website statistics

Is Running a Controlled Fall? Why the Gravity Model Doesn’t Make Sense

Posted by on Oct 22, 2012 | 54 Comments

How exactly do we run?

It’s a question worth asking for runners, coaches, and running gait theorists. Some would argue that gravity propels us forward. In fact, I used to buy into this model. It seemed to make sense- lean forward so you’ll off-balance and let gravity pull you down, which resulted in forward movement.

Then I started to read about running gait from researchers that actually study gait in labs. The gravity theory didn’t hold up to empirical scrutiny. About the same time, I started interacting with some of the “natural” running theorists. They had a much different take on how we run. The most influential have been people like Steve Magness, Mark Cucuzzella, Jay Dicharry, and Pete Larson. Their ideas of running gait were far different. More importantly, their analysis was based on actual observation using scientific methodology.

The following is a re-post from a discussion we’re having over at the Barefoot Runners Society. We’re debating how we run with an emphasis on the claim that running is a controlled fall. Here’s a summary of my counter-point. Note: the goal of the post isn’t a detailed biomechanical or physics explanation of running gait; it is part of a colloquial discussion.
Role of a lean in running

Regarding a workable model of gait- The spring model works pretty well. Mark C. uses an pogo stick as an analogy. This is how he describes the need to lean- it changes the vector, which results in forward motion.

For example, if you bounce on a pogo stick and remain perfectly vertical, you go up and down. If you lean slightly forward (or any direction), you bounce that way.

Likewise, if you stand perfectly vertical and run, you run in place. If you lean slightly forward, you move slightly forward. The more you lean, the more horizontal movement is generated.

Role of muscles in running

This model only accounts for the energy return (“springiness”) generated from the return of energy stored in soft tissues. It doesn’t account for the other factor in forward propulsion- muscle contractions.

It is possible to run forward by standing perfectly vertical. You can run quite fast, in fact. Why? Muscles generate most of the energy required to run.

Role of gravity

Gravity does play a role in running, but not as a propulsive force. It’s a force that must be overcome, along with drag. A simple diagram of pelvis position can explain what’s going on:


This is the concept Curb and I have discussed. At mid-stance, the hips are at the lowest position of the gait cycle (note the black dot below the line). As we toe off (NOT a good term, IMHO), the hips move vertically to a high point during the aerial phase. The only “falling” that occurs is on the downward arc immediately before touch down.

If we were continually falling forward, mid-stance would be the point where our hips were at the highest point. The hips would then lower as we toe off (due to falling), would reach the lowest point at the middle of the aerial phase, then begin an upward arc at touch down.

The Unspoken Question

The Pose and Chi crowd seem to neglect a critical part of their theory. If you’re falling, you need to expend energy to get back to a point where you can fall again. If we really were falling and we saw the movement pattern that would be the opposite of the above diagram, the point from touch down to mid-stance would require muscular energy to return to mid-stance so we could fall again.That energy expenditure would cancel out any possible gain from falling, plus more. To date, Ken S. is the only Pose coach that has addressed this issue… which is the reason I have so much respect for his ideas.

Why it’s important to study gait

We don’t necessarily need to study running gait to be able to teach people to run better. However, knowing all the underlying theories is a HUGE benefit to diagnosing problems or mastering a variety of teaching methods. I used to teach high school psychology. It was immensely helpful to have a strong background in the topic. I also taught history and didn’t have a strong background. I was a great psychology teacher and a terrible history teacher.

It’s also important to think critically about our own beliefs. Read the first edition of my book, then read some of my recent blog posts. I have different opinions on a TON of topics because I question my own beliefs regularly. While I really believe my above comments are the most accurate current model of running gait, I’m always looking for better models and will ditch this if something better comes along. I’m a science dork at heart, and that comes out in my inherent skepticism.

Folk knowledge versus science

Bare Lee made a comment awhile ago about the difference between folk wisdom and science. As barefoot runners, we tend to rely on folk wisdom, or tips and explanations that get us to more or less do the correct movements but have no basis in science. Examples would be:

  • Lift your feet
  • Bend your knees
  • Pull your feet up
  • lean forward at the ankles
  • Run with a cadence of 180 or 180+
  • land pancake flat
  • feet land under the center of mass (gravity, hips… whatever)
  • stand up straight

This also includes explanations of movement like “running is a controlled fall.” It also includes claims like barefoot running reduces injuries, makes you faster, more efficient, etc. All of these ideas aren’t grounded in empiricism or are obviously inaccurate, but are effective as teaching methods because they work for some people some of the time.

However, we can’t confuse folk wisdom for science. Science is a controlled method of inquiry, and none of the above claims has been firmly confirmed via science. As such, it’s important to figure out exactly how we run… or even if there’s a right way to run. Many topics can be added to that list: what are the role of shoes? Is there a right way to teach people to run? If so, what is the most effective method? Is there a way to reduce injuries? If so, is it universal?

Folk wisdom is an adequate way to answer the questions in the absence of empirical research, but it shouldn’t be a replacement.

So, anyone care to add to the discussion? Feel free to comment. Since this topic tends to evoke emotion, I’ll delete any comment that isn’t civil in tone.


Be Sociable, Share!
Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Related Posts:


  1. alvinj88
    October 25, 2012

    so evolution running concept was right?

    • Jason
      October 25, 2012

      I gave away my DVDs some time ago… in which regard is it right? I recall liking the ideas presented, I just need a refresher on the concept.

  2. Miss.R
    October 24, 2012

    Thanks for such a great article here. I was searching for something like this for quite a long time and at last I’ve found it on your blog. It was definitely interesting for me to read about web applications and their market situation nowadays. thanks one
    hotel london

  3. Leard
    October 23, 2012

    I’ve been mulling this over for a while, and there is obviously a lot of simplification going on. One thing that pops out at me is the reference to “center of mass”. The actual center of mass of the system is going to move substantially depending on how your arms and legs are positioned. In the mid stance pictured in the diagram, your COM would be well in front of your hip joints. The majority of both legs, the whole torso, an at least one arm are all in front of the hips.

    • Jason
      October 24, 2012

      Agreed. The drawing is an over-simplification of the concept and isn’t drawn to scale (I borrowed it from another source).


  4. Bare Lee
    October 23, 2012

    I’ve found it’s most helpful to coax the ground to push me up and forward rather than pull me down and forward. I guess that’s more bounding than running though, right?

    In any case, always fun to see people bandying about scientific terms without understanding the underlying concepts. However, when I want to learn about stuff like gravity, quarks, and so on, I tend not to consult accomplished runners. On the other hand, when I want to learn about training protocols, I listen to the guys who win races. We can’t all be good at everything :).

    • Jason
      October 24, 2012

      Are you suggesting we’re not biomechanical and physics experts? 😛

      Seriously though, this is my preferred method of operation: Write a post describing a phenomenon using colloqial terminology, then let more experienced experts make the necessary corrections and debate the merits of the argument. It’s how I learn from others since I’m generally too lazy to read and too dumb to comprehend the more formal sources of knowledge.

      • Bare Lee
        October 24, 2012

        I prefer to be obtuse most times, but let me clarify by saying I was teasing the Pose crowd and their looney tunes theory of gravity, not you. You’ve always called them on that. And I greatly appreciate your method. I often rely on you to digest the material for me, because I know you always bring a healthy skepticism and open-mindedness to any issue, as well as a sterling intellect. Plus you’re a bum like me.

        • Jason
          October 24, 2012


          Okay, here’s one. The spring-mass theory feels a little too neat to me… there aren’t a lot of detractors. Shit like that raises a red flag in my skeptical mind.

          What’s wrong with the theory?

          • Bare Lee
            October 24, 2012

            I’ll look into it, but right now I’m a little burnt out on running stuff. I tweaked my knee a few weeks ago, out of the blue, less than a mile into a 6-mile run, so I’m just trying to get back to some decent running. It still feels a little sore after a 2-mile run, but it’s getting better. Meanwhile I’ve added a bunch of lower body and legs stuff into my strength training to prevent this from happening again.

            Could you give me a few references though? I’ll look at your post of a few weeks ago, but if there’s anything else I’d appreciate having a look.

            I agree springiness is a good thing to be aware of, and perhaps the easiest thing to consciously optimize, but for me sinking my hips down just a bit has been perhaps the most important cue. That seems to help load the spring and also seems to help time its release into the recovery phase. I’m also hoping it will help prevent further knee-tweaks, since the hip-sink forces the knee to bend more without actually thinking about bending it, and thus avoids exaggerating it.

            But for all I know, all this is a figment of my imagination. The doc I saw for the knee said I got a few gait analysis coming my way, so perhaps I’ll take him up on it when my knees a little better.

          • Bare Lee
            October 24, 2012

            You mean this?:

            “A simple spring-mass model consisting of a massless spring attached to a point mass describes the interdependency of mechanical parameters characterizing running and hopping of humans as a function of speed. The bouncing mechanism itself results in a confinement of the free parameter space where solutions can be found. In particular, bouncing frequency and vertical displacement are closely related. Only a few parameters, such as the vector of the specific landing velocity and the specific leg length, are sufficient to determine the point of operation of the system. There are more physiological constraints than independent parameters. As constraints limit the parameter space where hopping is possible, they must be tuned to each other in order to allow for hopping at all. Within the range of physiologically possible hopping frequencies, a human hopper selects a frequency where the largest amount of energy can be delivered and still be stored elastically. During running and hopping animals use flat angles of the landing velocity resulting in maximum contact length. In this situation ground reaction force is proportional to specific contact time and total displacement is proportional to the square of the step duration. Contact time and hopping frequency are not simply determined by the natural frequency of the spring-mass system, but are influenced largely by the vector of the landing velocity. Differences in the aerial phase or in the angle of the landing velocity result in the different kinematic and dynamic patterns observed during running and hopping. Despite these differences, the model predicts the mass specific energy fluctuations of the center of mass per distance to be similar for runners and hoppers and similar to empirical data obtained for animals of various size.”

          • Jason
            October 24, 2012

            Yes, that’s it. So… any obvious red flags?

          • Bare Lee
            October 24, 2012

            I’ll read the article soon. Should I post my impressions here or PM you at BRS?

          • Jason
            October 24, 2012

            How about adding your comments to the thread in the Coaching forum over at the BRS? The limited nesting here is a pain for conversing, and I hate the BRS PM system.

          • Bare Lee
            October 25, 2012

            The best I can come up with, and this concurs with your skepticism, is:
            “Quantitative comparisons of the [spring-mass] model with human data suggests that, while capturing the vertical dynamics well, it does a poorer quantitative job of capturing the horizontal dynamics.”
            Page 14.
            This makes intuitive sense to me, as I was going to suggest that force application seems to be unaccounted for in the spring-mass model, but I really don’t understand the model very well in its technical details.

          • Jason
            October 25, 2012

            I think Magness does an adequate job of utilizing the SM model into a more comprehensive description of motion. Any thoguhts on his thoughts?

          • Bare Lee
            October 25, 2012

            Yes, Magness is my go-to guy for most things running. I discovered him last spring, or thereabouts, after a frustrating time trying to find straight answers on the barefoot sites. I have a very limited understanding of this stuff (I switched out of my biology major 30 years ago), but everything he said made sense, not just on form, but about cadence, HR, training protocols, and so on. That was my light-bulb moment; I realized there was no barefoot running form, just good form, and that the best way to learn about it, at least in theoretical terms, was to tune into those who practice or research this stuff professionally. For coaching, and specifically coaching cues, I’m on board with you: there are lots of ways to skin the cat, and each coach must figure out what works for each client. This is what we’ve called the difference between science and folk knowledge. And I think your system works best, based my own experience. Unfortunately I didn’t read you right away when I began looking into this stuff, but I’ve also concluded that (1) good posture and (2) proper foot landing are the main components of good form; everything else is tweaking. The main tweak I’m working on is sinking my hips down a bit. That seems to help with proper landing and release. Also, I’ve found that running at a faster pace leads to better form (as I later read you have suggested too), so it was nice to have this confirmed recently when I read that the Kenyans first learn to run fast, then add distance. It’s sort of the anti-Maff method, and it makes a lot of sense to me, but once again, that’s just based on my own personal experience.

          • Bare Lee
            October 25, 2012

            So yes, just to be perfectly clear (Channeling Tricky Dick here), I think Magness has a complete model of what’s going on. I was also relieved to hear Dr. Mark disavow the claim in his otherwise excellent video that human spring is maximized at 180 cadence, which seems absurdly exact given all the variables at play (different limb lengths, different paces, different genetics, different athletic backgrounds, etc.)

  5. Doug
    October 23, 2012

    1. If you want to use the scientific method to study the running by falling hypothesis, then do a video analysis of where the center of gravity is when barefoot running. I suspect that there will be significant differences depending on a number of factors.
    2. I’m not certain that a difinitive conclusion can be made by looking just at vertical movement of the hips–what about the involvment of the upper body in establish a center of gravity of the entire body? There may be potential and kinetic energy components there that act in concert with the hip motions mentioned in the article.

    • Jason
      October 24, 2012

      Indeed. Video analysis is inherently flawed because it will only show external movement. It does nothing to tell us what’s going on internally. For example, what muscles are activated in what sequence? How much energy is being stored and released in tendons? When is that energy being released? In what direction?

      All are questions that cannot be answered via video analysis, yet have a profound impact on gait.

  6. Miss.R
    October 23, 2012

    Thanks for share

  7. Recorrido » Desabaf… Leituras de 3a Feira
    October 23, 2012

    […] Um pouco de corrida, gravidade, biomecânica… […]

  8. Rick Merriam
    October 22, 2012

    As always, thanks for posting Jason!

    In order to have an efficient/effortless propulsion, you have to have three planes of motion throughout the chain, and it all starts when the foot comes in complete contact with the ground. So, the retrograde forces between the foot and the ground play an integral part in the human chain’s ability to store elastic energy.

    These principles apply to heel striking and midfoot/forefoot striking.

    When the entire foot touches down for the initial contact, the ground is pushing up, and the front-foot only has one place to go which is up and away from the ground, i.e., inversion–a component of supination. Then, the calcaneus (heel), a component part of the rear-foot, is driven out which is an eversion of the calcaneus at the subtalar joint, i.e., pronation. In order for the body to come over the foot with little to no effort, the front-foot has to be supinating at the same time that the rear-foot is pronating at the subtalar joint. This is mainly due to the fact that the front-foot and the rear-foot are moving in opposite directions in order to store the appropriate amount of elastic energy that is needed to overcome gravity and lock-up the rear foot for an effective/explosive propulsion right before heel lift.

    Essentially, you have to have efficient motion of the subtalar joint in order to have quality motion from above at the ankle joint, e.g. shock absorption. When both joints are functioning optimally, the lower leg can then follow the talus down and in at the subtalar joint which allows for an eccentric lengthening of the posterior calf musculature in all three planes, e.g., the achilles tendon.

    Ultimately, the lower leg has to be driven in all three planes of motion in order to load and/or dissipate force. Example: the lower leg is driven in (medial/internal rotation), out (abduction) and forward (dorsiflexion).

    Then, the swing leg is free to move in the open chain, and the arms combine to drive movement at the right time, in the right plane and at the right joint from the top down. Essentially, muscles throughout the entire chain have to cooperate in order to provide the posterior calf musculature with the green light to unload and unleash the stored elastic energy just prior to heel lift.

  9. Ben W
    October 22, 2012

    Nothing to add here, but I will say thanks for showing how an internet discussion can be fruitful. I’m enjoying the information and the civility.

  10. Amy
    October 22, 2012

    This is quite complex for the ordinary person, but for a runner, this opens up more complex questions than simple answers. Many thanks for this article!

    • Jason
      October 24, 2012

      The more you know, the more you realize you don’t know much.

      I like topics like that. 🙂

  11. Lee Parker
    October 22, 2012

    Gravity may not be “propelling” me when I run, but it sure as hell encourages me to get my foot down before I face-plant. I also try Ken Bob’s (I believe) idea of not bobbing up and down which means my hips stay pretty much level – no or very little up and down movement. All that said, it still looks like a controlled fall to me. If my foot doesn’t come down when and where it should, I’m gonna hurt and you got me stumped if gravity isn’t to blame.

  12. Damian Stoy
    October 22, 2012

    Love the discussion! Keep it going for sharing knowledge is essential to progress. Personally, I’d like to see more discussion on the role of elastic coil and its role in movement.

    • Jason
      October 24, 2012

      Elastic recoil obviously plays a major role in gait, but I’m not aware of any research that has directly measured the effects. I *think* the amount of energy return has been calculated using cadavers, but I’m not sure that generalizes well to dynamic gait.

  13. Dave
    October 22, 2012

    I have to say that the engineer in me cringes when I hear about using gravity, a force which points straight down, as a way to propel yourself forward, 90 degrees perpendicular to the gravitational force vector. Any 1st year physics student will say this is hogwash; no empirical studies necessary. It falls under the same category theoretically as perpetual motion machines.

    Running downhill is a different story of course as you are moving in a vector with a vertical component, aligned partially with the gravitational force vector. Same argument with uphill running, except that gravity is of course working in the wrong direction here.

    THAT BEING SAID, leaning forward slightly is largely necessary to initiate acceleration. This lean is a temporary state which you will naturally auto-correct for, unless you have a propensity for face-planting. I.e. You will return to a (nearly) vertical alignment very quickly.

    Note also, that due to slight atmospheric drag, you will have to lean ever so slightly forward so as not to be ‘blown’ backwards. I.e. The force vector of air resistance will increase with speed. You will naturally lean forward such that the moment of rotation around your center of gravity is net 0 with air resistance trying to rotate you one way, and gravity the other.

    Viewing a runner moving from your left to right, air resistance will try to rotate them counter-clockwise (head pushed back), and gravity (due to a slight lean) will try to rotate you clockwise (nose towards ground). While running steady state, these forces will balance. BUT, gravity is in no way propelling you.

    • Jason
      October 24, 2012

      I have yet to meet an engineer or physicist that agrees with the “gravity propels/ running is falling” theory… which is a major source of my skepticism. 🙂

      I agree with the lean- it serves a utilitarian purpose… but fascilitating falling is not one of them.

  14. David Stretanski
    October 22, 2012

    Interesting post and that you assume ALL runners will use the same gait cycle you diagram. Your assumption/diagram implies ALL runners will bend the knee and “sink” their COM height in the first ~part of the contact phase.

    I disagree: It is very possible to go from touch down to mid-stance without “sinking” the COM but “falling” the COM forward/down by keeping the knee “soft” with limited and ~constant bend through this part of the gait cycle. The body/COM position at this mid-stance will be different of course and *forward* of the foot – which is very different from your COM at mid-stance *over* the foot. The subtle fall gait cycle allows a subtle fall in the first ~part of the contact phase, then a COM recovery in the second ~part of the contact phase. The difference is sinking under gravity vs. staying aligned to take advantage of gravity.

    I disagree that ChiRunning has neglected the issue of recovering center of mass (COM) height:

    This point and limiting knee “sink” was addressed here months ago:
    – This is a physics analysis.
    – Note the gait diagram posted here is not “opposite” of your diagram.

    And falling in terms of momentum addressed here months ago:
    – More physics analysis.

    I agree that if a runner uses your gait cycle as diagramed, they are not falling since their COM is rising when they have the opportunity (i.e COM in front of the foot) to fall forward. Their position negates their ability to use gravity as a propulsive force.

    I disagree that the gait cycle you diagram is required/used by all runners. I also disagree that COM recovery effort cancels out any possible gain from falling. The COM recovery can come from some combination of passive recoil and large muscles doing a relatively easy job.

    Interested in knowing why your gait cycle is required and what analysis you have done to conclude “that energy expenditure would cancel out any possible gain from falling, plus more”?


    • Jason
      October 22, 2012

      David, I agree with some points, but disagree with others. A few questions before I comment:
      – In your model you describe on your site, what muscles would be actived and at what time?
      – In your model, is there any vertical movement? In other words, if you measure a specific point on the torso, does it move up and down at all or does it remain perfectly horizontal (parrallel with the ground)?

      • David Stretanski
        October 23, 2012

        You could write a book on what muscles *might* be used when. Each person will be different of course. More appropriate for a blog post which I may do at some point.

        As my analysis blog post clearly states and shows, there is some vertical movement as the COM (and posture line) falls subtly forward/down, then recovers the minimal COM height loss for the next fall. How a person recovers COM height is a combination of passive and/or active effort; unique to the person, speed, terrain, etc., etc., etc.

        There is evidence on YouTube that you run with your hips at the lowest position *after* mid-stance. Consistent with other YouTube videos of you, I am looking at (1:14-1:15) and three still images I pulled from it:
        – The first one shows your touchdown with the “soft” knee. The heel is still up but a moment later you are down on the fullfoot.
        – The second one shows you vertical (shoulders over hips over ankles or in your terminology “mid-stance”) with the same “soft” knee at the same angle. There is little to no sinking from fullfoot (after touchdown) to mid-stance as in your model.
        – The third one shows your foot still completely on the ground, your posture line now tilted slightly forward with COM in front of support *AND* your knee is still “soft” at the same angle. Given the geometry, your COM (hips) is slightly lower; which this third image confirms against the background. So based on this video, your hips are lowest *after* mid-stance. This is very similar to my model/diagram and inconsistent with your model/diagram and your statement “At mid-stance, the hips are at the lowest position of the gait cycle.”

        Here are the three images left to right:

        One reason you are very unlikely to sink is your cadence. At a fast cadence (90 at the 1:14 point in the video) you have limited time to sink. People who sink tend to have a slower cadence closer to 80 because they are on their feet a long time; and therefore have time to sink. They also tend to reach more which puts them more on muscle from touchdown in front which continues as they move over their foot.

        If your feet/ankles are very relaxed as I believe you promote, your video suggests you are falling with each step. Here are four things needed to fall with each step:
        – Foot on the ground for friction at the base of your column.
        – A relaxed ankle hinge which does not resist the subtle fall.
        – The COM moving from over the ankle to in front of the ankle while the leg/knee angle stays ~constant. The duration of fall and angle is subtle, much less than most people think.
        – The minimal COM height loss is recovered as noted above. The video leaves clues as to how you are instinctively doing this through a combination of passive and active effort. But, only you can body sense how this is happening. Advanced ways to enhance this COM height recovery can be taught.

    • Ken S.
      October 22, 2012

      You did a good job of pointing out some of the things that were also bothering me about this explanation. Thanks!

    • Jason
      October 24, 2012

      Okay, finally got time to resopond.

      Regarding your article:

      – “Maintaining momentum” learning cue is good; I agree it’s a useful way to think of running.
      – Your formula- agree here, too. However, I would say gravity would be a “deceleration” in the formula.
      – I really like your use of describing running as reducing resistance.

      Regarding the numbered points:
      1. Agree.
      2. Agree, but the relative position of foot strike to COM varies with pace. For example, it would be possible to land too close to your COM. Same deal with cadence- it is possible to have a cadence that’s too fast.
      3. Agree about hinge resistance. It’s important to keep the swing leg relaxed. However, the big toe does “lock” in position immediately before the foot leaves the ground.
      4. Agree about hip mobility. I would argue this is necessary because hip extension is the primary action that propels us forward.
      5. Regarding arm swing: agreed.

      Your explanation and analysis of my 2009 vid reinforced a point Ken S. and I were moving toward… we’re describing the exact same actions using a different explanation of the phenomenon. The “falling” described by Chi/Pose is the exact same thing that occurs in the latter half of the arial phase of the spring model (as opposed to the “opposite” explanation I gave in the OP).

      In the diagram on your site, the body is at the highest point when both feet are on the ground. In the spring model, the same thing happens.

      In your model, the body is at the low point at approximately mid-stance. Same deal with spring model.

      This isn’t surprising since both models produce the same results… what I teach and what you teach look remarkably similar and have the same benefits and drawbacks.

      If we assume we’re talking about the same motions, the question then becomes *how* motion is achieved. Based on your diagram, the “falling” occurs between the picture on the far right (both feet off ground) to the middle picture, which would be the forward movement attributed to gravity. While gravity would play a role in forard movement, it’s limiting forward movement, not fascilitating it. That’s the point the engineers make- gravity works bertically, not horizontally. The forward motion is produced by the force generated between the middle picture and the far right picture. SOMETHING has to propel the runner forward and up.

      I think your model as described in the picture is a much more accuate representation of Chi (and Pose) than almost all people present. I’ve heard countless Chi and Pose coaches claim gravity is the force at work in that “middle picture to far right picture” phase, which would make the claim that gravity is causing your body to go move upward away from the ground. Of course, that idea is flat-out stupid… you can’t “fall” upward.

      Regarding your comment about people using different gait patterns- I agree. Most runners (the runners we’re trying to convert) run with a sped-up walking gait which is fundamentally different. I’d argue Chi and Pose runners use the same gait pattern as barefoot and other “natural” runners… which is what we both diagramed.

      Regarding the energy expenditure needed to lift the body off the ground versus energy “recovered” vai gravity- that’s basic physics. I’m NOT an expert by any means, I’m just relying on what smarter people have told me: the energy required to escape gravity is always going to be greater than any energy that can be saved with gravity. I’ll have to rely on others to conform this.

      Logic can confirm it, though. If the falling saved enough energy for it to be a neet-zero proposition, perpetual motion would be possible here on Earth.

      A more applicable explanation- if it were a net-zero formula, running would require zero energy. Anyone that’s run a long distance knows that’s not the case. 😉

      The question of muscle use is a topic that fascinates me- what muscles are activated at various parts of the gait cycle. It’s obvious different runners use drastically different activation patterns. The question- is one pattern better than others, and does that pattern generalize to all runners? Is the optimum affected by terrain, environmental conditions, or shoes that alter motion and geometry? Will that pattern develop on its own based on the brain always working toward peak efficiency? Can we teach it or does our attempt to teach it interrupt the natural gravitation toward peak efficiency? In fact, that will likely be an entirely new post.

      • David Stretanski
        October 24, 2012

        Many of your comments are covered in my posts so I am not going to restate them again. I think you said it best in another comment … “It’s how I learn from others since I’m generally too lazy to read and too dumb to comprehend the more formal sources of knowledge.” I am shocked that a fellow coach serious about helping others would take this approach.

        I have just four comments of the many I could make:

        1. You stated “However, I would say gravity would be a ‘deceleration’ in the formula.” So you are saying gravity can decelerate you in the horizontal plane but it cannot accelerate you in the horizontal plane? I thought your position was the gravity was a vertical force and does not effect things horizontally? Please explain.

        2. Falling when your foot is on the ground is falling *forward*. Falling through the aerial phase with foot off the ground is falling *down*. These are two completely different physics scenarios at different times in the gait cycle. There is no possible way that “we’re describing the exact same actions using a different explanation of the phenomenon.”

        2b. In my falling model, the COM is at the low point after mid-stance. This is what your video shows you actually do when you run. The sinking model you presented in your post is that the COM is at the low point at mid-stance (hips over ankles). Per your video, this is not what you do.

        3. I challenge you to provide your complete free body diagram analysis proving that gravity cannot pull a column down and forward. A simple analysis of how a tree cannot fall over (meaning down *and* forward) will work. Or you can use the column model with conditions/assumptions for falling as I have here:

        Just saying “gravity is a vertical force” or “gravity works vertically, not horizontally” is not a full physics analysis. It only considers one force in a possible system of forces. A full physics analysis determines the *net* force on an object considering all the forces on the object. Only the *net* force will determine acceleration/motion moment by moment. The direction of individual forces may or may not be the direction of acceleration/motion. This is “basic physics”.

        4. I have not suggested it is a net-zero proposition or formula. There is no perpetual motion here. It is simply a question of adjusting effort to utilize a force of nature to some degree.

        • Jason
          October 24, 2012

          I’m pretty sure a falling column is not a ‘full physics analysis” of running gait, either, but I’ll let the readers decide. They’re a smart bunch.

          Regarding gravity- my point was that it’s something that has to be overcome. As such, it is a ‘loss” in your formula, not something that can be used as a “gain.”

          And see my comments on the new post. 🙂

  15. briderdt
    October 22, 2012

    What so many of these things ignore is resolution of moments about one’s center of gravity, and the role that push-back plays into a smooth gait. The foot can indeed land under the CG and maintain a smooth gait without falling BECAUSE of push-back during contact phase of the stride. The body constantly rotate like a pendulum about its own CG, and this acceleration is generated by a complex sequence of muscle actions, but it allows the push-back to work and keep the CG moving at a steady speed and with little vertical movement. Models like these take ALL of the rotation and moment resolutions out of it in the name of simplification, and in doing so lose the connection to the real world.

    • Dave
      October 22, 2012

      Right, assuming no air resistance, running on level ground, and maintaining a steady pace: runners alternate between a state of having ground contact slightly in front of CG (initial contact) to having ground contact slightly behind CG (toe off). If you average the location of CG relative to place of ground contact, you will find CG EXACTLY above the point of ground contact. Air resistance or acceleration moves the CG slightly forward of this point.

      • briderdt
        October 22, 2012

        Again, this neglects the resolution of moments about the CG and the effect of pushing backwards with the foot. But I guess the dogma means more than physics.

        • Dave
          October 22, 2012

          Without air resistance, there cannot by definition be any force exerted backwards by the foot unless one is accelerating. If we put air resistance back into the equation, you do indeed push backwards slightly, and as you imply, this will create a moment of inertia that needs to be counteracted by an opposing moment created from gravity and a slight forward lean. This backwards (accelerating) push will exert a force exactly balancing the force exerted by wind resistance.

          I.e. In a perfect non-resistance setting, your body will not lean forward, but in the real world a slight lean will be necessary to maintain forward momentum. I am not sure however how much lean is needed to do this – would it be noticeable?

  16. Bill
    October 22, 2012

    Hi Jason,

    I liked this article. My take away is this. It is important to understand the basic principles behind barefoot running. For me, what I have found helpful is trying to determine how these principles apply to my physique. I’m a short, stocky guy, about 5’4″ tall. I think there may be subtle differences between say me, and a someone who is 6’4″ tall with long gangly limbs. Would you agree? Just a thought.

    • Jason
      October 24, 2012

      It’s a good question. I think the fundamental principles are the same, they just look a little different to the outside observer. Individual “customization” will always make gait look slightly different in different individuals.

  17. Ken S.
    October 22, 2012

    Hi Jason,

    There are some problems with your explanation here. However, as you have pointed out, I don’t think we are really all that far apart in our thinking.

    I really would like to have a chance to hash out a discuss our ideas and our differences. I believe, at this point, we are mostly divided by a common language.

    We could do a lot for the running community if we (and a few others) discussed these differences in disciplined way based on mutual respect. We might not settle anything, but I think we could potentially set precedent for more intelligent and thoughtful discussion.

    What do you say?

  18. Frances
    October 22, 2012

    There is a similar fight that goes on these days over in the world of teaching voice. The muscles and mechanisms that coordinate to produce the singing voice are all INSIDE and way less observable than the muscles used for running, so the lore that builds up around how to sing properly gets even more vague and mysterious.

    But in recent times, vocal scientists are studying these mechanisms and bringing new ways of talking about the actions of the muscles and the breathing apparatus. A lot of the teachers who use imagery and expressions from the history of teaching voice in the past are resistant to the science. They use expressions and ideas that may be useful to helping a voice student get a result, but from a scientific standpoint are inaccurate, misleading, and confusing. A certain type of student might get very confused when told to do certain things and there are some kind of people who need the science to help them understand what they’re aiming for.

    Sometimes the scientist’s ability to explain what is happening, does not help achieve the results. Sometimes the folk wisdom and intuitive kind of way of getting someone to do something can work better, and then the scientist can go ahead and tell you why that works.

    I think running, like singing, has aspects of art and science to it. There is a mystery component that science can’t quite get a handle on, and that’s when the intention and feelings of the individual has to come into play and figure out what science tells us on a more personal level.

  19. trissa
    October 22, 2012

    OH! Here’s a thought:
    What does body TYPE have to do with efficiency?

    I noticed that those who excel at running SEEM to have a consistent body type. While there are, of course, exceptions to this, overall, it appears those who receive the award money in big races are “built” the same..

    Is it training/technique….or body type?
    All of the above? Born in a country NOT known for wearing “Foot Coffins”????

    • Jack Harris
      October 22, 2012

      I like this question, but do you mean body type as in body fat distribution or as on longer/shorter limbs? You must also remember that correlation does not equal causation. Is a person faster because they are lean, leaner because they spend a lot of time running, or is it a third factor that causes both?

  20. Kyle Kranz
    October 22, 2012

    Something that stuck out to me from Ken’s website is that he says the Pose runner does use energy to get back into the Pose (thus pushing his hips back up) however the Pose runner does not use energy to run forward, as they are simply falling forward. Pose runners do of course push, but they push up to get into Pose, not to move forward.

    You basically said the same thing Jason, but I really like how Ken puts it, from a Pose runner’s POV.

    • Jason
      October 22, 2012

      Ken’s writing is awesome. As he’s noted, Romanov probably encountered significant language barriers when writing about Pose. I believe Ken’s descriptions of Pose align that theory with other theories (like the spring model) much better than any other writing I’ve read. The take-away: Ken should be the official Pose America spokesperson. Every Pose coach should read his thoughts… it would clear up A LOT of misinformation.

      • Ken S.
        October 22, 2012

        Thank you to both of you.

    • Ken S.
      October 22, 2012

      Thanks! But actually the runner does use energy to move forward. It’s in the form of potential energy being converted to kinetic energy. Potential energy is supplied prior to falling by the runner. Just to clarify 🙂

      • Kyle Kranz
        October 24, 2012

        Right on. I could have clarified better.

  21. trissa
    October 22, 2012

    Your most cogent article to date.

    yeah — I think at heart you are a science nerd, hence the bent towards “stirring the pot.”

    🙂 Tris