Runblogger marathon foot strike study

Earlier this month scientist, writer and Runblogger Pete Larson and colleagues published a study of runners completing the Manchester City Marathon in Manchester, New Hampshire, USA on 1 November 2009. The focus of the work by Larson and fellow researchers was to analyze the foot-strike patterns of runners passing through the 10km and 32km points during the race.

You can find Pete Larson’s post about the paper here. The study, amongst other things, showed that the running form of marathoners changed between the 10 and 32 km points. No surprise there you might think, we all start to fall apart when we get tired. But this study is interesting and worth writing about for a number of reasons, including:

  • there is not much observational research about that looks at how runners move outside of the lab;
  • this study is mostly concerned with the non-elite regular running community, an area that is crying out for this kind of in the field research;
  • asymmetry (variability between left and right feet) in foot-strike was examined; and
  • it specifically looked at changes in foot-strike patterns that occur as a consequence of fatigue.

When you have this kind of information available it allows you to learn something new, confirm or challenge things you thought you knew and importantly offer up some potential explanations of why certain observable phenomena may or may not be occurring. What I’ve done in this article is look at some of the observations and conclusions made by Larson et al and add some further thoughts for discussion.

Why do runners that start out forefoot/mid-foot convert to heel striking late in the marathon?

The researchers observed that a number of runners converted from forefoot or mid-foot striking to a heel-toe pattern late in the race. The explanation offered by the authors is reasonable and one I’ve read in other studies, which is this phenomenon is caused by plantaflexor fatigue. The calves, soleus and deep compartment muscles get tired from the eccentric loading caused by the load and stretch cycle associated with forefoot and mid-foot running.

I agree this is a factor, but a complementary explanation could be that hip extensor fatigue also plays a part. A key element I have discussed in my own writing and observed in coaching running technique is that recreational runners tend to over use the quadriceps and hip flexors. This can be reasonably explained by the muscles at the rear of the hips and legs (buttocks and hamstrings) not being strong enough or not being activated at the right time during the running gait cycle. Therefore the quads and hip flexors take over to drive the leg ahead of the body to create stride length.This is exactly the way I used to run – see image opposite.

These runners usually don’t have a powerful push off or get much air between strides, they also mostly heel-strike. In this study almost 90% of 800 runners were shown to be heel strikers and many of them might exhibit the muscle activation pattern I’ve described above.

My take on the reason why runners tend to end up heel striking late in the marathon is that they have exhausted all the strength, spring and pop in the buttocks and hamstrings, the stride starts to shorten. Inevitably to maintain pace, the runner starts using the quadriceps and hip flexors more than they did early in the race when they were fresh and bounding happily on their forefoot or mid-foot with plenty of drive from the haunches.

Photo by: Erica Sara Neuman

Did this happen to Tsegay Kebede in the closing miles of his 2:07.12 3rd place in the 2011 NYC marathon? I doubt he’d have looked like this early in the race. Did anyone catch some video of him or photos inside the first 10k to prove me right or wrong? I did read that Kebede was struggling with an injury in the lead up to NYC, so perhaps this played a part in why he looks so different than during this speedier 2.05.20 2nd place finish behind the late Sammy Wanjiru at the 2009 London Marathon.

Not all heel-striking is bad or the same

The authors quite rightly point out that there’s a big difference between heel-strikers and how hard they might hit the ground. They couldn’t measure loading and impact forces in this study because of the Tsegay Kebede Running Form during London Marathon 2009absence of equipment that you might have in a lab or in a controlled study on the track. But you don’t need a lab and a white coat to know some runners will be much heavier heel-strikers than others.

In my opinion this is because of variability in the ability to activate the buttocks and hamstrings around the moment of contact with the ground. So even if a runner like Kebede (an Olympic Bronze medalist) starts heel-striking late in the marathon, he’d still hit the ground far lighter than many other heel-strikers because of his ability to get good drive from the buttocks and hamstrings.

Shoe selection for marathon runners?

Most runners observed in this study were wearing traditional cushioned running shoes. The observations documented in this study indicate that many runners that started the race as forefoot or mid-foot strikers ended up heel-striking. Based on these results, it would be hard to make a case for wearing genuine racing flats in the marathon for the majority of regular runners.

Photo by: Pascal Terjan


Even among elite runners, there is a tendency to go for something with a bit of heel such as the Adizero Adios to perhaps ward off plantaflexor fatigue and offer some forgiveness if late race heel-striking occurs.

However, there must be a reasonable number of incidental heel-strikers included in this study who might otherwise have landed a bit more forward without a big chunky heel getting in the way. For these runners who are marginal or light heel strikers, then perhaps moving toward a marathon racing shoe (not a flat) such as the Adios or similar models might be worth considering. This would also seem a reasonable piece of insurance for the marathon distance even if you’re regular running is done in closer to zero drop shoes.

Unevenness more common that you might think

One of the curiosities of the coaching work I have been doing this year with Mark Gorski is the number of runners that exhibit variation between what the left and right side of their bodies is doing. This study confirms that there is a reasonable number of runners that heel-strike on one side and land forefoot or mid-foot on the other side. I’ll put this down as confirmation of something we’ve observed and thought we knew.

While we coach from a philosophy of looking at the hips down (alignment, activation and strength) we have also noticed that among many runners there is variability in foot-strike patterns. Often this is associated with the aforementioned buttock weakness and/or overuse of the quads and hip flexors. These major variations appear to contribute to running injuries, another area that needs further study amongst recreational runners.


I’m sure many readers of the Larson et al paper might have unanswered questions, or their own theories on what is behind the various observations made. Running is such a complex and variable physical activity that understanding what you can observe is extremely difficult and fiendishly hard to prove. I take my hat off to Pete and colleagues for putting together some fantastic observations and conclusions, many of which could form the basis of future research.

Written by

Other articles you may enjoy:

, , , ,

4 Responses to Runblogger marathon foot strike study

  1. Daniel November 30, 2011 at 11:30 pm #

    How they strike the ground at the end of a marathon is also a function of wich shoe they trained in. What is they trained in zero-drop shoes, like Fred did? He sustained an injury, but that hardly seems surprising… I would like to know your thoughts on this

    • Brian December 1, 2011 at 7:27 am #

      Hi Daniel, thanks for the comment and question. I can’t offer any thoughts on whether or not Fred sustained an injury because of the shoes he was wearing,we’d have to ask him that. In relation to training it makes sense to train in the shoes you intend to run the marathon in and have taken them on a few training runs longer than 32km. If you can do that successfully and hold your technique together without getting sore or injured then there’s a good chance you can wear those shoes in the race and maintain similar form. I’m not saying runners can’t wear flats in the marathon and sustain a forefoot/mid-foot strike, some can, but this study suggests they would be a small minority. The majority of regular runners would need to be more cautious and stick to something that will protect them a bit if they heel-strike from the beginning or later in the race. Brian

  2. FRED BROSSARD November 30, 2011 at 5:52 pm #

    I do agree with your conclusions. I recently ran a marathon wearing Saucony Hattori, which are real zero-drop ultralightweight shoe. During the race I tried to stay focused on midfoot striking. I succeeded for the whole race but encountered pain in the hip flexors starting from the 30 km.

    • Brian November 30, 2011 at 8:57 pm #

      Hi Fred, Thanks for the feedback and very interesting observation about your Marathon performance. The Hattori is definitely a very flat shoe, I had my first run in them last week and found them challenging over about 5km. You must move very well to wear them for the entire Marathon! Brian