I was so excited to actually be in Berlin to watch the marathon in person this year. I took my camera along to get some high-speed (aka slow motion) video to use for analysis, as I’ve done every January at the Great Edinburgh Cross Country.
I learned, however, that there’s a big difference between videoing a road race and a cross-country one. For instance, the fleet of cars, trucks, motorcycles and bicycles that got in between me and the runners. And also the width of the streets — my camera won’t zoom when recording at high speed, and because I turned out to be on the far side of the street from the runners, they ended up tiny.
What can I say, I’m no videographer. Now I know what to watch out for next year, though!
I make all these apologies because I’m about to share a video with you that is truly not good. It’s of Wilson Kipsang and Kenenisa Bekele around the 37k point, Kipsang in the lead.
There are a few things you can see in this video–perhaps better because it’s low resolution and the runners are small. When detail is absent you see the big picture better. I even use this sometimes in working with clients–to decrease detail I half-close my eyes and the simpler form I see shows things I didn’t see before.
Kipsang, as always, is running with his arms swinging low, near his hips, and his shoulder blades pulled slightly back. This gives him a gliding style of running, with an easily moving pelvis.
Bekele, as always, keeps his hands very close to his chest, swinging them up to a little above heart height.
In this video you can see Kipsang’s lower legs, as compared to Bekele’s larger leg action. You can also see how much less Kipsang turns his upper body than Bekele. The gap in space makes the comparison not immediate, but if you watch it twice you will see it.
The second time you watch, notice how the forward movement of Bekele’s shoulder as he moves into flight makes a single long line of his side from shoulder to foot, creating his big leg action.
My initial impression was that this gave Bekele a greater stride length than Kipsang. But after watching my own video and others’ many times, I don’t feel I can accurately eyeball their relative stride lengths.
People get very caught up in what happens in the hip joints to create stride length — ability to hyperextend the hip of the back leg and flex the hip of the front leg, which gets translated to hamstring and hip flexor flexibility.
But of course stride length isn’t just a result of how great the angle is between the two thighs in flight, but also the distance covered in flight. And that’s where my questions about the difference between Bekele’s form and Kipsang’s lie.
Now that you’ve seen these things, take a look at some “real” video from the broadcast of the race and, with your eyes primed from squinting at my bad video, you’ll see these same things again — Bekele’s considerably larger upper body movement and how it generates a large leg action.
Notice also Bekele’s very loose, easy head movement. This kind of freedom in the upper cervical spine makes all the large movements beneath it possible, including his upper body rotation.
In my experience working with runners, the “differentiation” Bekele has in his upper body, meaning his ability to articulate all the different joints and allow them all to contribute motion, normally translates to lower percieved effort, increased stride length, and greater speed. Kipsang’s upper body appears to move as a single unit; seen from the front his head doesn’t move relative to his neck the way Bekele’s does, nor do his shoulder blades move as much relative to his ribs.
However, it looks like Kipsang may be covering as much ground per stride as Bekele, at least in my video, and this would be by virtue of his good pelvis movement.
So this leaves me with a buring question: do Bekele’s larger upper body and leg movements (which occur together and enable each other) indeed translate to lower perceived exertion, or even lower heart rate or longer stride length, than Kipsang’s at the same speed?
I’ll be honest, I’m a huge Bekele fan, on an aesthetic level the beauty of this run was simply awesome.
I’m sorry to not be commenting on any other elite men or any of the women at all, but my video isn’t usable and the price of watching in person is that I wasn’t harvesting screenshots on my computer through the race. So my respect to all the runners (including Kippsang, of course), and I promise to do a better job next year.
One of my favorite things to do after watching a marathon is try on the different runners’ styles on my run the next day. Give it a go yourself — try out what you picked up from watching Kipsang and Bekele and let me know what you discover.
Jae, I couldn’t help but notice Bekele’s range of motion from the knee. He seems to be using almost all of it, almost locked at footstrike to almost kicking himself in the butt. Nevertheless, the range of his thigh seems very normal compared to other runners.
I agree, Steve. After sleeping on it I wasn’t confident at all that there was a difference in their stride lengths, so I’ve revised the post to reflect that.
Of course, if they are running the same speed, the runner with the slower cadence has the longer stride length. I didn’t spend much time/effort looking at it, but to my eye it seems like their cadence was pretty darn close (and hence stride length too).
Yes of course, David. But I don’t feel like either my tidbit or the German broadcast clip give me enough info on their cadence overall, though Bekele’s cadence did seem higher than Kipsang’s coming up to the finish line. At that point, though, the way the camera was swooping around I couldn’t tell if the gap between them was growing or not. This is the sort of thing I’d have a better feeling for if I’d watched the whole broadcast rather than being there on the street. Tradeoffs.
Running style is an overrated independent variable of running. I rest my case in two great runners who, both, have terrible running styles. Tsegaye Kebede and Paula Radcliff
Hi Marco. This is the first analysis I’ve done in years where I’ve forgotten to give my usual disclaimer that running form is just one of many assets a runner may have and we can’t fully ascribe performance to it. I’ve always known the minute I forgot to say this, someone would leave a comment saying “Paula Radcliffe!” so thank you for that satisfaction. I disagree with you about Tsegaye Kebede though, the way he runs works really well. Anyway, the thing about running form/technique is, who wouldn’t want another asset if they could have it?
Jae, Wouldn’t it be more important to focus on the relative contact time with the ground, centre of gravity when contacting the ground and how much energy is being lost as they transfer between strides. The relative stride length and mobility differences of each athlete will most likely not allow them to run with similar stride technique. Obviously both techniques are very efficient in their own way.
Ground contact time is definitely something I’d like to know as well, Owain. But as a consequence of my Feldenkrais training, mobility differences or, rather, movement strategy differences are the main thing I focus on. They are within the realm of things runners can make choices about when they can feel what they’re doing and understand their options. For instance, Kipsang could learn to bend his elbows more, move his shoulders more, and differentiate his upper spine, and he would immediately develop a bigger leg action and run somewhat like Bekele. And vice versa, Bekele could lower his hands and his upper body would move less and so would his legs. In my professional experience, Bekele’s strategy is the better one though a number of runners have done very well with Kipsang’s strategy (Desi Linden most clearly, but Meb can sometimes move in similar fashion). But watching my own video and then the broadcast clip made me wonder why…
Hello Jae – a slightly different request …
Have enjoyed your observations on elite strides – particularly these two as well as Rupp and Kipchoge posts. But as you encourage readers to consider implications for their own stride,I am always left wondering about relevance….let me explain.
This Berlin footage is of runners that are literally running at a pace that is faster than most folks reading here even can hold a 30 second stride for (430-445 pace per mile). I am a 2:41 marathoner – and my cadence / effort / lean / arm action / etc is far different at 7 min vs 6 min vs 5 min pace. So what do these elites look like on video at more “normal” speeds? Is there more insight we could observe or adapt in watching Bekele doing say his warm up at 630 or 7 min pace per mile pace – truly effortless for him but near race pace for us common folks? Thoughts?
That’s a great question, Scott. I think, though, that the question of what pace these runners are running is perhaps less significant than what effort. The overall dynamics I write about hold true for all runners — for instance the relationship between the shoulders and the leg range of motion. In essence, they’re true at every speed. The main aspects of coordination that change with speed are also pretty standard: amount of forward lean, the degree to which trunk action facilitates leg action, the amount of “snap” in the arms and their range of motion. But my sense is that these things are linked to effort, not necessarily speed. In other words, your form and Bekele’s form will go through similar changes as he approaches 2:03 marathon pace and you approach 2:41 marathon pace. From the outside you won’t look exactly the same but you’ll look more similar than if you both ran 7 minute pace. A famous scientist once expressed the view to me that you can’t learn about human evolution and how the gait of running works by looking at elite distance runners because what they’re doing is highly abnormal for our species. I disagree, because you can’t reach a pinnacle of performance as they do by using your body in any other way than how it really works. Or as close to that as any given individual, with their personal movement history and experiences, can get.