I came to running in triathlon in 1980 as a swimmer first.  In swimming I had learned that technique is everything, so naturally I started trying to find out “how” to run.   However, in the early 1980s hardly anyone considered how you ran as important, so I didn’t have much luck and just experimented.   This led to the inevitable failures and successes but no unifying approach to carry me forward over time ever emerged.

Flash forward to 1996 and the beginning of the USA Triathlon National Teams program which I began as USAT’s first National Teams Coach.   I knew my primary weakness in this job was my ability to improve running technique.   So being the type of athlete and coach who recognizes that you have to develop your weaknesses, and not just rely on your strengths, I made this my primary focus for new learning.   By good fortune, Nicholas Romanov became known to me and we began to communicate.  First I simply tried the simple ideas he was promoting myself with very rapid positive results.   The dominnent things I can remember were that I could suddenly run much faster and that my quads stopped being destroyed by longer runs.   Next we brought him out to Colorado Springs to learn his concept, the Pose Method, first hand and ultimately the basic ideas became the basis for how we coached running in the National Teams program for years.   At its simplest, the Pose Method, Chi Running, and barefoot running all hone in on the same basic idea – that many runners over stride as result of recovering the leg using relatively more hip flexion and less knee flexion (by swinging the whole leg forward like a pendulum), which results in projecting the lower leg and foot well forward of the center of mass at ground contact.   This prevents the runner from absorbing energy elastically with maximum efficiency and creates reactive forces directed upwards through the body initially, rather than reactive forces that propel the body forward, a concept sometimes called braking.   Eventually the full foot does come down to the ground and some energy is certainly absorbed elastically and reactive forces eventually do  become propelling, so these differences are small and hard to see at normal running speeds.

When running without shoes over hard surfaces our bodies revert to a different approach.   Rather than using a pendulum from the hip to return the leg we flex the hip and knee simultaneously and to the same degree (like marching versus walking).  This action lifts the foot in a higher and shorter arc upwards towards the rear end (butt kicking) and results in the foot landing on the ground with the lower leg at a more neutral or rearward angle, and closer to the center of mass, avoiding the phenomenon of “braking” and allowing the midfoot to engage the ground first, followed later by the heel as the ankle dorsiflexes.  Our bodies are clearly designed to land on the midfoot, which then engages the plantar reflex earlier as well as making full use of the ankle as a second class lever to receive and return the forces of gravity which both drive us to the ground with each step and can then be returned to propel us forward elastically.   Essentially, the Pose Method and Chi running are systems to teach us how to do this, should we not be able to simply start running barefoot on hard surfaces all the time.

The simplest biomechanical outcomes of such a change in running technique are simply stride rate and frequency. When we take steps without overstriding, the steps become shorter and must be taken more quickly at a given velocity of running.  Now to the science.

In my research and statistics class my students develop and carry out original research projects every semester as I believe and practice experiential learning – learn by doing. We don’t put these projects through IRB review or publish them – they serve simply as a methodological learning technique.   In one of my favorite projects over the years we tested barefoot versus shod running over 100 meters on a turf football field to see which was fastest (hand timed), using a within comparisons approach – each student ran in both conditions with order randomized.   Further, we counted steps and then used simple calculations to determine stride rate and length.   The result were that students ran significantly faster barefoot (about 3/10ths of a second for 100 meters) using a higher stride frequency and shorter step length than when running shod. Astoundingly, such basic science on this elemental question has never been done or published, at least to my knowledge.   This study suggests we can run faster barefoot when the surface for running is not inhibitory to doing so.  In so doing our mechanics change by increasing stride frequency and taking shorter steps.

As I described in my first post to this blog, sport science rarely creates new advances in technique or training.    Rather, it usually just helps us to understand why they do or do not work.    The focus in sport science, to its detriment, is quite often only on the mechanism (how something works), and not on the outcome (how do you run fastest in this instance).   The largest example of this is the huge body of literature on running economy.   Running economy (versus mechanical efficiency) is simply the idea of measuring how much oxygen is required to run at a given steady state pace.  In activities like cycling we measure mechanical efficiency – how much oxygen is used to perform a given rate of work or power, because we can directly measure work.   But until recently, rate of work or power could not be measured in running, largely due to the fact that running employs elasticity to a large degree to achieve the outcome – how fast you run.  Consequently measuring economy (oxygen cost at a constant running velocity) is a default position.

We think economy is very important to running success – largely because running economy helps explain performance differences amongst homogeneous runners of high ability.  Here is the classic study on this topic: https://www.researchgate.net/profile/Gary_Krahenbuhl/publication/15744615_Running_economy_and_distance_running_performance_of_highly_trained_athletes/links/59d698f8aca27213df9e81b4/Running-economy-and-distance-running-performance-of-highly-trained-athletes.pdf

In addition, a long history of sport scientists manipulating stride frequency and stride length, going back into the 1970s,  has consistently shown that increasing someone’s stride frequency and reducing stride  length, in comparison to their freely chosen approach, increases oxygen consumption (makes economy worse).   Here is the most recent review: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4887549/

Hence, the conclusion in sport science has long been:  “Don’t change someone’s running technique”.  But there is a problem – nobody ever measured the effect of these changes on performance.

When we started using Pose Method performance principles with our currently overstriding elite triathletes, who then inevitably increased their stride frequency and reduced stride length at a given velocity, we nearly always saw their running performance improve.   Two notable cases come to mind.  The first, Hunter Kemper, was a notably overstriding and fairly mediocre NCAA Division I runner his sophomore year when I began to coach him as part of our collegiate program.   You will find he was near the front in the world junior triathlon championships that year off the bike, and then fell back to fifth or so after running about 35 minutes for 10K.   However, Hunter has high motor ability, and with some basic cueing, he returned the next year with a far more Pose like stride and rapidly became a 30 minute 10K runner and conference champion.  By 2005, he was considered the dominant runner in elite triathlon worldwide.  The second, Susan Bartholomew (now Susan Williams), came to elite triathlon following her NCAA swimming career and with an even more pronounced overstride as well.   Similar changes were certainly part of her improved running which later resulted in her becoming the first Olympic medalist in triathlon for the United States.

It was with this mindset that I set out to do my dissertation research examining the effect of the Pose method on biomechanics and economy found here: https://www.ncbi.nlm.nih.gov/pubmed/16195026

This particular study is often referenced by those who continue to choose to accept the “don’t change running mechanics” paradigm because we found that upon increasing stride frequency and reducing stride length in our Pose Method treatment group, as a result of 12 weeks instruction, running economy became worse.  Of course this study can easily be interpreted as simply another study showing that changing away from selected stride frequency and length increases oxygen consumption and makes economy worse.  Further, as a dissertation student, I accepted the instruction of my mentor to “not bother with performance measures”.  However, the problem was that the study subjects in the treatment group kept telling us about anecdotal incidences of performance improvement.   This was so prevalent that I felt compelled to formally survey them after the study about their intent to keep running in the “new way” even knowing that the result of the study suggested that their changes would hinder their performance.  All but one subject in the treatment group said they would continue with the changes, in spite of the study results, citing their own believe that they could run faster in the new way.

I had experienced my own profound running performance gains using this approach so I also never even considered changing back once the study was done.   However, the outcome of the study always troubled me; although I attempted to publish it without bias in any case.

Fast forward once again to more recent times and several things have happened which help me to understand the conflict this work created – how could one become less economic yet able to perform better with a change in running technique that resulted in increased stride frequency?   First, one of my colleagues on the Pose Method study, Graham Fletcher, later published similar work looking at performance, which showed that Pose Method changes do result in relatively short order running performance improvements, as we had consistently experienced. That study is found here: https://journals.sagepub.com/doi/abs/10.1260/174795408786238506

The second development is that we now have technology which purports to measure the power associated with running, although still not directly.    Finally, a pair of physicists and runners using this technology has recently shown that increasing stride frequency and reducing stride length results in a reduced running work rate and an increased oxygen cost at a given running velocity, although they have not yet formally published this work.  A summary can be found here:  https://blog.stryd.com/2019/12/06/the-impact-of-cadence-on-the-running-economy/

Of course by reducing the mechanical work of running at a given velocity by increasing stride rate, possibly as a result of an improved elastic energy return created by recovering the leg in a more spring like manner and landing more efficiently on the ball of the foot, we may be offsetting any negative effect from an increase in oxygen consumption, even in the short term, resulting in better running performance.   Their work does not answer that question, but it does help to explain our outcome and offer a way by which individuals whose economy was clearly worse after changing technique could still be running faster.    Further, with time, the most central way to improve economy begins to work – that being repetition.

One can find another example of this concept in cycling.  While virtually every study of cycling economy shows that pedaling a larger gear at a slower rate reduces oxygen cost, all the way down to about 60 RPMs, no elite cyclist of any note in modern times (lets say the last 30 years) has ever created a new standard in a typical time trial using these kinds of slow cadences.   In fact cadences of 90-100 RPM are more likely to occur.  This recent review highlights the use of increasingly higher cadences as power outputs increase over shorter time trials.

https://www.ingentaconnect.com/content/sabinet/ismj/2009/00000010/00000001/art0000

When we pedal more slowly using higher forces to produce a given rate of work we not only recruit more fatiguing fast twitch muscle fibers, we may also inhibit the interplay between propelling  and opposing forces created at the pedal itself.   Consequently, all competitive triathletes should strive to “learn” to pedal at higher cadences to allow for the possibility of creating higher power outputs and faster time trial performances.

As a practical example of this concept I began my cycling career in 1980 at a time when the 40K national record was around 54 minutes (it was in fact set in my first ever cycling race by Tom Sain, at the Arizona District Time Trial championships).   Over the next couple of decades Tom Doughty lowered the record to 52+ minutes using a huge gear and pedaling at ~60 RPMS.   Following this a long period of record swapping between Kent Bostick (also a U. of Arizona grad) and John Frey (from up the road in Albuquerque) ensued which ultimately dropped the record into the 47 minute range.   In observing this progression (and having had occasion to sit on the wheels of both) the lasting impression left in my mind was how the huge gear slow pedaling style employed by both while in the 54 minute range was ultimately replaced by a smaller gear higher cadence approach as the records got faster and faster eventually reaching 47 minutes.   I will note that the introduction of aerobars also greatly influenced this progression.

My conclusion is that making changes in the running technique of overstriding triathletes, in line with barefoot running, Pose Method and Chi running principles, will increase stride rate at a given velocity and facilitate faster running performance.  This is generally applicable to athletes whose stride frequency is between 60-85 strides per minute during steady aerobic running initially.

As to the method of doing so, great care has to be exercised if a barefoot running approach is used, although this is the simplest approach.   I suggest artificial turf, very limited initial running without shoes and a very slow progression.   As most modern people have spent nearly all of their lives in shoes, the idea that they could simply shuck them and run their normal volumes on typical road and concrete surfaces, without some injury occurring, seems improbable.   When using Chi or Pose method approaches, similar caution should be used to avoid large amounts of calf soreness and or injury initially.   Over time a stride frequency of 90-100 strides per minute at steady aerobic efforts should be the target.

I can further conclude that increasing the ability to pedal at higher cadences in cycling will facilitate a similar potential for performance improvement in that sport as well when one is not able to sustain cadences of 90-100 rpm comfortably.  This is easily accomplished by simply introducing cadence feedback and gradually progressing the maximum cadence at which one begins to “bounce” in the saddle in short efforts, while gradually increasing cadence targets for more extended riding.