Marathon WR in Slow Motion: Dennis Kimetto Running Form

I do like an opportunity to watch elite athletes in slow motion. They make it all look far too easy!

Here’s a great video showing Kenya’s Dennis Kimetto in the closing few kilometres of running a World Record time of 2:02.57 in Berlin this year.

Some Observations

Clearly this is footage pulled from TV, rather than good quality high-speed camera footage. So we can’t get too into the more subtle areas of running biomechanics. But we can look at a couple of key areas…

Running Technique Quick Guide [FREE PDF]

Cadence / Stride Frequency:

Earlier the race, I picked a few of the leading group of male runners, and simply counted their cadence (strides per minute). As you’d expect, there was variation between the runners, but all those I observed were in the 180-190spm region. At the pace these guys are running, that shouldn’t come as a surprise.

What I found interesting is that in this footage collected nearly 41km into the race, Kimetto’s cadence is still well over 180spm. That’s over three steps per second – which you can observe using the race timer in clip! Anybody who has really pushed themselves in a race will know that cadence (leg speed) is usually one of the first things to drop as fatigue kicks in.

We can assume that Kimetto had to work pretty darn hard to set a new WR, which makes the consistency in his cadence impressive!

Remember though that cadence is variable in determining running pace, not a constant… more on that here and here.

Stride Length & Foot Strike:

The tongue-in-cheek title of Craig Payne’s article on this run made me smile:

Heel striking in a cushioned shoe with a 10mm drop and he still managed to break the world marathon record!

It’s important to remember that not all heel strikes are made equal. Also that there is a big difference between the glancing / proprioceptive heel strike we see from Kimetto, and the heavy, over-striding heel strike we see from many less skilled runners.

I’d also argue that on his left, Kimetto is heel striking a little more pronounced than on his right, which looks to be more on the borderline between heel strike and midfoot… Hard to say with this quality of footage through! This just goes to show how even the elites often display notable asymmetries in running gait.

“Isn’t he over-striding?”

That’s a question I’m often asked when people see elite distance runners and show surprise at how far ahead of their hips they strike the ground.

I’ve been guilty in the past of judging distance runners on whether they land close to under their hips / centre of mass, or not. Using this as a simple marker of whether they’re over-striding or not. This is unfair I now feel…

When working with sprinters, I do want to see them strike the ground as close to under their hips as possible when sprinting at maximal velocity. Marathoners are not sprinters!

When working with distance runners, the over-striding marker I now look for is whether the point of initial foot contact comes as the ankle is positioned underneath a flexing knee (regardless of foot strike pattern). What we don’t want to see is a more extended knee aligned behind the ankle as the foot strikes the ground – essentially running with the brakes on!

Kimetto achieves this ‘ankle under flexing knee’ nicely as the foot strikes the ground.

I’ll cover ‘markers for over-striding’ with reference to the differences between sprinting and distance running in another article 🙂

Feel free to share your thoughts on this footage in the comments section below…

Last updated on March 2nd, 2021.


  1. I notice the whole forward lean. When his rear leg is straight (almost) his body is 10 to 15 degrees over vertical

  2. I loving watching the video. Such an effortless looking stride, despite the quality of the footage. It does look like he initial strikes in front of centre mass, although hard to tell from the footage quality, but it doesn’t look like it is loaded until it is well under and before you know it the foot is coming up and through again. It makes me feel like a plodding elephant and I didn’t think that my stride was too bad these days 😉

  3. I agree, and it’s really obvious to me, he lands on his heel more with the right foot. That can also be explained by the fact that it’s past 40km already and the fatigue is obviously there. At somewhat a totally different level, I felt the same at the end of my last marathon, I could feel that one my foot was more heel striking than the other and that I had to concentrate to go otherwise but almost couldn’t.
    I agree with Richard just above, he’s so bent forward that although he’s overstriding, his knee still lands right under him. Impressive.

  4. You can say what you will about (potential) problems in his mechanics, but the bottom line is that this guy holds the world record. Could he be a little faster if he corrected these issues? Sure. I can only dream of being 1/2 as efficient as he is. At this point in time, I’m just pushing myself to see if I can get on the same hour as the WR.

  5. Just more proof that over time, over many miles, our body develops form that is best and most natural for our bodies. That not everyone had the same exact bone structure and angles so not everyone is going to run the exact same.

    1. Would have to stress that people don’t just “DEVELOP” form over time. Good form must be repeated and developed with your mechanics analyzed repeatedly. Once this is done, you can then determine your deficits which will need to be worked on (flexibility, strength, and/or coordination). Most of the runners that have efficiency or pain issues have underlying orthopedic deficits that subject the individual to being a poor and inefficiency runner. It’s not just a case of not have the appropriate form and practicing your form and not just strengthening or stretching… If you have crap tools or a poor ability to use good ones, it’s going to affect your end product… that’s why we have sports medicine physicians and sports specific physical therapists that have years of training and education to determine what aspects are wrong with you before you decide that it is solely a “form” issue. It’s never just one issue and each individual will have varying percentages as to how a change of form and just strengthening will improve their situation.

  6. Whatever the mechanics say, one can but only admire such a fantastic effort, 2:02:57 is truly amazing …. for now. It will go again no doubt to below that time. Beautiful long GLIDE slow-mo.

    There’s alot been said about running styles and though not text book, one can still achieve great times, PR and her bobbing head and the 2:15 at London, which still stands today as the world’s best women’s time.
    Unknown to me a the time, i last ran my marathon a decade ago, in 2:38:00 and was not aware of anything but running, over-striding probably, heel-striking probably, novice in the terms of Kinetic Revolution approach to modern day long distance running and training.

  7. Refreshing to have this discussion. I’ve been trying to modify my style for 18 months after a lifetime of running and ive only just managed to shake lower leg problems that have arisen, By reverting to my original style, which is a much slower version of Kimetto!

  8. The forward lean’s really interesting. I’ve recently been trained to use my glutes properly and, I think because of this, I’m consciously pushing my back straighter to better engage them. Am I then doing exactly the wrong thing by doing this?

    I’d be grateful for any advice, I’m a distance runner and no sprinter!

  9. The irony of sports science. Break it down into component parts and analyse. All very well if you look at the right things and their order of relevance. Issue here may be wr time has been in some way attributed to his running style, may be despite his running style, or a mixture of both. Be careful of making broad inference. I’d focus more on v02 max, I doubt efficiency of gait is spectacularly greater than that of his competitors.

  10. He’s not heel striking at all. His foot is accelerating backwards into the ground and he appears to be midfoot striking on both sides. Video is a little fuzzy but what is critical is not when his shoe touches the ground but when his weight is on his shoe. I suspect that with high resolution/60fps video you’d not only see the negative shin angle but also that his center of mass is dang close to over his foot when his foot is fully weighted.

  11. Totally agree with you Kevin. Full foot compression takes place not when the heel of the running shoe touches the ground, but when full compression of the support leg and foot takes place as the body (COM) pushes down on it. I’ve seen videos where people are making a conscious attempt to land their foot under their GCM rather than it being a natural process and it looks very funny. They end up taking small, choppy strides which creates a non-fluent running motion.