Blog : News



Break Down of Energetic Cost for Each Component of Running

About a month ago at the Running Medicine Conference at the University of Virginia, I had the pleasure of listening to a presentation by Rodger Kram, PhD; a researcher at the Locomotion Lab at the University of Colorado. Life has been busy lately so it’s taken me a while to get around to writing this. (Thanks to Dr. Kram for permission to use pics and reviewing this for me before I posted it)

Dr. Kram frequently publishes research papers in journals but the topic of his presentation at the Running Medicine Conference was on “Disintegrating the Energetic Cost of Running”. In this case, disintegrating means to break down into its smaller parts. Basically, through a number of different studies, he and his students and colleagues have been able to look into how much each different task of running costs us – energetically speaking.
Table 1 outlines the basic major components of running:


In order to find out what the energetic cost of each component of running was, Dr. Kram and his grad students would measure various runners to find out what their energy consumption was normally, then invent some sort of apparatus to eliminate the energy cost of each of the factors listed in table 1. So, for example, to eliminate the energy cost of body weight support, they constructed this apparatus:

Body weight picture

They found that as they progressively offloaded the vertical component of body weight, the runners used about 26% of the energy compared to when they had to run with 100% of their body weight. In other words, holding up your body weight accounts for 74% of the energy of running. The study is found here.

In another study, they eliminated the cost of forward propulsion by fabricating this gizmo:

Forward Propulsion

They found that by eliminating the work needed to propel the body forward, the runners used up to 41% less energy. In other words, 41% of the energy for running is used for forward propulsion. The study is found here.

To eliminate the energy needed to swing the leg forward for each stride, they constructed this:

leg swing assist

They discovered that the action of swinging the leg forward for each stride accounted for about 20% of the energy cost of running. The study is found here.

Now, if you’ve been paying attention up to now, you should have realized something.


Obviously, there is a problem when you have just a few of the components add up to 135%. As Dr. Kram explains, that is due to synergy. In other words, there is a cooperative action of two or more movements, muscles, nerves, fascia etc.

In order to see how much synergy there was in these three tasks, they combined the vertical body weight support, the forward propulsion and the leg swing assist all in one contraption rivaling the Death Star:

All combined assistance

They did a bunch of trials with different combinations and eventually figured out that because of synergy, the vertical body weight support, forward propulsion and leg swing accounted for 87% of the energy used for running. The study is found here.

table with 3 combined

Next on the list was arm swinging. In this case, Dr. Kram had a different hypothesis. The way he figured it, swinging the arms actually assists the energy of running by counter-rotating the upper body. If we didn’t swing our arms, it would cost more energy to run. So, they ran normally, then with the arms behind the back, folded across the chest and hands clasped behind the neck.

no arm swing running

In the end, they found that they were correct – eliminating the arm swing increased the energy cost by about 4%. The study is found here.

table with arm swing

At that point, Dr. Kram and his grad students came up with a bizarre notion: Since the arms weigh about 10% of the body weight, and propelling the body weight forward and lifting it vertically costs energy, it costs a certain amount of energy to have the arms on your body. Since swinging the arms saves 4% of the energy cost, but the weight of the arms costs you about 9% of your energy, you could potentially save about 5% of your energy by amputating your arms. So, they paid a grad student…….just kidding!

Moving on, Dr. Kram and colleagues have assumed that running with the feet narrow, (not zero width or cross-over) costs less energy and helps balance when compared to running with a wider base. So, they had runners run on a treadmill with varying step widths and measured oxygen uptake again:

step width

In the end, they found that when runners ran with the feet at widths less than or greater than their preferred step width, there was greater metabolic demand. In other words, they concluded that step width costs zero energy. This could be based on the idea that running with an unfamiliar gait pattern increases metabolic cost. The study is found here.

table with step width

The next factor for energy cost was balance. Balance competency is highly variable. We use this as a movement screening tool in our office and it’s surprising how come runners are unable to perform a single leg stance for more than a few seconds. They are moving their arms around, the foot is moving all over the place and then they have to put the other foot down. Well, how much energy does it cost in running to simply balance yourself? To answer that, Dr. Kram and colleagues hooked runners into the following contraption:

Lateral stabilization picture

Pretty simple apparatus, yet ingenious really. They developed something that stops lateral swaying, yet allows the runners to swing their arms freely. After measuring the metabolic cost without the apparatus and then with it, they found that there was about a 2% energy savings when the apparatus was used. That is to say, balance when running on a very flat surface constitutes 2% of the total energy expenditure. It may be greater when running on trails or uneven terrain. The study is found here.

To summarize up to here, we have the following table:table with balance

So, next on the agenda is shoes. Obviously a contentious topic these days. At this point of his presentation, Dr. Kram recapped a 1984 study in which found that for every 100 grams of shoes weight, there was a 1% increase in metabolic cost. However, there are many other studies that have not reproduced these results; for better or worse:

shoe energy costs from different studies

This may be because when compared to shod running, barefoot running reduces energy cost due to less weight on the foot, OR, running barefoot increases metabolic cost due to a lack of shock absorption. Nobody knew for sure. So Dr. Kram’s student team set out to find the answer. They performed a number of trials with subjects running barefoot or shod, plus running barefoot or shod with 150, 300 or 450 gram lead strips attached to the barefoot or shoe. When they measured O2 intake and CO2 production and calculated metabolic cost, they discovered that when given equal mass between barefoot or shoes, shoes save 3-4% in metabolic power. WHY? Well, we don’t know for sure, but the hypothesis is that shoes provide shock absorption and without it, your legs have to absorb the shock. Absorbing shock with your legs costs energy. Study found here.

table with shoes

So, the best case scenario is as follows: find the lightest shoes that provide ample cushioning (from an energy perspective) Why is that? Well, because as they found out, mass is bad but cushioning is good…to a point (as you will see below).

In this last study that Dr. Kram discussed, they wanted to measure the energetic costs of the surface you run on. To accomplish this, they used a very rigid surfaced treadmill with a steel deck and no give. Then they put slats of foam that are used for shoe cushioning and modified the treadmill to look like this:

foam cushioned treadmill

The researchers had experienced barefoot runners run on the hard steel deck, and then run barefoot with 20mm of cushioning, then with 10mm of cushioning. What they found was there was a sort of “sweet spot” of cushioning for reducing metabolic energy. 10mm was better than no cushioning, but 20mm was too much cushioning and energy expenditure increased.

The surface you’re running on can vary energy consumption greatly. For example, Lejeune et al., (1998) found that running on sand costs an extra 60% compared to running on a hard surface, yet Kerdok et al., (2002) found that with the right amount of compliance, the surface can actually decrease energy consumption by as much as 12%.

table with final combined

So, the surface you’re running on could be one of the highest costs of energy when running, or could actually decrease the cost of energy.

That concluded the end of Dr. Kram’s presentation, but he left us with some thoughts….

  1. They can’t account for 100% of metabolic costs of running because there are so many variable factors.
  2. You can account for nearly 90% of the metabolic costs with 3 factors: vertical body weight support, forward propulsion and leg swing
  3. Arm swing actually saves energy, shoes don’t make that much of a difference, but surfaces can make a huge difference
  4. There are many factors that can account for the total metabolic costs such as the horizontal braking forces, ventilation (breathing), cardiac work, and decelerating and reversing the forward leg swing, they just haven’t developed a way to measure those factors yet


- Active Spine and Sport, ,

1 Comment

  • Thoughts on minimalist running | Are we any closer? Reply

    May 19, 2013 at 7:01 am

    […] thing that absolutely do not boil down to the shoe you are wearing (or are not wearing). In fact, a very clever study, albeit a bit contrived, recently showed that ~80% of running efficiency was derived from vertical […]

Leave a Reply

Your email address will not be published. Required fields are marked *