Running Form Comparison: Bekele, Gebrselassie, Farah

Running Form Comparison between Bekele, Gebrselassie, and Farah in the 2013 Great North Run

I came across this video clip via Peter Larson. Despite the fact that Bekele is turning through much of it, it gives us an opportunity to look at the running form differences between three great runners: Kenenisa Bekele, Haile Gebrselassie, and Mo Farah.

You may have first looked at footstrike, checking whether they heelstrike and/or overstride. These questions are so much the focus of research and discussion about running form these days that it can sometimes seem like they’re all that matters.

However, footstrike is really just the tip of the iceberg, and if you are trying to change your footstrike you need to know something about the whole iceberg, not just the part that’s most easily visible. So I’m going to walk you through a few of the things I look at when I evaluate a runner’s form and maybe this will give you some new ideas about what to feel for in your own form and look for in others’.

Before I go any farther, though, let me clarify that this commentary represents me adapting my evaluation methods to video. In reality I rarely use video to work with runners because what you can’t see in video dwarfs what you can, and my analysis doesn’t depend generally on particular moments in the gait cycle but how the cycle unfolds – in short, I analyze movements, not a series of positions.

That said, here are three very interesting positions I’ve pulled out of the video:

1. Midstance

The first collage is midstance for each of the runners, plus or minus the margin of error of my reaction time on the YouTube “pause” button. If you wondered whether any of these runners were overstriding, this should set that question to rest.

Yes, yes, I know these are not pictures of footstrike! Frankly looking at initial contact of a foot with the ground can be confusing because whether it’s the impact-transient-causing type of event depends on what is on the foot, how stiffened the leg is on contact, how the weight progresses over the sole of the foot, and even how fast the runner is going. At the speed these guys were running the foot definitely touches the ground out in front of them, but their bodies are moving so fast that by the time there’s any significant force through that foot they’re on top of it.

With so many conflating factors in visually determining whether a runner is overstriding I’ve stopped caring much about the question unless it’s extreme, and I’ve switched to checking where their weight is in midstance instead. If a runner is overstriding in a meaningful way, their hip joint will be behind the ankle in midstance. These guys all have their standing hip joints exactly over their ankles in midstance. Clean as a whistle. (And in case you’re wondering how I determined midstance, I picked the moment their heads were lowest.)

So they’re not overstriding. However there’s a major difference between the first two and Farah. Bekele and Gebrselassie are both leaning forward (And please note in this part of the gait cycle it looks like “bad” leaning, just from the pelvis. It isn’t, as you’ll see in the other pictures.) If you drew a line from ankle upwards through the hip joint and continued up to head height, both their heads would be in front of it. However the back of Farah’s head lies on the line. He’s much more upright than the other two. (In this context I’m using “upright” to mean how close to vertical his spine is, not how straight it is.)

2. Toe-off

In this picture you see the three at the moment of toe-off. As you can see, their shoes have just lost contact with the ground. I’ve taken a shot at drawing lines from hip joint of the toe-off leg through the neck to the head, it looks a little sloppy and I didn’t have the control to get them exactly where I felt they were most accurate. (If anyone knows of a better tool for doing this than picmonkey.com, I’d be grateful if you’d leave a comment pointing me to it!) In pictures where someone has drawn a line through the body I often disagree with their placement of the line as being not anatomically precise. It’s easy to draw these things in a way that confirms your bias. I’ve aimed my lines to run from hip joint parallel to the spine, through the neck and into whatever part of the skull they arrive in. I didn’t get all three necks the same, Gebrselassie’s should angle yet a little more forward. Even so, you can see that Bekele and Gebrselassie are leaning forward about the same at this point (Gebrselassie actually a little bit more) and Farah is quite a bit more upright. You can also see that Farah’s chin is more tucked than the other two. In this picture it’s hard to see Gebrselassie’s chin but overall in the video you can see that he tucks his a little more than Bekele.

Aside from lean you can also see that Bekele and Farah have their left shoulders and hands behind the line I’ve drawn while Gebrselassie’s shoulder is right on it and his hand is in front of it. This shows something that will be even clearer in the next set of pictures, which is that Gebrselassie uses less trunk counterrotation than the other two when he runs. You can also see here that his back leg is not as straight as the other two have theirs. This is connected to the counterrotation issue – less rotation gives him a shorter stride length which he compensates for with a higher stride rate. I counted the stride rates of the three while watching the actual race and got approximately 180 for Farah, 188 for Bekele, and 204 for Gebrselassie! All of them are reasonable for runners of their level but it’s worth observing that Gebrselassie and Farah use opposite strategies for speed – Gebrselassie takes shorter, super-quick strides with minimal torso rotation while Farah takes super-long, slower strides with huge torso movement. Bekele is right in the middle in balancing stride rate/stride length.

3. Maximum Trunk Counterrotation

In this final collage I’ve selected what looks like the point of maximum trunk counterrotation for each of them, which happens in the flight phase of the gait cycle. These pictures are the hardest to compare because each runner is captured at a different angle and Bekele’s head is still turned, which limits how much he could turn his upper body the other direction. It looks from the leg and foot position like I’ve caught Farah a little bit later in the gait cycle than the others but I’ve watched the video a million times and this seems to be the moment of maximum forward movement of his shoulder, so I’m going to go with it.

The first thing that stands out to me is that the difference between the forward lean of the first two runners and Farah has become very large indeed. His spine looks absolutely vertical. We’ve seen this change in the angle of his trunk to the ground over the course of his gait cycle, and this reveals the biggest difference between him and the other two: Farah moves his trunk in the sagittal plane when running and they do not. He goes from a moderate forward lean at midstance to almost upright, then leans forward again for the next stride. The other two maintain a steady lean through out their gait cycles.

You can easily see an indicator of Farah’s sagittal-plane movement in full-speed race video, for instance the 10,000 meter final from the Olympics. When seen from the side, his head clearly bobs forward and backward (relative to the speed of his body overall) in contrast to the heads of all the runners around him which just move smoothly forward. Farah positions his trunk so that when he pushes off in late stance his head is behind the line of force and is pushed backwards – in other words, pushing off makes him arch his back a little. And then he has to counter that movement by activating his trunk flexors to bring both his head and leg forward for his next footstrike. Very few world-class runners do this; I believe very few runners who do this rise to world-class level because it is less economical than positioning the torso so the force in late stance is transmitted through the spine to the head, pushing it smoothly forwards instead of backwards. However, a runner’s success is created by a constellation of attributes and economy is only of them, and Farah depends on it less than the other two.

What causes Farah to organize his running in this more expensive manner is something that can’t be determined visually; I’d have to put my hands on him and feel what his neck is like, his ankles, his upper spine, and so forth to know the answer. (Incidentally, the level of tension a runner would feel in their neck, face, and jaw running like this would lead to a lot of grimacing.) In my experience, people instinctively choose to be economical when they can feel how to do it, and if a runner isn’t making that choice there is probably what Feldenkrais practitioners call “parasitic effort” getting in the way.

There’s still more to say about this: the possibility that Farah is at maximum trunk rotation later in his gait cycle than the other two points to the overall larger trunk movements he makes, much of which can’t be seen from the side. His head moves more side-to-side than other runners’, as do his shoulders and ribcage. When he stretches out his stride at the end of a race he makes his fabulous stride length happen without overstriding through this huge trunk movement. The possibility that it might be slightly out of sync with his legs – slightly delayed – coupled with the sagittal-plane trunk movement so unusual for a runner add detail to the impression I always get that he’s “pumping” not only with his arms but somehow with his torso, pushing harder against the ground and driving his legs through the whole cycle by tremendous auxiliary motion of his trunk. This is strength, not economy, that he calls upon to perform his best.

You can see a small difference between Bekele and Gebrselassie’s backs in these pictures as well. Gebrselassie is a little straighter, you could almost say his back looks arched as well. That difference would likely seem larger if Bekele were looking forward and his shoulder were able to move naturally – he would look even more relaxed and forward-leaning, based on how he looked in the rest of this race and how he usually looks. Gebrselassie has always had a slightly more arched back than the average East African distance runner, and a tendency to tuck his chin as well though under pressure he lets it move forward in more efficient fashion – the opposite of Farah’s response to pressure. Bekele by contrast has always had an exceedingly clean lean and easy trunk mechanics free of distortion. If I had to be any runner in the world other than myself, I’d want to be him.

(If you haven’t seen the whole race, you can probably find it on YouTube and if you’ve read this far you should really just go watch it! Then read the rest of this blog post.)

Under pressure at the end of the race, Farah pumped in all the ways I’ve described, reaching out of his form for more power anywhere he could find it. He found it and closed on Bekele in amazing fashion. Bekele, nearly overtaken, seemed to smile calmly and then inexplicably sped up without appearing to do anything different, and won the race.

Does Bekele’s win mean his way of running is best? I think at this level “best” is a vague and possibly useless word. Farah beat Bekele in the Olympics 10,000m, after all, and Gebrselassie is more accomplished than either. I think we could eventually agree that Bekele’s style is most efficient and Farah’s least, or, to flip it around, Farah’s is most powerful – even extraordinarily so – and Bekele’s least, with Gebrselassie striking perhaps the perfect balance and possessing more versatility than either.

For me this becomes an aesthetic question. They are three extraordinary runners with very different styles, bodies, and personalities, and so they give very different gifts to those of us watching. I think we love to watch Farah run because he embodies desire and force of will, Gebrselassie displays amazing speed and lightness, and Bekele channels calm and grace.