components of movement in different axes and how those axial movements are distributed. At this point we’ve tested a number of individuals who range the spectrum from people who really don’t exercise much at all up to collegiate and national-caliber distance runners, and we have a good idea what distinguishes a good runner from an energy expenditure standpoint.
And that’s what it boils down to. Everyone has a sort of engine that they were given—VO2max and the metabolism that comes along with it. The question is what can you do with that? The really cool thing about running is that you can have somebody who has a relatively small engine but they can do a lot with it—and that’s running economy. We have some examples of that in our data set, of people who are not gifted with a big aerobic engine but do a lot with that engine. On the other hand we also have some athletes in the database who were gifted with some incredibly large engines but are not especially economical.
We’re getting into some of the nuts and bolts of what objectively differentiates a good runner from a poor one and how you get there.
What are the key differences between the running technique of good runners and poor runners?
In one study we looked at untrained runners who were just recreationally active. We compared these runners to some collegiate runners. As you might expect, the untrained runners actually moved a lot more in space than the highly trained runners. They bounced up and down more, they moved side to side more, and they also accelerated in the anterior-posterior direction more, which essentially means they’re slowing themselves down and then speeding back up, and that wastes energy. That’s about as clear-cut a comparison as you’re going to see.
Now, we have also compared a number of runners of similar fitness but different ability levels, and what we’ve seen is that certain things that might be considered important really aren’t. For example, you might think that bouncing up and down more wastes energy because it’s working against gravity, but it appears, in trained runners, it actually doesn’t. And it actually might be advantageous. A lot of the good runners we’ve looked at do bounce up and down a fair bit compared to their peers at the same fitness level who aren’t as good. This is still less than those who don’t run, so it seems there’s a “happy medium”, if you will.
One of the other things that is apparent is that they may not accelerate as much in the fore-aft direction because they’re not slowing themselves down as much when they land and that means they don’t have to speed back up to maintain a constant speed.
Another observation that’s more esoteric is that the good runners appear to be less constrained in the anterior-posterior axis. When we use a tool from chaos theory, called entropy analysis, we see that entropy is greater in the anterior-posterior axis in good runners. The thought here is that they’ve learned through repetition to select the optimal patterns to run so that they accelerate and decelerate less in the anterior-posterior axis. This is something they don’t need to focus on, and as a result their running is less conscious or “forced” in this respect.
These characteristics of good running technique that you’ve identified—they are the result of just running, right? It’s not as if the runners who accelerate less in the anterior-posterior access consciously modified their form toward that end, and they’re probably not even conscious of that difference in their stride, no?
It doesn’t appear that there is any particular running approach, or form, that is common. If you take 10 different good runners and look at them compared to 10 different poor runners, there is really no single common theme that distinguishes the form of the good runners from that of the poor runners.
Now, if you look at the untrained person, there are a lot of differences. The trained runners have more of a forward lean, and are probably running more toward the midfoot than the heel. But within the population of trained runners there is nothing you can capture on a camera and put on a billboard and say, “This is good running form.” At least not at this point.
The runners on the Eastern Michigan team that we’ve tested all train in a very similar way but they have very different running styles. And you can actually look at some of them and say, “Wow, that person has really horrible form,” and they actually come out pretty good in our testing.
The point I’m getting at is that the only common theme in terms of the methods used to acquire the ability to run well are running a lot, and probably running fast and running against people who push you to push your limits. If we look at individuals who run a fair bit but don’t train with a group, they typically don’t exhibit some of the characteristics we ascribe to some of the best runners.
This is more of a qualitative than a quantitative analysis, but it makes sense. Again, from the field of complex systems, it’s what some people might call self-organization. When you run against people who push you to run faster, you find the optimal way to run that speed—or you don’t. Running is so complex that it’s difficult to put all of the pieces of the puzzle together consciously. The only thing you have to do consciously is try to run fast, and then the pieces of technique required to do that fall into place unconsciously. In other words, running technique improves through a process of self-organization, and that’s something you see in a lot of complex systems.
You’ve gathered pretty good evidence that running a lot, running hard, and running in groups leads to more economical running. Do you know of any evidence that conscious manipulation of the stride—I’m thinking of technique systems like Chi Running and the POSE Method—leads to more economical running.
No hard evidence. I won’t pick on anyone in particular, but there are a lot of running approaches where a person will espouse a particular form or technique for running and will say that running is improved and economy is improved, but there’s really no hard evidence that a conscious, technical approach, or a conscious change in form, can improve economy.
I think it’s a mistake to fall into the trap of saying, “So-and-so runs really fast, run like him.” For any successful exemplar of a particular form you can find a counterexample. The complexity of running leads to any number of different solutions to the problem of running.
The traditional view is that long-term improvement in running is primarily the result of changes in fitness. Are you not finding evidence in your work that long-term improvements in running performance are equally the result of improvements in running technique?
In a homogeneous population of highly trained runners, the skill of running, which imparts the running economy, is probably equal in importance to fitness. The skill of running becomes increasingly important as you get up into the range of people who have fairly high VO2maxes. A substantial deficit in VO2max can be made up with skill.
Who are the most economical runners and why?
There’s a well established trend in exercise physiology studies that running economy is inversely proportional to VO2max. So if you have a high VO2max, running economy will be poor, and if you have a low VO2max, running economy will be good. One of the questions I’ve always asked is, why? A VO2max is largely given through genetic inheritance. Why would it be that someone who has been given a big engine would also be given a poor economy, if economy is also a physiological thing that you are given?
What I think happens is that if you run a lot against other people, and you have a small engine, you have to make up the deficit you have with respect to the people who have bigger engines. So if you’re going to compete with them, you’ve got to figure out a way to run fast, and the only way to do that is to learn to run more skillfully, thus more economically.
Then if you look at the other side of the equation, an individual with a VO2max of 80 doesn’t really have to learn to run very skillfully. They’re winning a lot of races by virtue of their fitness, at least until they get to a higher level. So what happens when you have areas of concentrated running talent like in Africa or some of the running groups here in America, you get this selective pressure. So in these groups people who have big engines to begin with are forced to learn through competition to learn to run more skillfully.
You can maybe see it in the recent case of [Chris] Solinsky breaking the American 10K record. He’s running with a group of high-caliber runners who may be pushing him to run fast, faster than he might without the group to push him. He needs a fast group, though, not just any group; otherwise, he’s not being pushed.
Based on your research, what else can you advise runners to do to improve their running technique?
I would encourage people who choose to follow a particular running approach not to follow it blindly. I was speaking to a person the other day and he was telling me how he had tried this one running approach and had become more economical and he’s become a better runner as a result. I said, “Well, how do you know?” He really didn’t have an answer. His running times hadn’t gotten appreciably faster. His results on a performance basis were similar to what they were before he had learned this approach.
You need to have objective measures to base your results on if you choose to go down this path, otherwise it may lead you nowhere. Again, running is a very complex thing and there really is no approach that works for everybody.
The other thing you can do is try running faster. How much faster is the question. You want to avoid running too fast too much. But adding in faster running with caution will probably challenge your body to learn to run more skillfully. Running faster is a good way to run faster.