Don’t run with gravity

In the past years several schools of thought like ChiRunning and Pose-Running have gained some traction especially amongst recreational runners. These methods are promoting the idea of gravity doing propulsive work in running, amongst several other claims and proposals regarding running mechanics.

Many runners and trainers have written positive articles regarding these techniques (e.g. in Proloop) whereas many others have been fairly critical. However these critiques seldom use logic and physics, but personal arguments or arguments like “there are no new running techniques because everything already exists in the minds and works of the great coaches of the past”.

In this article we want to give a sound and clear refutation based on the basic mechanical claims of the gravity running models in question. We acknowledge that not every technique instruction given to a runner has to be a physically true statement, but can also help as a psychological cue. However, deriving such instructions from a flawed mechanical model of running is not sensible and may even be dangerous to the health of the runners.

We will point our attention to three popular statements, which also form the core of the running technique models in question. Refuting any of these statements implies the refutation of the whole model.

1) Gravity does propulsive work in running
2) A runner should land below or slightly behind the center of mass
3) The body’s forward lean determines speed

Ad 1) Gravity does propulsive work in running

Work is a fixed term from physics and is defined as force acting on a body over a certain time. Positive work results in a raised energy level, which generally means a higher velocity when we talk about running. The word “propulsive” indicates that we are talking about the horizontal movement. So in essence gravity doing propulsive work in running means that gravity is the driving force increasing running velocity.

However, from a mechanical point of view this is impossible. Gravity acts in a vertical direction and therefore can do work on a body in only two ways:

a) It can increase its vertical downward velocity.
b) If the body is fixed at a certain point (like the foot standing on the ground) which is not exactly on gravity’s line of force then gravity causes a rotation about the fixation point.

We don’t have to consider the first option any further, because this is only about vertical movement. The second option however is claimed to be the driving force behind a runner running forward and is sometimes titled “gravitational torque”. However, the reality is that we actually create a counteracting torque by our own muscular effort to NOT let gravity rotate us in any direction. This is the only way to keep running with a stable body posture and to prevent stumbling. If gravity would do net rotational work with each step it wouldn’t be long before we would do somersaults or just fall flat on our face. This applies to constant speed as well as to accelerated running.

Ad 2) A runner should land below or slightly behind the center of mass

Landing below or even behind the center of mass is mechanically impossible during constant speed running. For the sake of argument let’s suppose a) you run at constant speed and b) land exactly under your center of mass.

From this landing position gravity will cause a forward oriented rotation of the body until toe-off. As mentioned in 1) we have to actively counteract this torque to maintain the cyclical running pattern. To do so, the ground reaction force vector needs to be in front of the center of mass. Remember that our assumption b) was to land exactly below the center of mass? Well then the consequence is that we will create a forward directed ground reaction force during the whole stance phase. In other words we are accelerating with each step.

Wasn’t our assumption a) to run at constant speed? The deduction of what it means to run at constant speed while landing below the center of mass leads us to a contradiction with our own assumptions. Landing below or behind the center of mass is simply not possible at constant speed running, and only for very few steps during the most extreme accelerated running.

Ad 3) The body’s forward lean determines speed

Body lean is defined as the angle between the vertical and the line connecting the foot’s point of contact with the center of mass. ChiRunning, Pose-Running and other theories refer to this angle at or shortly before toe-off as “body lean”. The claim is that you run faster by increasing this lean angle (in spirit of the idea refuted above that it is the gravitational torque which creates forward movement).

It is very simple to demonstrate the lack of a causal relation between the body’s forward lean angle and running velocity. The longer you stay on the ground at a certain speed, the bigger this angle will be, because the body travels further during ground contact. This shows that the body lean angle can be altered independently of running speed just by modulating ground contact time, which in return demonstrates that there is no causal relation.

However, it should be mentioned that during accelerated running this angle really is important, because you can create greater horizontal forces in more forward leaning body positions (see “Body position determines propulsive forces in accelerated running”).


In summary theories like Pose-Running, ChiRunning and all other running models with similar ideas start from a mechanically false set of premises. Deriving instructions from false premises often results in asking runners to perform movements which are mechanically impossible. From our broad experience in recreational and elite running in different sports and with athletes of different age groups we know that for some these instructions still were helpful. However, we have also seen many people struggling with their practical implications and even worse, injuring themselves while trying to implement them into their running movement.

If the basic premises of a theory are incorrect, the whole theory with all its implications has to be rejected and reassessed. Hence, we highly recommend to reconsider running technique instructions on the basis of basic mechanical laws and logics. Don’t be fooled by gravity.

Florian Kugler & Ruben Jongkind, February 17th, 2011

Recent articles