If there is one thing that I have learned in 15+ years of teaching and studying biology, it’s that variation is the norm rather than the exception. Variability is the reason why biology is sometimes called the “sloppy science,” and it’s the reason why undergraduate biology majors at my college are required to take a course in statistics. Statistical methods are the tools that we use to tease out pattern from often messy piles of data, and they allow us to come to not always very clear-cut conclusions. The upshot of this is that in any sample of data there is going to be variation, and there are going to be outliers who don’t follow the statistically supported pattern.
As an anatomist, variation is inherent in nearly everything that I teach and study, from the frogs that I wrote my dissertation on, to the cats, pigs, and sharks that my students dissect in the laboratory. Variation is the bane of my student’s existence in the anatomy lab, and it often comes as a surprise when I tell them that the exact pattern of branching blood vessels in one cat may not be exactly the same as that in the cat on the next bench over. Some cats are muscular brawlers, whereas others have muscles so wispy that it’s a wonder that they could support their own body weight. Anyone who has taken human gross anatomy in medical school has probably observed the exact same thing in the human cadavers that they dissected. Variation is normal. In fact, it’s essential – variation serves as the raw material upon which natural selection can act in the process of evolution, and it’s ultimately the reason why we can now run down the road on two legs instead of four.
So what does this discussion of variation have to do with running? I’ve been thinking a lot about running form lately, and have gone so far as to start experimenting with my own stride. It is therefore with a great deal of interest that I read the following quote by Alberto Salazar: “There has to be one best way of running. It’s got to be like a law of physics. And if you deviate too much from that–the way I did in my career–it can be a big handicap.” From a purely biomechanical perspective, I tend to agree that there is probably an optimal way for a human to run, but the biologist in me keeps coming back to the topic of variation. Humans are highly variable animals, probably much more-so now than when we were developing our distance running skills millions of years ago out on the African savannah. Take one glance at any gathering of humans and you will see short people and tall people, thin people and heavy people. You will see some with biomechanical abnormalities. If you could look beneath the skin, you’d also find variation in bone structure and muscle composition, as well as innumerable other anatomical and physiological differences. Getting back to Salazar, variation is why they had to develop a shoe insert for Dathan Ritzenhein’s unique metatarsal structure. Given this, I find it unlikely that there is a single “perfect” running form that will equally apply to every single human being who runs. I may not be an expert on running mechanics (not even close!), but I know a thing or two about variation, and the realist in me says that what works for a sub-5:00 miler may not work for those of us who run our miles at an average pace of 8:00 per mile or higher. Variation in all of its forms needs to be taken into account when talking about things like an ideal running form.
My goal in this post is to address the topic of variation in running form as it applies to a small sample of elite runners. This past April two of my students went down to Boston and filmed the first 1000 or so runners to pass the ~17.5 mile mark in Newton during the Boston Marathon (on Washington St. near the intersection with Commonwealth Ave. – see picture at right from Google Street View – it was a relatively flat stretch of the course). The film was recorded in slow motion at 300 frames/second (see below), which allows a high degree of accuracy and detail when examining aspects of the running stride. Despite this, it can still be hard to see exactly what is happening in a streaming video, so what I have done here is to extract still images from my videos for four runners at a standardized moment in the gait cycle (I will be doing more of this at different points in coming posts). The runners that you will see are four of the top five finishers in the 2010 Boston Marathon: Robert Cheruiyot (the eventual winner), Tekeste Kebede (2nd place), Meb Keflezighi (5th place), and Ryan Hall (4th place). In the yellow singlet behind Cheruiyot you can see Deriba Merga, who rounded out the top five by finishing 3rd – I have omitted him below since he is somewhat obscured by Cheruiyot in the video.
What follows is not an exhaustive analysis of running stride in these runners (I’m hardly qualified to do that), but rather an attempt to point out that even at the highest level of competition, running form is highly variable. Why is this important? The main reason in my mind is that you can find innumerable published descriptions of how best to run, or what the optimal running form is. Often these descriptions share key points (e.g., avoid overstriding, avoid excess vertical and horizontal motion, etc.), but they also often differ greatly in the details and it’s hard to know just what you should be doing out on the road or trail. For this reason I’m going to take a look at three aspects of the running stride and discuss the variation seen among these four elite runners. We’ll start with one of the most hotly debated aspects of running form these days – footstrike.
If you want to start an argument among a group of runners nowadays, all you have to do is bring up the topic of footstrike (or how it relates to shoe choice, but we’ll avoid that equally controversial topic here). A lot of runners are interested right now in the idea of adopting a more “natural” midfoot or forefoot footstrike (I’m very admittedly guilty of this myself and have been playing with chaining my form for most of this summer). Others think this change unnecessary, and don’t buy the argument that moving away from the ubiquitous heel strike will improve efficiency or reduce injury risk. It’s not my goal to settle this argument here, and I think there’s a lot of work that needs to be done before we can come to a clear conclusion on the issue. Nevertheless, I like to experiment, and the evolutionary biologist in me is a bit enamored with the idea of running as our ancestors did. Maybe I’m crazy, but I’m having fun.
Back to the point, the ultimate question is whether one style of footstrike is “optimal” or better than the others. Famed runner and Nike Team Coach Alberto Salazar clearly thinks that midfoot/forefoot is the way to go, and was recently quoted as saying the following about heel striking: “It’s like having a tire with a nail in it.” He has gone so far as to recently convert two of his elite runners, Alan Webb and Dathan Ritzenhein, from heel striking to midfoot striking, and it’s going to be very interesting to watch how these guys do going forward (Ritzenhein is running the NYC with his retooled stride this Fall). Barefoot and minimalist runners also advocate for a midfoot/forefoot footstrike, mainly due to the fact that scientists like Daniel Lieberman have shown that this is the way humans run when not raised in built up, high-tech shoes (not to mention that it’s very hard to heel strike while barefoot or in an un-cushioned shoe). Lieberman also showed that heel striking is associated with a greater initial impact force, though how that plays out in its potential relation to injury risk remains to be seen. On the other side of the issue, you have the running shoe companies, who have been producing shoes for nearly 40 years that are usually designed to absorb shock associated with a heel strike. The vast majority of runners out there run in these shoes and are heel strikers, and there is some risk to moving away from a shoe style that our body has adapted to. That’s not to say that it can’t be done, but it takes time and effort, and many don’t see conclusive evidence of a benefit big enough (or any benefit at all) to justify the effort needed to change (I have variously felt this way myself in the recent past).
One of the interesting notions that I have seen espoused in various places is the idea that elite runners never heel strike. I’m not sure where this belief comes from, but it’s out there. However, it takes merely a quick glance at the pictures below to see that this belief is untrue. Of the four runners pictured, all but one make initial contact on the heel (Ryan Hall is clearly landing on his midfoot). That being said, the degree of heel strike seen is highly variable. Robert Cheruiyot lands only very slightly on the heel, and it’s almost indiscernible when watching the streaming video (the pictures below depict the exact moment of foot contact with the ground). His weight almost certainly comes down on the midfoot, and his heel strike is about as mild as a heel strike can be. Tekeste Kebede has a more pronounced heel strike with a more extended leg, and Meb has by far the most pronounced heel strike of three (while apparently wearing the same shoe as Cheruiyot). Interestingly, both Cheruiyot and Kebede strike on the midfoot on the right side (not shown – I wanted to standardize the side shown on all runners), showing that there is variability even within individuals. I wouldn’t be surprised in the least if there was also some degree of variability in the same foot from stride to stride. The point is that even among elite runners competing in one of the highest profile races in the world, there is a high degree of variability in form, and they clearly have not converged on a single “optimal” running form.
One can come to two conclusions from observations like this. First, perhaps these guys are using the footstrike that works best for their individual bodies. Maybe a midfoot strike is what works best for Hall, and a heel strike is what works best for Meb. Alternatively, it might be possible that some of these guys are doing things that are not optimal and could be improved through a bit of stride tweaking. Based upon recent events, I’d suspect that if Meb went out to the Nike Complex in Oregon to work with Salazar he would be weaned off his heel strike as the first order of business. Meb won the 2009 NYC Marathon, but has also suffered numerous injuries in his career, both running related (he fractured his hip in the 2007 Olympic Trials), and flukey (he injured himself trying to escape from a bulldog). It’s hard to say what such a change might accomplish, but it’s certainly interesting to think about, which is why Ritzenhein’s performance at NYC this Fall will be a must-watch.
What does all of this mean to everyday runners like me and you? Well, if this level of variation is apparent in 4 of the 5 fastest runners at Boston, you can be sure that it would be found among recreational runners as well. It also shows that you can run really fast using either a heel strike or a midfoot strike, but it doesn’t say anything about whether altering stride from one form to the other would result in improvement. Maybe Meb would be faster with a midfoot strike, maybe he wouldn’t – we’ll probably never know. Incidentally, this is actually the major reason that has pushed me to try to move away from my own heel strike – the experimenter in me is simply curious to see what will happen. Will it make me faster? – who knows. Will it reduce my lifetime risk of injury? – maybe, maybe not. The one sure thing is that I’d never get any kind of answer to either of these questions if I didn’t try it.
Though less controversial than the topic of footstrike, the way to carry the arms while running is also a form characteristic about which I have read differing advice. Is there one best way to carry your arms while running? Perhaps, but referring again to our elite pictures, you’ll see a great deal of variation in style. Cheruiyot and Meb both carry their arms high with a pronounced flexion at the elbow. Kebede carries his arms a bit lower, just above the waist, whereas Ryan Hall has a unique style in which he keeps his palms open and carries his hands quite low, below the waistline (for more on the origin of Hall’s unique arm carriage, read this – thanks to Rick for alerting me to this in the comments). On my run yesterday I tried to pay attention to what I do, and noticed that I utilize a high, flexed carry like Meb and Cheruiyot. When I tried using Hall’s style, it felt awkward and uncomfortable, but far be it from me to tell Ryan Hall to carry his arms while running. What Hall does seems to work well for him, and like Meb and his heel strike, it’s hard to know if tweaking of the arm carriage for any of these guys would provide any benefit. Ultimately the point once again is that there is variation – elite runners at the highest level of competition do things in very different ways.
The final aspect of form that I’ll address here is body orientation. I have variously read that the torso should be oriented upright (near vertically) while running, or with some degree of forward lean. This is another case where I don’t really know what to believe, or if one method provides a benefit over the other. However, returning once again to our photos, we can see that Cheruiyot and Meb are fairly upright in their posture, whereas both Kebede and Hall seem to employ a slight forward lean. Once again, we see variability, with little ability to determine if each runner is doing what works best for them, or whether some of them are doing something less than optimal that could be tweaked to provide a performance boost.
The basic take-home message from this post is that even elite marathon runners exhibit variability in some of the most basic aspects of running form. If a “perfect” or “optimal” form exists for all runners, either only one of these guys is doing it, or none of them are. More likely, I suspect the answer is that each of us individually has a form that works best given our own anatomy and physiology. That’s not to say that we shouldn’t work on our running form – quite the contrary. The are likely ways that we can all improve, and experimentation, if done carefully, can’t hurt. Do some research and hit the track and try things out, record yourself on video and note your strengths and weaknesses, or maybe enlist the help of a coach – any of these things could potentially help you and your running.
I’ll finish with a few personal thoughts. I’ve spent the better part of the past year experimenting with my running form, particularly with a slow migration away from a pronounced heel strike. Will moving to a more midfoot or forefoot strike make me a better runner? – only time will tell. At least in my case I have some confidence in the path I have chosen. When one of the country’s leading anthropologists shows that humans evolved to run on the forefoot, I tend to listen (I am an evolutionary biologist after all!). When one of the country’s top coaches (Alberto Salazar) at what is probably one the most technologically advanced training centers in the world (Nike in Oregon) starts shifting some of the country’s leading runners from heel to midfoot, it’s not done without careful consideration and study. We have both our own evolutionary history and presumably cutting-edge modern research at Nike both apparently pointing in the same direction, and that is enough convincing for me to at least give it a try. I’ll never be able to run as fast as any of the elites that I’ve talked about in this post, but if my running form transition gets me a new 5K PR or a step closer to my much coveted BQ, the change will have been well worth the effort. If neither happens, at least I will have learned something and I’ll surely have fun in the process, and isn’t that what running is all about?
Update 8/10/2010: Amby Burfoot has added his thoughts on the issue of variability in running form on his Peak Performance blog: http://peakperformance.runnersworld.com/2010/08/aug-10-in-search-of-perfect-running-form-the-debate-continues.html
Update 8/11/2010: For another interesting series of posts on running form, check out this series from Jonathan Dugas and Ross Tucker of the Science of Sport blog: http://www.sportsscientists.com/2007/09/running-technique-is-there-right-way-to.html
Update 8/17/10: Added a second post on running form – this one addresses the question of where the foot should land relative to the center of gravity of the body.