Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Tuesday, April 01, 2008

Running technique: The Footstrike

Heel vs. Midfoot vs. Forefoot: How do elite runners land?

I've finally gotten around to this post, which is probably two weeks in the making, and it follows on from our recent series on running shoes. That series began by looking at whether shoes are in fact as much a cause of injury as a cure, and then evolved into a discussion of how the running market is evolving. Twenty years ago, it was all about motion-control shoes preventing overpronation to prevent injury. Today, it's all about running "barefoot in your shoes", as companies try to go back to "natural" without selling you "the Emperor's clothes", in effect! (It's quite a long post, my apologies, but a lot of important information to get through...)

The next logical question is to ask how is the foot supposed to land during running? This question evolves out of the discussion of shoes. vs barefoot running, and is often at the heart of discussions on running technique. Very often, debates of "technique" tend to start from the feet, jump to the knees ("lift your knees") and then skip to the arms, and that's about it! We won't go into too much detail on technique today, focusing instead on only one of many aspects - the landing of the foot, and particularly, whether the elite runners tend to land on the heel, the midfoot, or the forefoot.

Elite runners footstrike patterns

Perhaps surprisingly, there are very few studies looking at elite runners and footstrike patterns during actual races. Despite this, until recently, the overwhelming majority of coaches and experts were advocating that heel-striking was the most effective technique, simply because most athletes did it. That claim will come up again, but the perception that it was most effective has, over the last few years, been changing. And with the advent of Pose and other running techniques, as well as the observation that not all elite runners are landing on the heel first, people have now begun to advocate that forefoot landing is better! So we have this 180 degree shift, often in the absence of any substantial data to support the claim.

I am sure that many will have seen this kind of assertion (this one is from Wikipedia):

Leaning forward places a runner's centre of mass on the front part of the foot, which avoids landing on the heel and facilitates the use of the spring mechanism of the foot. In other words, landing on the heel is bad, to be avoided...
Or there is this, from Gordon Pirie (admittedly somewhat older):
"Running equals springing through the air, landing elastically on the forefoot with a flexed knee..."
But what is "better"? Where science has yet to catch up with opinion

It's important at this point to ask the very pertinent, but infrequently asked question: "What does 'better' mean?". In other words, when people are advocating that it's 'better' to land on the forefoot, what do they mean? Is it faster? More efficient? Less injury-prone? The fact is, the word "better" is used without studies specifically looking at any single one of these aspects. And the 'prudence concept', as applied to science, says that you cannot say something is "better" unless it's been studied and compared to the alternatives. Unfortunately, the science lags behind in this regard.

So for example, above you have the quote that you are supposed to land elastically on the forefoot. That implies performance and efficiency, which might be true for short exercise, lasting a minute or two. But in an event like the marathon, are we sure it remains the "better" option? If you went out and ran 2 hours today, landing on your forefoot instead of landing as you've always done, what would be the likely outcome? Chances are, you'd be hurting for a few days, with calf muscles that you had perhaps forgotten you had! Worst case scenario, you'd be injured for months with an Achilles tendon injury. That is certainly not a desirable outcome. So there are problems with making sweeping statements about landing patterns.

But more than this, these kinds of statements are never grounded in proof. So for example, when it's written that you land "elastically", has anyone ever done the study of elastic energy return in different types of running? They haven't, but there is theory about it, and that's where these recommendations come from. So the approach in the discussion that follows is for me to adopt the role of "questioner", playing Devil's Advocate, with the humble admission that science simply does not know the right answer, only the possibilities...

Looking at one particular study - elite 21 km runners

So in the current climate where real evidence is scarce and opinions hold sway, let's take a look at one study that has examined footstrike patterns during running events. It was done in 2004 in Japan, and published in 2007 in the Journal of Strength of Conditioning (not sure of the reason for the delay - it happens sometimes in science!). The full reference, for those interested, is Hasegawa et al., J Strength & Cond., 2007, (21), 888-893

It was performed at the 2004 Sapporro International Half Marathon in Japan. The scientists set up a high speed camera (very important for accurate collection of information - beats YouTube science any day!) at the 15km mark of the race, and captured most of the runners coming through. In total, they were able to observe the foot strike of 248 men and 35 women, and characterize them as either heel-strikers, mid-foot or forefoot strikers. They also measured Ground Contact Time at the 15km point.

The finding - what do you expect?

Before giving their main finding away, take a moment to guess what they would have found...If you are anything like me, and have read the substantial amount on the internet and in books about how it's "better" (there's that word again) to land on your forefoot, then of course, your expectation might be that they found:
  • The majority of runners land on the forefoot
  • Those that DO NOT land on the forefoot are the runners who finish towards the back of the field
Well, if that's what you thought, you'd be completely incorrect...! Because the finding is the following:
  • The vast majority (75%) of the elite runners land on the heel
  • About 1 in four (24%) runners landed on the mid-foot
  • Only 4 out of 283 runners landed on the forefoot
Those runners that landed on the forefoot did not finish in the first four positions, so the common argument (a flawed one) that the best athletes are forefoot strikers is not supported by this finding.

Possible conclusions - how you read the study is influenced by what you wish to prove...

So, given this, one is tempted to say that the landing of the foot makes no difference to overall performance. Of course, this is not necessarily true. As I wrote above, science is often taken out of context, and this is one such example. You cannot, for example, rule out the possibility that these heel-strikers might be a few seconds or minutes faster if they just learned to land on their forefoot! Personally, I think that's highly unlikely, and what is more likely is that they'll end up in rehab for Achilles injuries, but even that is a "bald assertion", based only on opinion!

Now, however, here is where it gets interesting, and this is where the forefoot advocates got quite excited. When the researchers divided the finishers into groups of 50, they started to see something of a change in mid-foot landing as you moved further down the list. In otherwords, there was a higher percentage of midfoot strikers in the first 50 runners than in the second, and then third, and so on. The graph below shows this for heel-strikers and mid-foot strikers (I haven't shown forefoot, because it's so tiny and insignificant by comparison):

At first glance, the conclusion from this graph is that if you want to be a faster runner, finishing higher up in the overall order, then you should be a midfoot striker, not a heel-striker. That's how many people interpreted the finding. And this may well be true. Unfortunately, there is another possible reason it looks like it does - perhaps it's simply a function of running faster.

Speed and footstrike

In otherwords, you naturally shift your contact point with the ground further forward when you run faster. The average speed, incidentally, of the first 50 runners was 3 minutes 3 seconds per kilometer. The second group of 50 runners averaged 3 minutes 10 seconds per kilometer. Hardly a big difference, but given the range (the 50th runner is at least a minute behind the 1st runner), is it possible that groups of 50 is too big, and that all this "finding" represents is a speed effect on footstrike?

The point is, this study does not allow you to differentiate between three possibilities:
  1. Faster runners are midfoot strikers (could be co-incidence or some other cause); or
  2. Midfoot strikers are faster runners (and therefore we should all change our running style and land on the front part of the foot more); or
  3. All runners would eventually be midfoot strikers, if they just ran fast enough!
This is another classic example of how a scientific result can be taken out of context and applied to give advice that may not be 100% correct.

Personal opinion and implications of this study

My personal reaction to this research, when it came out, was that it disproved the popular theory that all runners should be aiming to become midfoot or forefoot strikers. Most of us (well, I'm in this group, apologies if you are not) are nowhere near the elite level, and we're often told by experts and coaches that the elite are landing on the ball of the foot or the midfoot, and so we should too.

But the next time you think of running like Gebrselassie and trying to land mid or forefoot, consider this: if you go out and sprint 100m, you're likely to run on your toes the whole way - because you're running faster, you land more on the mid-foot, or even the forefoot.

Sprinting as you are, you'll probably cover 100m in 14 seconds, which puts you only 1 second ahead of a Bekele or a Gebrselassie in a 5000m race, so is it any wonder they are midfoot strikers on the track - they're running as fast as most of us sprint? The point I'm trying to make is, if you ran the speed they did, you'd be a mid-foot striker too! But just as I suspect they change as they slow down, we all do. So why, and on what basis, should you try to run with the same foot strike when you are running perhaps 3 minutes per kilometer SLOWER than them? Again, these are relatively bald assertions, but hopefully you recognize the implication of speed on foot strike.

So when you go out and run a 3 hour, or a 4 hour marathon, that's another story altogether. And what the Pose running study at UCT showed me a few years ago is that if you change the landing of the foot, you predispose the athlete to injury - that study took a group of runners and within two weeks had them all running on the midfoot (please don't write in to say that Pose doesn't mean midfoot, because Romanov was the coach and he was happy with their technique!). Two weeks later, they all broke down with Achilles tendon injuries!

Why? Because sitting where you are right now, if I was to walk into your office or your home and take you outside and ask you to please run landing on your forefoot or midfoot, I can pretty much guarantee that the way you would achieve this is to point your toe down...you're probably doing this as you read this - contract the calf, and point your toe away from your body, like in ballet. Now imagine your body weight landing on that contracted calf muscle 85 times a minute for 4 hours. That, simply put, is a recipe for disaster.

However, if you can gradually change your landing, then I do believe that you can shift your footstrike. But it's a gradual process. And more important, what is the point? There is no evidence that heel-strikers are injured more, no evidence that mid-foot runners are faster and perform better than heel-strikers, and so the ultimate question is:

Why would you want to change your foot landing to begin with? Science has little to offer you in support of this. And so my advice, having read this far (well done!), is to forget about the possibility that you're landing "wrongly", and just let your feet land where, and how they land, and worry about all the other things you can when you run!

If there is one thing you change in your running, don't focus on your footstrike, but rather on WHERE your feet land relative to your body. Because if you are over-reaching and throwing your foot out in front of you, that's a problem, but what happens when the rubber meets the road is less relevant!

I'm sure there's more to this topic, based on your questions and comments. As usual, fire away! And remember the humble admission from earlier - science, believe it or not, does not know the answer definitively! (just as we can't tell you why Bekele is so dominant in World X-Country!)


Late edit to the post - the comments section of the post has been closed (29 Aug 2009) - too many comments flying in to respond (on top of the other 20 posts still getting comments! And I don't want to ignore people!)

I would recommend that you click on "Running technique" in the tabs at the top of the page, and read through those posts to get the full picture and context to this post


Anonymous said...

Is there a method to measure the respective force on the foot/knee/hip generated by a heel strike or mid-foot strike? If so, wouldn't that help answer the question whether heel strike or mid-foot landing is "better"?
Let's exaggerate and create an experiment just for the sake of argument: suppose someone jumps from a 3 meter platform and lands on the heel of one foot with a rigid but not locked knee. I would think that would generate more force (because of the rigid knee and the smaller area)than if someone jumped from that same 3 meter platform and landed on the mid-foot with a flexed knee. No pointy toes--just land.
Is running not just a series of continuous jumps?
I am a long-time runner--33 years--and have evolved from a heel-striker to mid-foot. This is consistent, I believe, with the genetic heritage of homo sapiens as long-distance, barefoot, hunters.
Also, I am a recent convert to SOS and a big fan. Your comments would be most appreciated. Thanks.
Mike McGrath

steve said...

I'm a little biased, as I am someone who changed his footstrike from a heel strike to a midfoot strike, but I'm not sure you can make conclusions from just one study.

Though not scientific, a couple years ago, myself and two other coaches watched all the 1500m races from the Olympics and World Championships from the 1960 games to present. We were paying close attention to running mechanics and footstrike and slowed the video down to do a quick visual analysis on the top runners. The vast majority of the runners landed with their foot underneath them or close to underneath them landing somewhere on the midfoot/forefoot. Of course with such raw video it's impossible to tell where exactly they landed, but the point is you could clearly differentiate from the runners who landed heel first reaching out, with those who landed somewhere not on their heel. I wish I had the numbers still, but there were very few runners who landed heel first.

Secondly, there has been one other study done on this subject, but it was a long time ago. In a paper on running mechanics by famed sprint coach Tom Tellez, he cites and then quotes a study done by Toni Nett titled "Foot Plant in Running" that was published in both Track Technique and orginally in "Die Lehre der Lechdathletik." Nett tested runners from 100m to marathon and found that the majority of runners did the following:
"At that fastest pace (100-200m) the ground is contacted on the outside edge of the sole, high on the ball-joint of the little toe.

400m runner's contact point lies a bit farther back toward the heel. It appears to be somewhat flatter than the 100-200m sprins.

In the 800m, the foot is planted throughout most of the race, on the part of the outer edge of the sole within the metatarsal arch. The position of the foot seems almost flat on the outside edge.

At all distances beyond 1500m, first contact with the ground was with the outside at the arch between the heel and metatarsal."

Take that for what it's worth.

Ross Tucker and Jonathan Dugas said...

Hi mike and steve

Thanks for your input. To address each separately:

Mike, thanks for the positive words and compliments, as always, they are much appreciated!

With regards to the force on the knee, there certainly are ways to measure it - you need some pretty specialised equipment, but it's done fairly regularly, and it's possible to work out the forces,the power and the work done by the various joints.

In fact, it was this kind of measurement that was done in the Pose study a few years back. What was found then was that during Pose (let's oversimplify and call it fore/midfoot running), the work on the knee was reduced compared to heel striking. However, the work on the ankle was increased. The hip was not affected. THis is typical of studies where the forefoot is compared to heel strike run - knee is reduced, ankle is increased.

The implication of this is just what was mentioned in this post - people were predisposed to developing calf muscle and achilles tendon injuries.

In terms of comparing the impact when jumping off a 3 m platform, the one big issue with running is that you are not coming down vertically - there is a substantial vector in the direction of movement, which means it's not entirely a case of loading the body weight. And you never really have a rigid knee, even as a heel striker, but point taken.

Again, I think that if you take a heel striker and try to change them overnight, you're asking for trouble. Many elite and non elite runners have run for years as heel strikers, no injury problems, and as this study shows, no problem running fast (3 min/km for most). So I'm not sure that we're better off trying to fix what does not seem to be broken. Of course, one could always argue that they'd be even better as midfoot strikers, that's the debate,I guess...!

and then Steve, thanks for your comments and additions to the argument.

In my defence, I will say that my conclusions are most definitely not based on just one study, but on the collective evidence from this, and other studies and experiences, many of which have been discussed in the series on Pose running done previously (see tabs on the top of the page for links - under Series) and on the running shoe series.

You're quite right, this study doesn't allow a conclusion to be made, though I was at pains to "hedge my bets" when I admitted the science did not have the answers!

With regards to your analysis of the data from the Olympics, sounds very interesting. I would suggest, again, that a big factor determining the preponderance of fore- and midfoot strikers here is the speed - 56 seconds per lap will do that! But I most definitely agree with you on the "reacing out issue". What I try to instill in all athletes I speak to, and its based on sound theory, as I'm sure you know, is that you must keep the landing point beneath you so as not to brake. However, it's still possible to be heel in this position, and some athletes, I dare say, would really struggle if you forced them to change this style.

Finally, thanks for mentioning that study - I had not ever heard of it, but it certainly does highlight the impact of speed on footstrike. What needs to be asked now is what happens in the marathon? Are we realistic in expecting the forefoot/midfoot landing?

Both you and Mike are cases of "evolution" changing footstrike, and I'm pleased you wrote in, because you demonstrate that it can be done. And I certainly don't dispute that, but I do think it important that people don't just go blindly into it. My attitude is to leave it if it is not broken, and to look for gradual, incremental change if it is. But to me, training, training, training, is always the key to injury, not technique!

Thanks for the input!

Mike T Nelson said...

Hi there! Cool post and excellent job on the site! I just linked it to mine.

Perhaps this all gets even more muddy when most runners (or athletes in general) that I see have very immobile feet/ankles which leads to a whole host of other problems up the kinetic chain and muscular "shut down" at the hips for starters (via the arthrokinetic reflex)

Keep up the great work!
Rock on
Mike N

Andrew said...

Speed is definitely a factor. On a treadmill next to a wall mirror I once compared my form when running at paces between 5:30 and 10:00 per mile. It was impossible for me to heel strike at fast paces, but very difficult to forefoot strike at slow speed. In fact when I tried the latter it looked like I was jumping up and down.

I also notice the difference in muscle use. After long, slow runs my quads are the most sore; after short races or speedwork my hamstrings are the most sore. That makes it pretty clear who's handling "impact" duty - at faster speeds I'm naturally generating increased horizontal force relative to constant speed-independent vertical force due to gravity.

I'd be interested to see some discussion addressing differences in physiology. Among humans there is surprising genetic variation in bone length, muscle mass, and their relative proportions (e.g. some have shorter tibias relative to the femur or relative to overall height; some have giant quads and small calves or vice versa, etc.) Not to mention simply comparing bigger runners to smaller ones. At 193cm/88kg my large frame results in long strides with more mass behind them compared to a 165cm/55kg runner; maybe my forefoot and/or achilles can't absorb that kind of inertial load even if it were more aerobically efficient to do so.

Thanks again for another informative article!

Anonymous said...

Firstly again congrats for taking on this subject... one of the most heated in running forums, despite what video evidence says.
My favourite btw is 2002 London, simply because of the large front pack together for about 30+ km. Observations from that shows one runner who you could say (at the frame speed) was slightly forefoot, other than that the others were midfoot/heel... this agrees with both what you and commenter Steve says.

In terms of the full gait swing, if you freeze frame (yes I have wasted much time to prove my point band learn), you will find , in common, high backlift with low knee lift, and thus lower leg swinging freely forward to land - this is important- not in front of body, but directly under (the lower leg slightly back from verticle). This is hard for most objectors to accept, since their argument against is based on 'overstriding' and 'braking' force.
My belief is (I have seen some studies too) that such landing in turn is the cause of high backlift through released energy and thus the pattern continues.

In fact, most 'easy striding' runners already do this.

Now most shoes show wear on the heel.. thats where we put our 'Shoe Goo' right, and that is because the landing is midfoot/heel- a slight touch of heel or even a slight shift backwards on footplant.

btw, are you sure the pace was 5:03-5:10/km and not PER MILE. Seems to slow if it was the Sapporo International.


Unknown said...

I love your technical posts most of all. But I'm always afraid of reading anything about technique because I'm just sure that mine will suffer by just thinking about it.

Keep up the good work guys,


cassio598 said...

Firstly, I would say the essential element in this debate is reall whether one should reach out during ground contact and overstride or land with the foot underneath the body. The issue of which part of the foot one should use for the later is just a proxy.

Secondly, kudos to andrew but shame on Ross. If you'll remember your intro physics course, vertical and horizontal motion are completely independent of each other. When measuring the force during the landing phase, you are indeed measuring the body weight's acceleration all by itself, PLUS any additional force caused by the brief change in forward momentum.

Rahul Verghese said...

Excellent post.
Wonder if there is any study done to see how elite mid foot strikers run when they run 2 mins/mile slower than their marathon race pace, and how they run when they sprint.
So IF it is a Heel, midfoot and forefoot thing for three different speeds for elites, does it work the same for the rest of the crowd - maybe it does.
Would be great if a University or a New Balance or Nike etc took up some of these studies on a more comparative basis so that learnings can be obtained and then transfered across to the vast majority of the population.

Ross Tucker and Jonathan Dugas said...

Hello everyone

THanks for the great comments and discussion. I'll try to address each one as much as possible in one reply, hope I miss out as little as possible;

Mike, thanks for the link and the positive feedback. I think your point on runners having differences in mobility of the feet and ankles is a pointer towards why sudden, "textbook" driven changes in technique don't work too well. If a runner has a certain "style" (for want of a better word), making a sudden change to it ignores the possibility that the style has evolved out of necessity, perhaps due to these individual differences. It is thus a symptom, not the cause of any problem, which is why I would again argue that one should not make sudden sweeping changes. Slow progress if any, is the way to do it.

Andrew, I honestly don't know of any work looking at anatomical differences like bone length and ratios and stride mechanics. Presumably there are some, I will have a look to see, but I've never come across any. That would be one likely explanation, however, for why when you take a group of people and compare running styles, you get such large variations.

I'll respond to the other comments in a separate reply, else it gets too long!


Ross Tucker and Jonathan Dugas said...

The next batch of replies:

Oliver, you're quite right - the foot should land beneath the body, which means that the old adage of "drive your knees forward" (given by most coaches I've ever heard at school especially) is not quite correct. Having said that, the knee lift of the elites is still relatively high, particularly for the track runner, but it is mostly passive. You can quite easily tell when a runner is "reaching out" in front of them, and I certainly agree that this is counter-productive. I would say that if one wants to work on one aspect of technique, it is this one - don't worry too much about how the foot lands, but rather about where it is in relation to your body.

And as for the pace, you're right - my mistake. The actual pace was 3 minutes/km, I corrected it in the post, but unfortunately the email had gone out by then. So if you read the article on the site, you'll see the right values - a simple, but significant typo on my part! Sorry...

Cassio, same as for Oliver - it's the reaching that is so important, but not so much the landing. if 75% of the elite runners in a field going at 62to 63 minutes for a 21km race are able to land on the heel, then it's difficult to say that it's inefficient, though as I wrote, one can't conclude that it isn't.

Second point, I have to defend myself against the "shame on Ross" call!!! If you read the post, I'm not trying to link the vertical to horizontal at all. Let me put it this way: If you were caught in a burning building and had to jump out of the first floor window, which would you do:
1) Jump straight down so that you land vertically, or
2) Take something of a run so that you land and then fall forward, some horizontal momentum carrying you immediately past the point of impact?

I think most people know that if you're going to do this, you make sure you don't land straight down, you get some horizontal momentum.

So when I responded to the question my Mike at the top, his point was to jump off a 3m high platform, and my point was in running, your foot hits the ground not directly, but while your body weight moves over it. It's not a static system, which means the loading is not the same. So yes, you're taking gravity times mass as the force, but it's quite different from coming straight down, because of that horizontal vector.

Then finally to Rahul, that would be one interesting study to do. The other is to take the non-elite and test them at a range of different speeds, which would show you the effect of speed in the same person - we've compared across the speeds, but in different people, never the same.

As for the research by the shoe companies themselves, i suspect that much of it may exist, but it never reaches the public domain because it's seen as a competitive advantage. For example, I know that adidas do substantial testing in quite reputable universities. The problem is that the work never gets peer-reviewed and published, and instead we are asked to believe the conclusions drawn from it. Perhaps there could be a 6-year waiting period, after which work gets published, it would be good to see.

Thanks again!

Unknown said...

Brilliant post!

I've been trying to change my striking position over the past few months, constantly feeling as if there was no real evidence or concrete reason for doing so.

Ross Tucker and Jonathan Dugas said...

HI Andre

Thank you! My advice? If you want to change something in your running, try to focus on your hips, and your position when running, especially downhill. If you can get your hips a little further forward, you'll notice the effect is to bring your landing position more underneath your body, and that is a desirable change, for which there is at least theory supporting it. What is important is that you don't actively change the landing pattern by contracting your calf to point your toe - that leads to injury, I believe. So let it happen passively, if at all.

Good luck

Anonymous said...

Fascinating post, I really enjoyed it, as usual. I had a run on Monday night, which confirms what Ross writes. During our club run, we sometimes run the "Measured kay." It's a km stretch, that some runners decide to run at full speed, just to see what you can do. It's just a little fun, and in my infinite wisdom, I attempted to run it under 3 minutes.
So I head off, at a fast and free stride, but very conscious of running normally, which for me is heel striking. At 500m I glanced at my watch, 1:35, slightly behind the pace, so I increase it, now I was going under 3:00 min/km pace. I immediately noticed I started running on my toes, automatically. Quite a strange feeling, when I was conscious of it.
Just my own little experiment.

Anonymous said...

Just a little n=1 insight... I switched from heel to mid-foot about 8-9 years ago. Yes, definite achilles soreness at first... I'd say it takes 1-2 years for a full adaptation. My main pros: increased cadence, can run downhill faster, no more knee soreness ever. Cons: calf soreness until fully adapted, most run shoes suck. Thanks for exploring this topic more!

Anonymous said...

Hi Guys:

Very interesting stuff, and good job on keeping the discussion science-based!
Like many of the other bloggers (particularly "Mike"), I am a mid-foot/forefoot convert, having been taught to strike with my heel, when I was younger. From personal experience, I have found midfoot/forefoot running to be a more efficient way of running, and I have been injury-free since (5-6 years now).
One thing that keeps me "pro-forefoot" is simply looking the way animals run in the animal kingdom. They use their legs like springs and are perhaps the ultimate example of efficient movement. Our calf muscles and tendons perform plantarflexion and have been shown to function as springs, suggesting that we ought to make the best use of their potential. I know R. Macneill Alexander has written much about this, and I wonder if looking at how animals move might offer some more concrete evidence (if not a good argument) in favor of using the spring-like, propulsive muscles of our triceps surae.

Lastly, it seems obvious to me that a group of heel-striking runners in a study would become injured from being asked suddenly to run on their forefeet. Like anything there is a necessary adaptation period, and arguably, a good deal of "unlearning" which needs to first be done.

Thanks always, Guys! Good job and keep on keepin' on!


Champion0607 said...

I have been the subject of numerous comments on my "long stride" over the years (6'1", 34" inseam). A couple years ago I started doing "strides" in an attempt to decrease stride length, move my landing point more under my body and increase cadence. Originally, my cadence was in the low to mid 150s. Never did I even think about where I was landing on my foot, just tried to get it under me. My main points of concentration for the strides were cadence, foot landing point, and vertical "smoothness" (no more bounding). Anyway, after 2 years I've increased my natural cadence into the mid 160s and I feel like I do a lot less "bounding". My heart rate seems to have come down at the same aerobic pace also (8:00/mi).

So my question is why would landing on your forefoot provide a better "spring" than landing slightly on your heel and using the running shoe sole as the spring? Since springs tend to wear out, I would much rather replace my running shoes than the tendons/muscles in my foot.

Another note. I remember my Dad training for marathons 25 years ago. The emphasis back then was "running on your toes", and I can still distinctly remember how huge his calves were. If your calves have to work that hard, you'd think they would be using a decent amount of oxygen/calories. I would think you'd be more efficient using bones/tendons to support the vertical component, and letting your leg muscles concentrate on the forward component. Just a thought.

Anonymous said...

Thank you. Thank you. Thank you. I've had debates about pose vs. heal striking. As opposed to forcing anyone to change their technique, if it's working, stay with it.

If you're getting injured a lot or having similar problems, you can always try to make adjustments.

One size definitely does not fit all.

surfchicky23 said...

Hi there,

I find that shoes that simulate barefoot are great for short distance running. They tend to be uncomfortable in long distance as they dont have as much support/cushioning.

Do you think its possible or even practical to design barefoot simulated shoes to have more cushioning?

Thanks, Sarah.

JM said...

Interesting read. Got the mail from a friend of mine and just now subscribed to the mailing list.

This whole debate has been fascinating, to say the least. I am relatively new to running, only taking up the sport about 1 year ago. I have, for the fun of it, played around with various running techniques. I just cannot run heel first and really struggle to run forefoot as well. My natural gait seems to be midfoot, supported by a pressure plate testing done when I fitted my first shoes. When I run very slowly, midfoot strike. When I sprint, also a midfoot strike, but leaning more to the forefoot.

For me personally I also then find the midfoot strike to be the most economical.

Noel said...

I don't think your complaint against mid-foot striking is valid. Of course you'll get an overuse injury if you change gait and then run for long distances without any build-up. You'd also get an injury is you changed from a mid-foot strike to a heel strike and did the same. I believe the Pose training emphasises lots of prehab work to avoid this issue.

Another confounding factor: I expect pro runners have spent a lot of time working on their gait, probably under the assumption that a heel strike is optimal. So just looking at what the pros do is not sufficient to (dis)prove the hypothesis either way. Sure, it is a useful piece of research but much more is needed before this issue can be resolved.

Anonymous said...

Have not got time to read all comments so sorry if already been said. If a runner puts a shoe under his foot (i.e. wears a shoe) when he/she attempts to land around about the 5th metatarsal (which is where they would land if barfoot) then of course the heel section of the shoe will hit first because it is tapered that way!

Ross Tucker and Jonathan Dugas said...

HI Noel

Thanks for the comment - I have no "complaint" against the midfoot landing, incidentally, I think you've read it this way for some reason. My complaint is with the advice that is given to people to change their landing because heel striking is "bad". And while you may not have been exposed to this advice, is most definetely exists. I have seen first hand many runners who are pre-occupied with the way their feet land. They needn't be.

You write that the elite runners work a great deal on technique - not at all! At least, none I have ever know even care or think about it - they run, simple as that. I've not worked with any runners (elite among them) who get stuck in this mindset of changing their foot position. Technique to them involves arms, knees, balance, and avoiding overstriding. But even that is natural, not learned. They are not consciously going out there to change the way they run. Now, you may say this is a mistake and they're missing an important piece of the puzzle, but really, they train, they do technique work (not including foot strike, I might add) and they work as hard as possible on running. The "technique movement" is a man-made, created one, that has "sold" the idea of changing a running style based on no evidence that it works.

So that's my issue - the advice, and not necessarily the midfoot landing - 25% of elite runners land on the midfoot, and I'm sure they're happy there, and will stay there. 74$ land on the heel, and they too are fine in that position. But the advice that is given to runners is flawed, because it presumes that heel striking is bad, and tries to change it, without consideration for the consequences.

You appreciate the obvious fact that changing it will lead to injuries, but others have not, and a good many calf and achilles injuries are the only tangible outcome from this advice.

Now, in this comments thread, there've been many examples of runners who have changed, either naturally evolving, or gradually changing their stride pattern. So of course it's possible. But as a scientist, in the complete lack of evidence to support it, the concept of changing it because it's "better" is difficult to justify.

Anonymous said...

Hi Guys, just came across your site, fantastic resource and great work, thank you.

I put my hand up and say I coach all my guys to run fore/ mid foot (not heel) and my reasoning is due to elasticity/ stretch shortening cycle.

Prof Craig Sharp wrote a grate piece on the elasticity in animals (his background pre Human’s!) and humans (can’t find it at the moment but will try and locate it) highlighting the stretch shortening cycle etc.

I have personally been through the whole prescription of orthotics (£600 later) and was told I was pronaiting, flat footed etc and this was causing injuries (pre Ironman but just kept getting blisters) so had to look at the biomechanics of running, and in my quest for answers went backwards in coming forwards.

I have looked at young people, babies, toddlers… very flat feet, arch grow as they develop. Importantly, most run on their fore/ mid foot which must aid in the development of the tendons/ muscles in the foot.
They must then get to an age where shoes/ trainers play a part and bingo they heel strike. You see this in adults, put them on treadmill with shoes on and 99% heel strike, take shoes off and bingo land softer and on fore/ midfoot!

I then look up the chain, soft knees (I find if forces the cycling of heel/ knee), hips forward (very difficult to switch on glute s with bum sticking out), tall with high chest all linking with a foot strike that is under the body.

With the pendulum type running action you see from most heel strikers, they use a lot of quad action to swing the foot through. With soft knees and more of a cycling action using hammies I find I can “sell” this action to triathletes as they don’t want to be using the same dominant muscle group they use in cycling (quads) for running.

It takes time to get all this correct, but one you get it, it feels effortless.

One of the major reasons I feel we see most elite or recreational runners heel striking in my opinion, 1. Due to what happens with shoes/ feet growing up and 2. Most coaches do not teach running technique (the mortal words – stride out/ relax! is about as far as it goes).

Not exact science, but at least my study group if more than 8 participants!

Thanks guys

James Beckinsale M.Sc
Optima Racing Team .com

Anonymous said...

I don't usually disagree with you about technical stuff, Ross, but this time...

This mantra, "25% of elite runners land on the midfoot, 74% land on the heel" is being repeated so often in your piece and in the comments that it appears that everyone is accepting the finding as fact. I'm not.

It is virtually impossible to categorise a bunch of runners as heel-strikers or not by studying video, not matter how sophisticated it is. The only definite categories you are going to pick up are at the extremes.

Here's an example of two fairly elite runners in the closing stages of the 1974 European championship 5000m:

The lead runner, Foster, is "obviously" heel striking, while Viren is going to land flat footed. This picture was taken with 5 laps to go, and they are in the middle of a 60.2 lap.

Compare this picture of the two: http://www.jamd.com/image/g/1253106?partner=Google&epmid=2 lo and behold Foster now appears to be about to land on the forefoot, while Viren seems about to land on his forefoot. (Foster won by 7 seconds.)

So, two (at least) points:

1. Viren -- whether a forefoot or midfoot striker -- is "obviously" not a heel-striker... or is he? The only way you can really tell is by looking at the wear pattern on his racing shoes. I bet you'll find wear on the outside of his heels. That's because almost everyone except sprinters, even classic forefoot and midfoot strikers (and I am one) also puts the heel down, however fleetingly, and to varying degrees. This is even true -- shock, horror! -- in runners trained in and properly utilising the Pose Method. And that is direct, in person, live, face to face, from Dr Romanov himself.

So it's not surprising the Japanese researchers reported 75% of the elites as being heel-strikers: the definition is so imprecise as to be meaningless in the real world.

2. Footstrike -- as does gait and cadence -- varies not only with speed, but how fatigued you are. Taking pictures of elites at the 15k mark in a 22k (odd) race only shows whether they are heel-strikers, or not, at the 15k mark in a 22k race. Let's see the pictures from the 5k point when they're all still fresh. Let's see the pictures at 20k when they're really fatigued and in the final 800 when they're raising their final gallops.

Given all that, let's not take this research too seriously, eh? An interesting piece of work but, in terms of advancing our knowledge in this area, not really up to much.

Ross, I won't say I'm surprised at you.... heh heh heh


Anonymous said...

That second link got a "lo" stuck oin the end...it should read

Compare this picture of the two: http://www.jamd.com/image/g/1253106?partner=Google&epmid=2

lo and behold...

Anonymous said...


I will try to answer your very passionate view.

First this though: "Ross, I won't say I'm surprised at you.... heh heh heh".... hardly the stuff of scientific debate, lets stick to science, observations and leave the bias aside.

I can lend you my copy (should I have admitted that) of London 2002. Gebre's 'real' debut, won by Khannouchi in a WR, a pack of 14 still together at this pace after 30km. Lets look at them after 5km and also after 30km, or anywhere in between. All bar one is an obvious midfoot/heel striker at this pace (72sec/400m).

Obviously, and as has been stated, at 60sec/400m the strike moves further forward... if they were to run at that pace, so your point on Viren/Foster is really not a revelation. If you look at the same marathon in 30min finish time cohorts, then you would see the reverse... the shift is towards heel, but still largely midfoot/heel.

These are observations, not arguments, yet almost everyone who has been exposed to Pose, barefoot etc will not accept it... nor will they take up the offer to look at it btw, because in my humble opinion people look for remedies to 'make them less injury prone' or 'make them faster' etc and feel that if it is a simple thing they have control over (change running style, shoes etc), then its easier than to accept their limitations (ability, biomechanics, injuries etc).

As I said, just imho.

The main thing is though, that a process of natural adaptation means that most are already running optimally for their individual circumstances, be it forefoot , midfoot etc and shouldn't change it.

But we also shouldn't think that how we do it should be how everyone does it... despite the evidence of observation.

Anonymous said...

Just another point (having looked at your argument on your blog).

What the midfoot/heel antagonists cannot accept is that, to land on the heel or midfoot/heel does not require overstriding and therefore 'a braking effect'. If that were so, then I would agree.
Rather, if you observe, they land 'below' the body, with the knee still bent, i.e. lower leg angling backwards.

The other point about adaptation and not changing style. It is my opinion, with 30+ yrs marathon running experience and a scientific background, that if a runner is able to run consistently week in week out, lets say >60km, and hopefully (but not necessarily) getting faster too, and is an 'efficient' runner (in layman's terms he would be on a 'predictor' curve or better over 5km to marathon), then whether that person runs on his heels or tippy toes, he shouldn't do anything to change it, and no-one can tell him its wrong.


Ross Tucker and Jonathan Dugas said...

Hi Simon and the subsequent poster

Thanks for your comments on the subject, I'll respond to each in turn.

First, just to clarify, I know Simon - we spent a great few days in Boulder pounding the pavements, hills and track, and so I take no offence at the last line, all in the spirit of good debate! I say bring on the arguments and debates, it's the only way we can evolve understanding! I can appreciate how it might be misconstrued though...still, no

Anyway, that out of the way, the first thing is to defend the study, and particularly it authors, because they actually took steps to prevent this effect that you're talking about, Simon.

Firstly, they didn't simply video the race, which I really can't stress enough is incredibly dodgy. A lot of people wrote in a while ago, to a previous post, talking about watching videos on YouTube. And while that observation provides the incentive for further research, I'd caution against making definitive conclusions. I do believe you can get very crude observations and tell the obvious footstrikes, but differentiating between heel and midfoot is very sketchy, unless you have equipment like they used in Japan.

To elaborate on that - they used high speed cameras, capable of taking pictures at a frequency of 120Hz. That means a picture was taken every 0.0083 seconds. So they have 120 pictures of a runner's foot every second. There's no error in that measurement, given that rate of capture.

They then defined heel, mid and forefoot with very strict definitions, and if you can get hold of that paper, you'll see the image they work from. So the point is we're not talking about your normal video or youtube here - this is serious capture power, 120 still images per second, and it's very easy to tell where the foot strikes first.

What they can't control is what you point out, and here, you're quite right - there is of course a fatigue effect, and a natural variation effect. But those things tend to "come out in the wash". So the study is not definitive - I will say in my defence that I never claimed it to be! But the weight of numbers - 3 in 4 elite guys running at 3min/km strike heel first, that's a wake up call for the "forefoot is better, just look at the elite" advocates.

So that' the value of the study. But yes, granted that you can't make the conclusion that 75% of elite runners are heel strikers all the time. One might turn that around and say that only 24% of elite runners are forefoot strikers SOME OF THE TIME, which is equally interesting in so far as it discredits the argument I've heard many times - "look at the elites, they run on the forefoot, so should you!". Fact is, they don't, at least most of the time!

Then to the response, well put, I agree with most of what you are saying. In particular, I like the final post you've made, concise and to the point. I agree that some people will run forefoot and cope fine. We've had many comments from guys who have changed their landing point and well done to them! They show it can be done. But it's the advice to people to change it because forefoot is better that is at fault, because there's no evidence for this. The debate can obviously continue, and likely will, but I struggle to find the justification for proactively changing someone's technique.

And I also believe that for everyone who has written to say that they have successfully changed their footstrike, there are probably 2 people out there who got injured doing it! Unfortunately, they are not likely to read this, so we may not hear from them, as they've probably taken up swimming or cycling, or quit altogether, forced by a persistent battle with shin, calf and Achilles injuries to hang up the shoes!

SO again, I can't stress enough - if you have something that is working, don't change it! If you feel a need to change, look at training first, second and third. And then consider shoes, stretching, strength and technique, and do it methodically and carefully. But don't jump on it as though it's the answer, because as 'non-blogger' has written, it's been sold as a solution when there's no reason to believe that it is (or at least, THE solution).


Kevin Joubert said...

As long as we are being rigorous about this, which studies show that landing with your foot underneath you is better than overstriding? I happen to agree, the braking force argument is compelling but I have never seen any studies about it.

Having checked out the pose, chi, and evolution running books and videos over the course of a couple of years, my own impression is that all these guys are trying to get us to stop overstriding. They all say it slightly differently but that is what they are getting after they just have different ways of getting there.

I suspect that going from a heel strike to a midfoot strike STRONGLY predisposes you to eliminate overstride.

Anonymous said...

Thanks Ross for the further explanation, and also for explaining that we do know each other and a bit of banter is par for the course :)

To non-blogger: you obviously missed an earlier comment from cassio598 to Ross's post which said, quite seriously, "Shame on Ross..." I was mischievously alluding to that with my "surprised at you" comment, which was very much tongue in cheek.

I took something of a punt with my criticisms of the study only having seen the abstract and your synopsis of it, always a dangerous thing to do. So Ross, if you have a minute could you add a paragraph or two describing the researchers' definitions of heel strike and so on and how they differentiated between them? Then we'd know what we are talking about.

Back to non-blogger: as I said, and Ross seems to agree, it is practically useless trying to categorise runners by using the freezeframe of slow-mo on a DVD or video, let alone, as Ross says, on YouTube. Now that's truly unscientific :)

My point about Viren and Foster was not that the strike changes at a specific speed, but that pictures of them taken at the SAME speed seem to show them doing different things. The Japanese researchers obviously took this on board, as Ross reports, getting 120 pictures per second of each foot. That's amazing!

Also non-blogger, I can't find where the trainee factoid "at 60sec/400m the strike moves further forward" came in. This makes no sense, as it must surely depend on the individual runner's ability and fitness where this transition happens, if it does.

On your second comment; "What the midfoot/heel antagonists cannot accept is that, to land on the heel or midfoot/heel does not require over-striding and therefore 'a braking effect'. If that were so, then I would agree. Rather, if you observe, they land 'below' the body, with the knee still bent, i.e. lower leg angling backwards"... I have to say that this is not what I observe day-to-day. Sure, some heel-strikers (whatever that exactly means) are quite capable of landing under their centre mass -- and I agree with Ross and everyone else that that is one of the most important things to concentrate on. (Equal priority, maybe, should be given to cadence? What do you think?)

However, I would say that many don't. They reach out, over-extend and hit the ground heel first with a locked knee way out in front of them. If, as everyone seems to be saying, this is often a style associated with relatively slower speeds, then there is little momentum available to overcome the inevitable braking effect. That's why running shoe manufacturers are obsessed with technological fixes aimed at facilitating "a smooth transition from heel strike to toe off" (to quote the latest New Balance advert from RW).

Your last para is a beaut, non-blogger, and unarguable.

Anonymous said...

Can someone please address the point of how shoes can change the appearance of first point of contact simply by the fact the heel of a shoe is built up. So if someone landed on the midfoot without a shoe on, then the exact same angle with a shoe on, the heel would touch the ground first - if the angle of foot to lower limb remained the same. So the first point of contact is with the shoes heel - but the first point of pressure applied may still be in the runners midfoot.

I would think this the most critical aspect of all, as the arguement may very well be largely mute if this point is not solved.

Anonymous said...

Most often missed in over emphasizing running technique is running action.

Not how you do it but the what of what you are doing.

How the foot strike is not the issue, the issue is what are you doing to get your foot close to your center of mass before the initial contact occurs. Once you get the what then you can move on to the how.

How will often relate to not arm actions but hand actions or lack of it.

To see a good example of this, watch the race of the womens 800 from 2007 World Track and Field Championship in Osaka, Hand range of motion and speed controls the what you are doing with your feet.

One other issue is the definition of over striding, most people assume this to be a reaching effect, but if you can understand over striding as the foot landing in front of the COM of mass, be it 1 in or 12 inches, and it is not moving backwards when initial contacts occurs then you are over striding.

That is why you find people such as POSE, advocate 180 steps of very small strides as a way to negate the "over reaching" principle.

If you go back and look at the tape you will find the commonality of those in the front have more active foot landings then those in the back, regardless of foot strike position. Along with that you will find they also have active hands over active arms.

Smitty Werbenmanjensen said...

Thank you for finally talking sense about footstrike.

Barrld said...

Ross--Many thanks for a great post/blogger exchange. Let me tell my little foot striking story:

I've run for more than 30 years and was a modestly successful recreational heel-strking racer in the mid-90s. Though I cut my racing back considerably in the late 90s I continued to run regularly and ran my last marathon in 2001. We had two children in 2002 and the time demands of parenthood cut way into my running and all other exercise. About three years ago, having surfaced from toddlerhood out of shape and maybe 12 pounds heavier than in my racing days I decided to get back into decent running condition with the hope of occasionally racing again. Six months in and my knees were painfully sore after even short runs (3-4 miles).

I worried that I had to hang up the trainers; then a friend sent me his POSE video and books and I spent six months trying to master the technique of fore-foot striking. I went through the sore calf and shin stage and my knees stopped bothering me, but I rolled my ankles a number of times and, most troubling, I hurt my hip. After a painful bout of bursitis and a cortizone shot I said to heck with it and went back to my old style. I also joined a local track club for speed training and the group motivation thing.

When running I don't worry about my feet anymore and think about good, relaxed upper body posture and smooth arm movements. I've found on the track that I can speed up (a relative term with me I know!) by leaning my upper body forward, which is a POSE method.
I strongly believe now that gait and foot striking should be be left untouched for virtually all non-elite runners. My knees don't hurt much either, maybe it was the weight.

BTW, last year was my most successful racing season in 12 years, so I must be doing something right.

Anonymous said...

I think this post is missing the point of forefoot and mid-foot forms. I'm assuming that it's not an April 1st joke.

Before I even heard of pose, chi, or barefoot running, I made a permanent switch to mid-foot striking after years of sustaining various injuries while building up for marathons. I wasn't concerned with over-using my calves at the time and after recovering from some initial soreness could run any distance without difficulty. I quickly found that a low, quick mid-foot stride let me reduce impact, navigate rocky trails and rough downhill stretches with dexterity and speed, and increase my overall speed on any terrain and surface.

I'm not a competitive runner; I do it for fun and the challenge of expanding my limits. Pain isn't fun. Whether or not contorting my body in an unnatural way for miles on end might add several more seconds to my mile has no relevance when I consider what's going to keep my on the roads and trails in the long run.

LKPT said...

The forefeet are not designed to take the impact of long distance running without eventual injury to the balls of the feet. However the knees are less prone to injury with this running fashion as the knee does not extend as much during forefoot landing. I believe the hips are the next area of possible injury/pain with the forefoot striker...where as a heel striker may experience eventual knee problems/pain after many years of running. LKPT

Dr.R.P. said...

There are other reasons other than speed for changing your foot strike. I ran for years landing on the front of my foot - it was the way I learned to run in high school xcountry. I had hip pain for years and only recently, after going to a physical therapist, found that the 2 were connected. I relearned how to run, with a heal strike, and have so far been pain free. An added benefit is that my feet would often be sore the day after a hard run (I have lousy arches and use orthotics), and have found diminished pain since reverting to my new foot strike.

Anonymous said...

Yes...less pain in the hips with heel strike...You may experience possible knee pain at some point or not...The thing about heel strike is that you are extending your knee to a greater degree...so greater hamstring flexibility would be beneficial...LKPT

SteveL said...

This is great information and very helpful. The one thing I think nobody is considerin is new runners. There's much discussion about changing from heel to midfoot or forefoot strike, but what do we teach young kids? What about me, a 41 year old taking up running for the first time last week? I have nothing that works for me already, and I'm trying to find what does. Sor far I can tell you I do not naturally heelstrike unless I concentrate on doing so. The same is true with forefoot, except when sprinting. I seem to land "midfoot"...but when I try moving even more foreward, my runs have been easier and the shin pain that has bothered me vanishes. There is also less braking and it feels lighter. As for my calves...unlike you long time heel strikers, my calves are not any worse off than the rest of my legs, so it doesn't much matter since everything needs to build up strength.

I think the ultimate question of what we should be doing needs to start from the perspective of somebody who has never run for distance. Once you've adapted your body to style, what YOU should do becomes a different question altogether. Don't study elites, study beginning JR high cross country runners who have had no coaching.

Anonymous said...

Excellent post. Incidentally, I have been experimenting with heel vs. midfoot vs. front foot strike techniques over the past couple of weeks. I am training for the Berlin Marathon in September 2008. My target time is 03:30:00 with a pace time around 5 min/km. So, not elite, but somewhat ambitious.

On Sunday I ran a 12 km "long run", and on purpose I just focused on letting my feet land naturally, not attempting to force heel, midfoot or front foot strikes. My main focus was really on maintaining a heart beat around 90% of max to teach my body to deal with lactate buildup. I clocked in at 60 minutes.

On the following Tuesday (2 days later) i again ran a 12 km. This was one of my easier runs where I keep my heart rate between 80%-85% of max heart rate. But, on this day, I deliberately focused on landing with my feet under my center of gravity AND on front foot striking. Towards the end of the run, my calves felt very different (tense and tight) from what they normally feel like. No surprise. But what really surprised me is, that my time was identical to that of two days earlier = 60 minutes. My average heartbeat was 86% of max.

So, on Sunday, I complete a 12 km run with average heartbeat at 92% of max. in one hour. No focus on foot strike.

On Tuesday (two days later) I complete a 12 km run with average heartbeat at 86% of max also in one hour. With focus on front foot strike.

What do you think? Can I conclude, that my cardiovascular system does not have to work as hard to accomplish 12 km in one hour if I land on my front foot? This would ultimately lead me to the next conclusion: Front foot striking is more efficient and therefore will make you run faster!

I can tell you, that one conclusion is firm and 100% proven: my calves are hurting like hell today (Thursday) and I can barely walk!! :-) My interval session this evening will have to be cancelled...



Anonymous said...

Thank you for the information about the japanese study. But I think you are missing the point a bit.

My experience is this: different shoes, different footstrike!

What do I mean? With many, not all, training shoes that are clumsy in the heel and have pronation or stability devices you HAVE to try to touch the ground with the forfoot first, if you want to run faster than a slow jog. If you dont't you risk dislocation a knee or ankle! It feels like that anyway.

Why train in clumsy shoes then? Well, lots of pseudo experts have made us. Don't just blame the shoe companies. Test in magazines like Runner's World have forced the companies to make shoes like that, if they wanted to remain in business.

My favorite shoe when I was in my upper teens, made by a company you never have heard of, was more or less put out of business because of those so called test.

Is it dangerous to train in so called racing shoes? My pair Adizero Kaha. I use them on some training runs, not every, have much more cushoning and stability than the training shoes I used 25 - 30 years ago.

Anonymous said...

were there any results about ground contact time?

I have to look at all research that attempts to describe human performance with a bit of skeptism. there are just too many known factors and who knows how many unknown factors that effect performance.

Running form is very individual - biomechanically we all move differently. making references from one study that captures a single moment in a 21 Km race is on very shakey ground.


Anonymous said...

hey guys great article, if you want more scientific info on technique and training check out a Dr Romanov and his Pose method of running.

Anonymous said...

I am wondering something. All of you are talking about measuring forefoot running as if it can be done just by looking.

My experience with it is that forefoot running cleared away all my knee problems which I believe were caused by heel striking. But if you watched me run, you'd see my heel touch the ground the instant before I rolled over my foot. What I mean is, I put the force on my forefoot, but I don't land directly onto it. It's as if I roll over my heel, and it is all very smooth and feels far more natural than any running I've done before. I'm inclined to think that's what is truly meant by "forefoot running".

Anonymous said...

I did research on fore and rearfoot strike running, there will be a publication soon. The aim of this study was to assess differences in muscle activity of the abdominus transversus muscle and multifidus during running in two different kinds of a running pattern - the rear foot strike running pattern and the forefoot strike running pattern.
This study has demonstrated clearly that the activity of the abdominus transversus is more symmetric in the RFS running pattern than it is in a FFS running pattern. The results indicate that superior core stability is achieved in a RFS running pattern.
The advantages of a RFS are:
- A slightly diminished impact peak at midstance compared with a FFS running pattern. This might be essential because the greatest potential for injuries during running is also at 60 % of the stance phase. In this mid to late stance maximal calcaneal loading occurs together with maximal Achilles tendon, plantar ligament and plantar fascia tension.
- The mean center of pressure pattern in a RFS running pattern is moving in an almost rectilinear motion toward the midline and the anterior part of the foot. The mean center of pressure distribution in a FFS running pattern however does not demonstrate this rectilinear motion. The first contact is made at approximately 50% of the foot length. After this first contact, the center of pressure is making a brief anterior movement, then changes its direction in a posteriorly / medially direction before moving anterior again. It seems reasonable to hypothesize that FFS runners are particularly at risk since they are experiencing large forces in a region not “designed” for these impact forces.
- Ardigo (7) found that the mechanical work (in watt/kg) was significantly higher (7-8%) for a FFS compared with a RFS for a running speed between 2.5 and 4.0 m/sec. The oxygen uptake however was similar for both groups, meaning that a higher storage and release of energy take place in the elastic structures of the lower leg with a FFS. The RFS running pattern at the same speed might be more gentle for the anatomic structures in both feet and lower extremities.
- A more symmetrical contraction of the transversus abdominus muscle means a better core stability for the RFS running pattern.
Relevance of core stability
A strong foundation of muscular balance and core stability might be essential for distance running. Weakness or lack of sufficient coordination in core musculature could lead to less efficient movement, compensatory movement pattern, strain and overuse injuries.

JSvD said...

Jan Swager van Dok

Anonymous said...

What a great discussion. I am linking to this site from my site www.fitbricks.com so my athletes can read this discussion too.

Although the athletes I train are not elite triathletes or runners, this discussion is very important. I would say 80% of the 10+ min/mi woman I coach are heel strikers. I too noticed as they got faster with more training, their foot strike moved up from heel to mid foot. I would never recommend quick changes in anyone's natural form, but work to change their form over many weeks/months. This goes not only for foot strike, but where the foot lands in relation to the body. I believe foot location actually helps these athletes the most to increase their run speeds.

Great discussion!

Diane Stokes

Anonymous said...

I was a heel-striker for 25 years when I developed a heel spur-like pain on my left heel. I was measured and found that I had one leg shorter than the other (right lower leg bone was shorter by 1/4 inch).

I decided to try landing on midfoot and the pain that had been off and on for six months went away immediately. That was 1.5 years ago and I've had no trouble since. My racing times are (still) improving and this switch seems not to have impacted my racing either way.

For me, this switch has been a help. Do you think I should have added another sockliner to the right shoe and kept heelstriking?

Anonymous said...

Interesting! As a midfoot lander in my twenties I became more a heel lander now in my fifties. And you say it is a matter of running speed? That could be very plausible (I don't run a sub 2:20 marathon anymore).

But maybe there is more involved, that also could contributes to the high percentage of heel landers in elite road races. When I jog comfortably on my running shoes, I land over my heels. But when I jog the same speed barefoot on hard pavement, I automatically prevent heel landing. I suppose nobody would prefer to hit the bricks heel first. You switch to your natural shock breaking system, by mid- or forefoot landing.
I've read the post on Quickswood about shoes and injuries. This story combined with yours brings the following question. Is it not possible that the increased comfortness of running shoes induces heel landing?

Gerard Smit

Anonymous said...

Not sure the high speed camera option is a valid way to evaluate foot striking.

From experience I know that I can land mid-footed with 90% of my weight on the heel or 90% of my weight on my toes...but externally it looks like my whole foot touched flat...but it makes a big difference. A pressure mat needs to be utilized to plot the landing's pressure profile.

IronBrandon said...

The most important statement in this, with regards to "changing" the forefoot running is:

"However, if you can gradually change your landing, then I do believe that you can shift your footstrike. But it's a gradual process."

If I may suggest Christopher McDougall's book Born to Run in which he gives plenty of research to back up the claim of many forefoot/midfoot strikers that heel strikers do indeed wind up with more injuries.

Anonymous said...

Hi Ross,

What an interesting blog discussing running techique and the footstrike. I thought it might be interesting to mention running footwear and the role it can play in a runner's techinque or running style.

First, let me mention that I am the General Manager of ECCO Performance and we have just globally launched (March 15th)a very new and innovative collection of running footwear called BIOM. The concept of BIOM was built on the running movement called natural motion. There are references made to barefoot running but it is somewhat contradictory to refer to footwear made to imitate barefoot running. Let me tell you and your readers a little about BIOM.

BIOM has been 2 1/2 years in the making. It is part of the natural motion movement, which is gaining much popularity worldwide. Other product/brands that re part of the movement are Nike Free, Newton, Vibram Five Fingers, etc. This movement is quite a shift in the current running market paradigm of stability and motion control.

Fact, in the last 20 years, with all the great techonolgy introduced in running (cushioning, motion control and stability), running injuries have not been reduced by even one tenth of a %! BIOM has been built to allow a runner to run on a more natural path as they would if they could run barefoot. BIOM can be used by any runner; heel striker, mid-foot striker or forefoot striker. BIOM is about minimalism, which ECCO believes can help stregthen the foot's muscles and ligaments and thus hopefully reduce running injuries. We have commissioned the University of Cologne to conduct an independent study of the effects of natural motion and BIOM on a runner's foot health/injury rate. Stay tuned for the rsesults in 12-18 months.

Until than, if you or your reader's would like to learn more about BIOM, please go to thebiomproject.com.

Happy Running,

David Helter
General Manager
ECCO Performance

Dave said...

I have gathered the opinions of some of the world's best triathletes and some independent physio therapists and composed it all into an unbiased article.
If anyone is interested to read this go to:


Ross Tucker and Jonathan Dugas said...

Hi Dave

Thanks for the link. I read the article - it's well researched and constructed, a good piece of work.

It also makes the same conclusion about changing the footstrike that I have made, so perhaps "closed-minded" (if this site falls into that category) brings us to the same position.

I must just clarify one thing - those viewpoints are not unbiased, because you're asking people who use the technique to evaluate it. It's like asking supporters of NFL teams to objectively rate their team's chances in the upcoming season. Or asking car-owners to give an opinion on the difference between German, Japanese or US-made cars. It's not actually unbiased.

"Unbiased" requires an evaluation of the scientific evidence, gathered through research where the researcher has used good methods to evaluate an outcome without any financial or other incentive to find one thing or another. I'd like to think that is what I've tried to summarize in these articles.

So it's a good article, I enjoyed it, but it's not unbiased.

And then finally, no one asks the very best runners in the world for their opinion. Yes, you've asked some great triathletes, but they are well off what I believe are the best runners - the East Africans. Their opinion is the one I'd love to see. I suspect they'd be completely mystified that the debate even exists. I can imagine them shaking their heads, laughing and having no idea how to answer the question. I believe that holds a valuable lesson.


tim said...

I am hardly an elite athlete nor a scientist but I started running after a 10 year hiatus and started to have calf compartment issues.

I knew I ran on the heel so I started reading ... Pose Method and Gordon Pirie etc.
Before opting for surgery, I thought I would switch my running method and shoes.
That was a year ago, I have had no issues since switching to a more midfoot method.

We are naturally fore-mid foot runners, all the evidence I need is watching my 3 yr old daughter casually chasing me in a game of tag.

Michael Yessis said...

One factor that appears to be missing in the discussion of whether it is better to land on the heel or midfoot, is running technique. There are different ways to execute a midfoot strike as well as a heel strike. Close examination of each can indicate the potential for injury as well as improvement of running speed.

In my work with runners, ranging from sprinters to marathoners, the first thing I do is film them in order to determine not only their landing technique but also how well they execute the other key elements in running technique such as the knee drive, pawback, pushoff and arm action. The knee drive and pawback are especially related to landing technique.

My analysis of running technique then tells me what changes should be made and why. Without complete evaluation of the runners technique it is impossible to single out only one factor and correct it. Many of the changes made have a direct effect on other actions. I discuss all of this to great lengths in my book Explosive Running. It has sequence pictures taken from organic digital videotape of sprinters, middle-distance and long-distance runners as well as specialized exercises to help make the changes and/or improve performance.

Michael Yessis, Ph.D
Professor Emeritus, CSUF
President, Sports Training, Inc.

FasterThanU said...

Everyone is driving me crazy with your heel strike crap!!

1) Better? What is your goal, speed or injury prevention? Do they coincide? The heel strike is obviously more prone to knee problems.
2) It was noted that the knee lift is generally higher for elite level runners...then everyone discredits it saying you don't want to overstride. So fix contact point, don't negate your lift.
3) Stretch-shorten cycle - If you don't understand why increased rebound without any effort is good for running, you might want to think about energy consumption while running.
4) The "STUDY" The so called scientist has so many fallacies of belief in this article, that I doubt it will ever be published. Fallacy a)"Most injuries occur at 60% stance phase" - most people run with a heel strike. Fallacy b)"superior core stability is achieved in a RearFootStrike running pattern" Immobilization of the core is not biomechanical stability. The result should be stated that the core is used less in RFS runners. Is this better?
If you read the bottom of his post it is even more of a stretch.
Fallacy c)"It seems reasonable to hypothesize that FFS runners are particularly at risk since they are experiencing large forces in a region not “designed” for these impact forces" This suggests that the force absorption occurs at the ankle.If you run ffs your knee is bent more, allowing you to absorb the force with your hamstrings and gluteals(the biggest muscle in your body) Seems like a good place to absorb force to me. And utilizing the SSC the spring from those muscle groups is superior.
5) The technique vs. training argument seems silly to me since all across the world they train much more intensely and frequently than we do in the US and with less injuries. Yes you can overtrain. Is technique the answer? Not always. Technique is usually nothing more than a symptom of underlying musculoskeletal dysfunction.The most obvious one in running is that of the low kneelift like I stated previously. Its not that you don't raise your knee, its that you can't do it effectively.


This is a great topic though.The main problem with these such topics is the ineptitude of the mind to adapt, not the body. haha

Running Diva said...

Interesting analysis. I learned so much!

Ron Wolf said...

terrific article, incredible (good) discussion. however, i was surprised that there was only the briefest mention of ChiRunning and that was in the comments. so i want to just add a pointer to ChiRunning's website to the discussion as a major component of that style is the mid-foot strike. i cross posted this blog to the ChiRunning site along with some comments that you can find at:


ChiRunning has successfully helped thousands to run without injury and, probably, to run more efficiently. so from that, i believe (biased, yes) that its valid to consider. one element of its success derives from ChiRunning taking a more whole body approach to running form. i don't think that any ChiRunning coach would be surprised that runners who are coached only to change footfall would come up with injuries. more things have to change, of course.

also, i'll offer a critique of the basic research behind this article - i find it notable and amazing that everyone who changed footfall had injuries in the next week. this leads me to wonder what else they were all doing because its very very odd to get a consistent result like this from any sort of study. for instance, were all of the runners also participating in a certain stretch routine? or plyometrics? etc. i would guess so, they were all on the same team under the same coach, right? so to blame the injury on the changed foot strike alone is just poor science. well, not really science at all....

Marty said...

You guys go way over board on this analysis. Faster runners typically wear minimal shoes, aka racing flats. Go ahead an try to land on your heel in those! Longer distance runners typically wear heavier shoes w/ arches and support ... but if you like the mid/fore running styles and wear minimal shoes for long distance, you would normally land mid and front as you would when you run/walk barefoot. Thus you may start with different forms with support shoes, but everyone runs the same when minimal or barefoot.

Salami said...

Good article. I've been skeptical about forefoot running since I first learned about it. It's good so see an article like this that clarifies just where the science is on the subject.
I'm 30 and just finished a 7th week and a walk/run program. I've been trying to land midfoot more often the last few weeks and the area under my calf/above the ankle is slowly feeling sore. I'm thinking this might be due to extra stress put on the Achilles area the original article warned about?
Second, I've tried to get some more work from my hamstrings instead of from the quads. I found it increases my run speed and I naturally land more midfoot. For me, this confirms the idea that foot strike is a largely a function of speed.
In short, I'll just keep moving at my slow, beginners pace, concentrate on having the foot land underneath my body, and just keep everything moving forward in an efficient manner.

Anonymous said...

OK, I didn't read all the discussion, but one important factor in the heel strike v's forefoot strike is the geometry of the shoe.

Typical running shoes have a 24 mm heel and a 12 mm forefoot, which is effectively a 1/2 inch heel left. Racing "flats" are 20-10 mm, although there is a small variation across manufacturers.

The study quoted in the original discussion was a Japanese group who looked at where the shoe touched the ground first, not body position or where the runner weights the shoe.

Take away the heel lift and you get a very different picture of how people run barefoot (I believe this to be the natural state of running.)

The bulk of the nerves and articulated structure of the foot is in the forefoot (front half, distal to the arch.) Running in a 1/2 inch heel lift creates unnatural afferent feedback - feeling the ground under your heel even when your foot comes down dorsiflexed or flat.

Proprioceptive reaction is to brake briefly (for about 1/30 second) when you feel the ground on your heel. Wanting to move forward though, we override this and accelerate again creating a stop-start kind of running gait.

If you level out a shoes geometry by reducing or removing the heel lift, you see a very different picture of how people run. In fact Prof Den Lieberman from Harvard department of Anthropology is publishing a study which shows how humans run naturally. This will likely be published in Nature or Science.

Ian Adamson
MS Sports Medicine
BS Bio-Mechanical Engineering

Ian Adamson said...

Ian Adamson

Gemfinder said...

Has anyone studied the recovery rate for runners with chronic knee pain changing to barefoot POSE?

I began running at 14. Knee pain began at 17, increasing every year till I gave up at 37. Could no longer run 2 miles without debilitating pain, beginning in the first 100 yards, lasting days, sometimes with inflammatory lockup.

At age 40, tried POSE on traditional shoes. Ow. Knee and arch pain on even brief runs. Gave up again.

At age 43, on a whim, I tried midfoot again, but on barefoot-style shoes, building calf and Achilles strength slowly over a few weeks (walking, then running 5 minutes, building up to 2 miles, recovering 4 to 6 days after every run).

The results were instant. For the first time in over 10 years, I can run 2 miles without pain.

Yes, calves hurt like hell at first, and would presumably have ended in injury if I hadn't used such long recovery periods. Took weeks to adapt.

Because the change was so dramatic, it occurs to me that a researcher might need relatively few samples showing similar results to reach statistical significance. Start with a population that is already suffering chronic joint pain, and test whether midfoot affects that.

This, in turn, would be a circumstantial argument that impact force was reduced. But more important, it would directly answer your question: why switch?

Anonymous said...

Hey guys, I am a 39 year old recreational runner who converted to "natural style running" about 6-7 years ago, mainly following Gordon Pirie's advice and incorporating some POSE drills. I am still running this way, and it works great.

Your articles correctly stress the long time needed to adjust - it will take some months to a year for most people, I think.

Another point that I want to make here is that in "heel striking" vs. "forefoot striking" the point should not be about which point of the foot touches the ground first, but through which point the deceleration/acceleration forces go in the load-bearing phase.

That is, you can strike the ground heel first, and still have the weight born by the forefoot/ball of foot, depending on the timing of the foot movement while the center of gravity is moving over the load bearing point.

In fact, in shoes with a higher heel it is necessary to run with heel contact first (e.g., in army boots) to avoid overuse injuries in the ankle tendons. Still, the feeling of running on the ball of the foot can be maintained in these shoes - but it would look like heel-striking if judged by the point of first ground contact.

This is clearly a problem with video analysis studies, since "heel strike" vs. "forefoot strike" is visually defined (I suspect), without taking into account the concept of a load-bearing phase vs. landing- and take-off phase.

Just my 2 cents,

Christian Lemburg