Running economy: Forefoot vs Rearfoot striking

It now appears from the most recent data that there is probably no difference in running economy between barefoot or minimalist running and traditional shoe running; and now we have a study showing that there is no difference in running economy between heel and forefoot striking. This study has been around for over a year now and we have been hearing about it at conferences, so its good to see it published in full. Here is the abstract:

Economy and rate of carbohydrate oxidation during running with rearfoot and forefoot strike patterns
Allison H. Gruber, Brian R. Umberger, Barry Braun, and Joseph Hamill
Journal of Applied Physiology May 16, 2013
It continues to be argued that a forefoot (FF) strike pattern during running is more economical than a rearfoot (RF) pattern; however, previous studies using one habitual footstrike group have found no difference in running economy between footstrike patterns. We aimed to conduct a more extensive study by including both habitual RF and FF runners. The purposes of this study were to determine if there were differences in running economy between these groups and if running economy would change when they ran with the alternative footstrike pattern. Nineteen habitual RF and 18 habitual FF runners performed the RF and FF patterns on a treadmill at 3.0, 3.5, and 4.0 m•s-1. Steady-state rates of oxygen consumption (VO2, ml•kg-1•min-1) and carbohydrate contribution to total energy expenditure (%CHO) were determined by indirect calorimetry for each footstrike pattern and speed condition. A mixed-model ANOVA was used to assess the differences in each variable between groups and footstrike patterns (α=0.05). No differences in VO2 or %CHO were detected between groups when running with their habitual footstrike pattern. The RF pattern resulted in lower VO2 and %CHO compared to the FF pattern at the slow and medium speeds in the RF group (P<0.05) but not in the FF group (P>0.05). At the fast speed, a significant pattern main effect indicated that VO2 was greater with the FF pattern than the RF pattern (P<0.05) but %CHO was not different (P>0.05). The results suggest that the FF pattern is not more economical than the RF pattern.

What is unique about this study is that they did not take a group of heel strikers and measure economy and then do an acute intervention and get them to forefoot strike and measure economy. They used two groups: a typical rearfoot striking group and a typical forefoot striking group. They then measured their running economy in their preferred foot strike pattern; and then got them to run in the opposite strike pattern. Heel striking was more economical for both groups!

Amby Burfoot over at Runners World covered this study a few days ago and had some quotes from the lead  author to explain the results. One of them was interesting:

Gruber raises one more point about rearfoot running. Her study shows that it appears to burn less carbohydrate than forefoot running. Since carbohydrate-sparing is the name of the game in long-distance races like the marathon, she believes rearfoot running could lead to improved marathon times.
Bottom-line: “Running with a [rearfoot] pattern might confer benefits in endurance events in both habitual [rearfoot] and [forefoot] runners,” writes Gruber.

Also interesting in Amby Burfoot review of the study was the interpretation of the study in the comments section by those who are not familiar with the interpretation of research. The response in social media to research that conflicts with ones world view is always intriguing to follow. It has the usual trope of fallacies and misunderstandings re things like sample size (which I responded to and there is nothing wrong with the sample size. They would have been happy with the sample size if the results were the opposite!); injuries got a mention (it was not even a study on injuries and we now know that the injury rates between heel and forefoot strikers are probably the same). There is also the usual smattering of comments that the research is irrelevant as they ‘forefoot strike’ and do just fine, therefore everyone should be doing it. Peter Larsen recently did a brilliant summary of the issues around that attitude: Be Careful About Converting Your Experience into a Prescription for All Runners.

As always, I go where the evidence takes me until convinced otherwise.

Gruber AH, Umberger BR, Braun B, & Hamill J (2013). Economy and rate of carbohydrate oxidation during running with rearfoot and forefoot strike patterns. Journal of applied physiology (Bethesda, Md. : 1985) PMID: 23681915

Last updated by .


16 Responses to Running economy: Forefoot vs Rearfoot striking

  1. Hans June 3, 2013 at 7:39 am #


    Interesting study, and results. As someone who has changed from an injury prone knee driving rear foot strike to a less injury prone hamstring pulling forefoot strike and find my running to be both easier, faster and more relaxed it obviously is somewhat suprising but it is what it is. But as a beginning barefoot runner I know it for me barefoot is clearly not more economical, at least not at any point in my one year learning phase. But it is highly interesting, fun and sometimes frustrating, and I am learning a lot about my own movement and running that I would not be aware of if only running with shoes, even with minimal shoes.

    I would anyway love to have videos of the the two groups to see what they are actually doing as one can rear foot or fore foot strike are in many different ways, not just in the strike of the foot, but in everything else that goes on above the foot. But, in this study both groups where more economical while rearfoot striking, regardless of whatever else they did.

    I would also like to see study’s in more realistic real life conditions measuring stuff that is really interesting, not just in these laboratory constrained conditions, measuring theoretically interesting variables. For example how fast one is over 100m, 400 m, 800m, 5k, 10k and so on. In the end, it is how fast you run, injuries and how you experience your running that counts. Of course a cushioned rear foot might still be equal or better for most people.

    Also, I would like to know if there are any studies that says anything about the foot strike relative to the center of mass, how far forward of center of mass the food lands? Theoretically, landing far forward of COM should slow you down, and the placement of the foot relative to COM could be more important to speed and economy than ff or rf strike. A cushioned light rear foot strike could easily be seen as slightly less straining that a ff strike, if the relative forward distance from COM is the same. Any thoughts or knowledge of research on this?

    Also, as a fun, but still relevant from a qualitative viewpoint of research, have you seen this video?
    The barefoot bushman clearly looks more economical and relaxed than the shod sub 3 hour marathon guy. Obviously a subjective interpretation, but all interpretation, even of “objective” data presented in the form of numbers, are still subjective.

    • steve ellery June 5, 2013 at 10:54 pm #

      Hi, this is an interesting study but does not show that either RF or FF is better – it simply shows that that running economy is optimised in habitual foot strike patterns……. I dont think you can then read this study as suggesting that the idea that for the same athlete, and alteration from a FR to FF strike pattern might not result in improved economy once habituation of the technique occurs ….. food for thought …..

  2. Hans June 3, 2013 at 12:59 pm #

    Seems like the video link was not pasted. Here it is,

  3. Mark Richard June 6, 2013 at 12:02 am #

    confirmation bias…

    • Craig Payne June 6, 2013 at 12:09 am #

      You going to have to explain how that study is ‘confirmation bias’?

      I don’t see that at all. The main supervisor of the project is a forefoot striker and has no reason to be biased. Are you alleging that they deliberately set out to confirm a bias in this study? You are on shaky grounds if that is your allegation. I have a high regard for that research group; are you alleging they did something dodgy to fake the data to get the result they got to confirm a bias that they had? What evidence do you have for that faking of the data?

  4. Brent June 6, 2013 at 4:13 pm #

    I agree with Steve Ellery. Very little can be said from these data, especially given the small sample sizes and limited sample pool. What about other confounding factors like posture or cadence? It would be interesting to see this study expanded to include not only more individuals, but also to have a broader age range and to have equal representation among the sexes.

    Are you actually using unbiased sampling of studies that you cite on this blog? It is a little misleading to write “It now appears from the most recent data that there is probably no difference in running economy between barefoot or minimalist running and traditional shoe running” because it implies that there is some definitive answer. This is just one study with a very small sample size. Perhaps you should do a comprehensive meta-analysis because that would be much more informative.

    • Craig Payne June 6, 2013 at 6:03 pm #

      What small sample size are you talking about? I see nothing wrong with the sample size given the power analysis and the statistical significance of the results.

      If the results of this study were the opposite of what was reported, would you still reject it due to the “sample size”?

      I fail to see what posture or cadence have to do with the research question in this study. The researchers had a research question and set out to answer it with an appropriate methodology. The issues of posture and cadence are factors in economy, but are different research questions and not part of the above study. If the results of the study were the opposite of what they found, would you still reject it because posture and cadence were not part of it?

      What biased sampling of studies an I citing? I do not think I have missed reporting and commenting on a single relevant study since I started this blog. When a new study is available, I quote it and comment on it, regardless of the results. Can you please show me which studies I have missed.

      As for a meta-analysis, maybe I will one day – given that the 5 most recent studies that have looked at economy between traditional and minimalist shoes have reported that there are no differences…..

      • Brent June 7, 2013 at 5:00 pm #

        12 males and 7 females were in the heel strike group and 14 males and 4 females were in the forefoot/midfoot strike group. The main finding of this paper is a negative result (i.e. they found no clear difference between the two groups in running economy). A logical question when analyzing a lack of pattern in any scientific data is: Was the sample size big enough to allow you to see a pattern despite the degree of variance?

        If there was substantial variation among the participants in posture and/or cadence, then that could have obfuscated the results. Differences between the experimental groups may have been apparent if they had accounted for posture or cadence. Or, it may not have made a difference. But without properly controlling for these things how do we know?

        Is it better to be barefoot or shod? The best way to answer that question in an experimental framework would be to split up your experimental subjects at birth, while also controlling for other factors like sex, race, height, socio-economic status, etc., and then see who grew to be the best adult runners. Obviously we are never going to see a study like that because it’s not feasible.

  5. Michigan Biomech June 6, 2013 at 9:44 pm #

    I agree about the sample size. It is not an issue except for the fan boys who do not like the results of the study. They are also happy to accept studies that have smaller sample sizes if they like the results of a study.

    As for biases. Every study that I have seen come out this year an running economy, injury rates and biomechanical ‘load’ are showing that there are no advantages of minimal running or forefoot striking over heel striking. That is not a bias. That is what all the recent research is showing.

  6. Will Musto June 9, 2013 at 5:31 am #

    It makes sense that heel-striking, especially over long distances, might be more economical as–over the course of a marathon–you’re not exactly moving at high-end speeds.

    Forefoot striking over shorter distances–and higher speeds–would have to be most economical. No??

    • Craig Payne June 9, 2013 at 7:45 am #

      Its complex and very subject specific; which is why the results are mixed in some of the studies. I not sure how much to write here or whether to write a whole new article!

      But the simplest way of looking at it is that traditional shod running carries a heavier shoe that will increase the effort to run. Running barefoot or minimalist does away with that weight, but you need to be up on the midfoot/forefoot, which requires more muscular effort that will also increase the effort to run. So its a trade-off between weight of the shoe vs effort of muscles, which is possibly why all the most recent studies are showing no differences between them in terms of economy. In the previous studies some showed barefoot or minimalism was more economical and others showed it wasn’t. If you look closely at all the data in each study, there is a significant amount of variation between individuals – so which one is more economical for one runner is not going to be more economical for another….it is subject specific.

      This would also explain why the study that looked at Pose running showed Pose was less economical (despite what the religious fanatics like to claim about the study) – the participants wore the same shoes – so they had the same weight on the feet in both groups, but the Pose group who transitioned to midfoot/forefoot had the increased muscle activity to get up on the forefoot or midfoot, so it is quite plausible that Pose running is less economical (which is what the evidence showed).

      Your point of being up on the toes to run faster is probably a biomechanical one to do with better lever arms etc allowing them to run faster; but is theoretically less economical over a certain distance due to the increased muscular effort to maintain it – what that distance is will vary substantially from person to person.

      As to which one is better for which individual runner over which distance is most likely going to be a lever arm issue as well. A simplistic eg would be an person with a shorter calcaneus –> the calf muscles in that person are going to have to work substantially harder to forefoot strike, making forefoot striking a very uneconomical option for them, compared to someone with a longer calcaneus and hence a better lever arm with less muscular effort to midfoot or forefoot strike. All the joints in the foot have a massive variation from person to person in the position of their axes of motion, so all tendons/muscles will have an almost infinite possibility of variations in lever arms and how hard or not the muscle has to work to achieve a certain task.

      Does that make sense?

      We only scratching the surface in understanding this from a ‘thought experiment’ perspective, let alone evidence based. To tout one as being globally more economical than other is just not supported by the evidence. What is good about the above study is that it adds more into the mix that the ‘thought experiments’ or ‘modelling’ need to be consistent with…. something the fanboi‘s are not happy with the direction the evidence is taking!

  7. Drew June 17, 2013 at 7:36 am #

    Wow, so much info! Trying not to drown in it!
    A couple of things occur to me, firstly, does the treadmill make a difference? Would the assistance provided improve the economy of the heel striker? Am I correct in saying that the forefoot striker has less time in contact with the ground, therefore would receive less assistance from the action of the treadmill? Potentially making the results look closer than they actually are?
    Secondly, my thoughts on minimal shoes and forefoot striking are formed from watching my kids and their friends run barefoot in the backyard – no heel strikers at all, so why and when do we change?
    Thirdly, I believe Kenyans are (mostly?) forefoot strikers yet the the report says that marathon times could be improved with heel striking? This, I’ve got to see!
    Cheers, thanks for the opportunity!

    • Craig Payne June 17, 2013 at 7:47 am #

      I not to sure how economy/physiological studies like this can be done not on a treadmill. I do know someone is writing a blog post elsewhere looking at the differences between the two in detail … if they don’t hurry up and publish it, I might look at it in more detail (HINT: Ian, get on with it!). I just got access to an undergraduate Honours thesis on the ‘Kinematic Analysis of Hip and Knee Joints between Barefoot and Shod Treadmill Running’ – I will blog in that study soon.

  8. Christian August 4, 2013 at 10:59 am #

    I love that there is such criticism, yet often those throwing the stones aren’t willing to get off their ass and conduct the research themselves. If you are going to criticise the discount for extending variables you could be here all day, but that is the whole point of matching an appropriate methodology with the hypothesis decided on. The whole point of the study was to explore variance in economy between habitual runners of different foot strike patterns. In specific regards to this hypothesis the methodology was sound. As for posture variances, I am not sure how on earth you would account for this in the context that some of you have suggested. The variability of posture within “posture types” itself is far reaching and in accordance with your own criticism of this study would be impossible to control (therefore you contradict yourself and display a poor understanding of research development).

    Take research for what it is, match it with clincial experience and ALWAYS remain open minded to the possibility that was truths you know now may be of less substance in the future. Change is the only constant, all you can do is excel in the immediate climate of research for any given moment.

  9. Geoff Hancock April 7, 2014 at 5:40 pm #

    @Craig Payne,

    Did you even read the abstract that you quoted in the article?

    “The RF pattern resulted in lower VO2 and %CHO compared to the FF pattern at the slow and medium speeds in the RF group (P0.05)”

    Then you say:

    “Heel striking was more economical for both groups!”

    Just because you get excited about it, doesn’t mean it is correct. Next time check your bias at the door, sir.

    • Craig Payne April 8, 2014 at 1:35 am #

      ….and what biases might they be? I only quoted what the authors said and found based on the data in the study. I assume that you have not read the full paper.

Leave a Reply to Drew Click here to cancel reply.