New studies on footstrike. Do faster runners heel strike?

Perhaps the most interesting finding in Lieberman’s work is that it may not be so much the barefoot vs. shod but the footstrike that is the important part. Brefoot running allows for the footstrike to happen properly, so they are interconnected.
Let’s leave behind injury prevention for a bit here and focus on speed. Is one footstrike better for speed in distance events? There are several theoretical arguments that point towards yes.

The role of elastic energy storage and return is one such factor. I’ve already mentioned this quiet a bit, but the Achilles tendon and the arch of the foot store a large amount of energy upon footstrike and then that energy is subsequently used upon take off. A forefoot strike has shown that it potentially uses this mechanism much better. One reason is that upon initial contact the foot is in better position to store the energy from the ground strike. In heel running, a great deal of the initial strike energy is lost. On a similar note, it is possible that a forefoot strike utilizes the stretch reflex mechanism better due to the position of the foot upon contact. With a forefoot strike the whole calf complex is in better position to be stretched and subsequently respond than in a heel strike.

Another potential performance enhancement is that it allows for shorter ground contact time while applying the same amount of force. Several different studies have shown the importance of ground contact time in running. In two separate studies, shorter ground contact time was correlated with top running speed and better economy. This shouldn’t be a surprise when you think about it. Ground contact time is going to be a result of mainly the person’s ability to produce force quickly, footstrike, and the ability to use the elastic energy mentioned above. Obviously footstrike plays a role in the latter two.

Let’s look at a couple of studies on footstrike. Several of which have yet to be published.
In the one study that everyone quotes, Hasewage 2007, they looked at footstrike at the 15km mark in a half marathon. The anti-forefoot strike people use this as justification in their decry of changing running mechanics. However, there are several problems with this view. First, let’s look at what the study said. In the study, out of the 283 runners, 74,9% were rearfoot strikers. The rest were midfoot and forefoot strikers. That leads many to conclude that rearfoot strike may be the way to go. However, if we look at a couple of other factors the picture gets a little more cloudy. When you separate out the top 50 instead of using the entire group, those who midfoot or forefoot strike jumps from ~25% to 38%. That significant difference showed that there was a tendency for more mid/forefoot strikers to be faster
Secondly, if we look at ground contact time, there was a linear relationship with the faster runners having less ground contact time and GC increasing as you got slower and slower. Basically, the faster the runner, the lower the GC, this isn’t unexpected. In addition, forefoot/midfoot strikers spent significantly less time on the ground than their heel strike counterparts (183ms vs 199ms). These findings led the researchers to conclude:

“The percentage of RFS increases with the decreasing of the running speed; conversely, the percentage of MFS increases as the running speed increases. A shorter contact time and a higher frequency of inversion at the foot contact might contribute to higher running economy.”

There are still two other factors that no one takes into account. First off, the video was taken at ~9.3mi into a 13.1mi race. In other words it takes place pretty late in the race when fatigue has already set in. Studies have shown that footstrike changes with fatigue. What happens is that more midfoot and forefoot strikers become heelstrikers. Thus, when you look this deep into a race, that potentially skews the percentages.
Secondly, the study was done in a large Japanese road race. While there were several elite Kenyans and other nationalities, of the top 283, the vast majority were Japanese. This is very significant. This means that technically, the results are only generalizable mostly to Japanese runners. Why is this significant? Because of how the Japanese historically train and how they historically run. Due to the heavy emphasis on very high mileage and moving everyone to the longer distances (half marathon and marathon) with neglect to the shorter distance races (1500,5k), the running style of Japanese runners is much different than Americans, Europeans, and even Africans. Similarly, the traditional ideas taught by Japanese coaches at the time favored a running style that was more flat/scoot around type running. Researchers and scientists will probably scoff at this idea, but go watch any video of top Japanese running and you will notice a visual difference. In addition, one of my coaching mentors started spending time going to Japan educating Japanese coaches on running mechanics at around the time of this study. We’ve spent many hours discussing what their views were, what the runners were being taught, and how there athletes were running.

Due to these factors, it’s impossible to take this study and generalize it to anyone except Japanese runners. In addition, the study shows that speed is correlated with both footstrike and ground contact time.

British studies:

In all of the British studies they looked at semi-elite/competitive runners during 800 and 1500m competitions. They looked at foot strike and ground contact time on each lap. This will not only give us an idea on foot strike implications but also on fatigue. The conclusions that can be drawn based on the research about fatigue and training are very interesting!

In the 1500m, the range of times went from 3:45 to 4:22 with the average being 3:56.
Once again, ground contact time-ground contact time was related to foot strike. Forefoot strikers spent 161ms on the ground compared to 169ms for midfoot and 192ms for heel strike. The difference between heel strike and the other two are pretty remarkable. What is interesting is that ground contact increase basically on every lap.
Footstike also changed based on lap. Initially on lap 1, 34.6% were forefoot striking, 46.2% midfoot, and 19.2% heel striking. On lap 4 the picture changed slightly. More of the midfoot strikers in particular had switched to heel striking (heel striking increase to 27%.)
What this means. Fatigue: “over the course of a 1500m race, ground contact time increased irrespective of footstrike position. This implies an element of fatigue, with runners presumably requiring longer to generate the same impulse.”

Before delving into the meaning of this, let’s look at the results of the other study on 800m runners quickly:

800m male runners- 1:47 to 2:01 (avg: 1:55):

- forefoot-35% Ground contact (156ms)
- midfoot-48% Ground contact (161ms)
- Heel-17% Ground Contact (177ms)
-Ground contact lap 1- 156ms lap 2-168ms
-Ground contact time and footstrike related to running speed.

What it all means:

What does this all mean? It’s hard to make a lot of conclusions since the hard data on the british studies could not be evaluated yet. The preliminary observations are very interesting though.

In regards to foot strike, there is a relationship between running speed and footstrike. Do all fast runners forefoot strike? No, but there is a tendency for the faster runners to forefoot/midfoot strike more so than the slower runners.

Is this a function solely of the speed that they are running? Speed of running certainly plays a role in where you strike to an extent but it’s unlikely that it plays as much of a role as people make it out to. Meaning that someone is not going from a straight heel strike while running easy to a forefoot strike while running 800m pace. The Lieberman study provided the first evidence showing that forefoot strikers struck forefoot regardless of condition. Similarly, if we look at the data in the studies above, you can see that the percentage of footstrike types is remarkably similar despite the significant increase in average running speed (from 63-64sec per lap down to 57.5sec per lap). Even though these are different people running each event, if foot strike was solely a function of speed, like many have claimed, then you would expect to see a definite trend away from heel striking as the group got faster.


Perhaps most interesting is what seems to happen during fatigue. Ground contact times increase in both studies, regardless of footstrike. If we look back at what typically impacts ground contact time, it provides some interesting clues. Ground contact changed even when footstrike did not, so we can eliminate that possibility. The other two possibilities are that leg stiffness and use of elastic energy changed, which is entirely possible, but impossible to know. There is some research showing fatigue changes leg stiffness and stretch shortening cycle fatigue. Lastly, the most likely scenario is that fatigue is impacting the body’s ability to produce force in as short a time period.

Lastly, in the 1500m study it was interesting to see a change in footstrike pattern. During the last lap, you saw an increase in heel striking. This would seem surprising as generally one of the faster laps in the race as people try and kick it in. The question is why do runners switch to a heel strike under heavy fatigue? I’m not sure I have the answer. One possibility is that stride length tends to decrease with fatigue and runners are trying to compensate by lengthening their stride, but instead of doing it by pushing off and covering more distance, they simply let their lower leg reach out. Another possibility is that fatigue may impact fine control of the lower leg.

Practical Implications:

What do we learn from all of these studies?

It lends credence to the idea that footstrike is important when we are concerned with speed and speed only. There is a tendency for faster runners to adopt a non-heel strike in a variety of events. In addition, these foot strike types allow for shorter ground contact times, which also correlate well with speed.  One other variable to consider is how that foot strike occured?  It's impossible to know but where the heel strike took place is incredibly important.  There is a big difference between striking close to under your hips to striking way out in front of you.  Perhaps we should consider looking at foot strike in terms of where it occurs in relation to your center of mass, instead of where it occurs on the foot.

Using this data, I’d recommend a switch to a more midfoot or forefoot running style if speed is your main concern.

Secondly, these studies provide some interesting data on fatigue and foot strike. Seeing that ground contact times lengthen, some training should be done to avoid this decrease. I’ve written an entire article (and done a presentation, which I have not posted yet) on a related phenomenon, Strength endurance work, that explains some of the ways to combat this fatigue. We need to train the body to maintain force production (and muscle fiber recruitment) under heavy fatigue. This means start off with being able to increase force production, move to being able to produce force quickly, then move to being able to produce force in heavy fatigued conditions.

Basically, strength endurance work combined with plyometric and power training would seem the best way to train for this type of fatigue resistance. In practical terms:

-Strength Training -> Power training/ Sprint training -> Strength Endurance (Circuits/hills) -> Strength Endurance under fatigued conditions (hard circuits/ 200m reps at 800m pace w/ bounding in between)

Read more about Strength Endurance Here



  1. Very well-researched post! This is an endlessly fascinating topic. As a barefoot ultrarunner, I am very interested in future research on fatigue as a function of footstrike... i.e. is there a biomechanical advantage of a midfoot/forefoot strike when covering very long distances?

  2. Kevin3:04 AM

    you dont have to accept this comment i just thought maybe you should see this

    jay johnson talking about sprint training :)

    i like your twitters by the way

  3. I am not sure I agree with your assessments on forefoot vs. heel strikers. In the What It All Means section, you wrote "there is a tendency for the faster runners to forefoot/midfoot strike more so than the slower runners." When talking about the top 50 runners in the half-marathon in Japan, you stated "midfoot or forefoot strike jumps from ~25% to 38%." While this is a significant increase, the overwhelming majority (62%) of top runners in this case are still heel strikers. This tells me that the tendency of the fastest runners is still to heel strike; to your point, however, forefoot runners as a whole tend to be faster than the masses that heel striker.

    In distance running, I am not a supporter of significantly adjusting gait/stride. I believe gait/stride will fall into line if arm carriage is made more efficient. Not everyone’s locomotion is the same-there is a natural tendency for the lower body to move a certain way. There are opportunities for all runners to make tweaks, but I think forcing a runner to make significant changes to their stride will likely lead to long-term injury.

    I liken this debate to trying to convert a sprinter to a marathoner. Those that excel at sprinting are born with a disproportionate amount of fast-twitch muscle fibers compared to slow-twitch, while elite distance runners have much more slow-twitch muscle fibers compared to fast-twitch fibers. It’s not to say that Michael Johnson couldn’t run a sub 2:10 marathon, however, it’s unlikely that he or any other world-class sprinter can also become a world-class marathoner. In the converse, Paul Tergat is very fast. But I’m fairly certain that even with the best resources, he would not come close to competing with Usain Bolt. An attempt may not lead to injury (other than to pride!) in this scenario, but it shows me that you can’t change who you are at the core.

    While I don’t have any hard data to back it up, I am willing to bet that most forefoot runners are born forefoot runners. Anectodotally, I’ve seen more injuries (failures) than successes in folks trying out shoes like Newtons and Vibrams. This tells me not to mess too much with Mother Nature!

  4. Thanks for the comments.

    We'll have to agree to disagree on this one.

    While the majority of the runners still heel striked in that study, the study has some flaws. First, it was done in the last 1/4 of the race where heel striking will be highest due to fatigue. Second, it was in Japan, whose runners are predisposed to heel striking.

    If you look at the other 4 studies measuring foot strike (granted they are in faster events), all four show predominance of forefoot/midfoot strikes in faster runners.

    We'll have to disagree on changing gait. Arm carriage helps, but it doesn't solve the problems. Not everyones running gait is the same, but that still doesn't mean that there isn't an optimal way for that person to run.

    I don't think your analogy works. Your running gait isn't something that you were born with that can only be changed slightly. I've seen athletes completely change mechanics for the better.

    In addition, our running gait changes drastically as we develop. While there have been no studies on this, more little kids run closer to correct than adults. Go watch kindergarden kids run around, especially barefoot, some will overstride, but most run pretty decently. I used to watch my little sister and her classmates run their P.E. mile every year. Most started out as wholefoot strikers, but sadly aren't any more.

    It amazes me that runners are resistant to changing mechanics. Every other sport, mechanics plays a huge role. I talked about this with Alberto Salazar for a bit and he basically said to forget about all those people who don't believe in changing mechanics, they're missing out.

    You hit the nail on the head with don't mess with mother nature! The problem is we've done that in wearing crappy shoes.

  5. I find at times there's confusion as to the difference between a forefoot and mid-foot strike. How do you differentiate?

  6. In regards to the study, I wonder if midfoot runners would change to heel strike if they weren't wearing shoes that allowed them to get lazy with form? I know that when I'm tired during a race, even my posture can change, but if I focus on maintaining posture, my speed improves and I can finish the race just as well. Be interesting to have all midfoot runners go barefoot. I don't think you'd see heel striking at the end of the race.

    Marathon Mikey, I forced myself to change and couldn't be happier with my new running style. It's been over 2 years since I started, and progress was slow at first, but I've learned much.

  7. Anonymous3:47 PM

    Sorry if I am just completely blind, but I am curious as to the citations for the papers (other than Hasegawa) that you are using as evidence, mainly the British studies as well as the studies that show fatigue increases rear foot strike in runners. Any direction in locating these (i.e. links from your site or authors/titles) would be much appreciated

    Thanks! and great work in putting together something you clearly love.

  8. I think associating a faster foot strike with non-heel striking runners could be misinterpreting or overlooking a possible underlying cause for faster foot striking.

    Upon reading the evidence it seems to me that where a person lands in relation to Center of Gravity (at a given speed) is likely to be just as good if not better predictor of ground contact time than type of foot strike. Thus the place we land in relation to COG is likely to determine the foot stike we use, rather than vice versa. (of course type of shoe, terrain, and other stuff is involved, but a hunch isn't a hunch if you can't ignore some of that!)

    I am making a few assumptions that I don't have answers for, but making a guess it would sense that the farther "back" a runner lands (at a given speed) the shorter their ground contact time is likely to be. I am also assuming that fore and mid foot strikers tend to land farther back than heel foot strikers. Thus (if all assumptions are true) explaining why mid and heel strikers have shorter contact times. Not because they are using a particular technique, but because where they tend to land.

    What would be interesting to know is where runners with shortest and longest ground contact times are landing. Don't look at it as "forefoot strikers are more likely to have shorter ground contact times", but rather a inquiry into where runners, that forefoot strike, are landing based upon their ground contact time. Where do short ground contact forefoot strikers land? Where do long ground contact forefoot strikers land? Repeat for heel and midfoot strikers.

    Of course there are going to be different distances from COM for different runners and where they land based on differences in height/leg length/shoe type etc. But maybe some interesting or useful patterns might emerge that deserve some attention. If nothing else it would be a fun thing to look at even if it turns out to have no practical application.


Related Posts with Thumbnails
Related Posts with Thumbnails