Acknowledging previous observations to be correct. Here's a physicist's perspective and an interesting proposed experiment to try
The acceleration sensors in the rS only measure acceleration at the site where the rS is mounted i.e. the foot. The magnitude of the acceleration is reported back to us in units of G's which are 9.80665 meters per second squared.
During running when the foot first strikes the ground there it undergoes an unavoidable and substantial upward acceleration as it's vertical descent is halted abruptly. So the rS captures that part. And there is a force associated with that acceleration which you could approximate as (F = mass of foot x G's).
... my foot ought to have about 2 pounds mass, and so my typical rS impact G's of 10-12 means the foot at impact is seeing momentary 20 to 24 pounds of force. As soon as it's vertical motion stops, that force drops to zero, but the upper body is still decelerating, albeit in a slower better cushioned fashion, and is bearing down on the foot via the ankle joint.
...following the impact, as the runners body goes forward and the the full momentum of the upper portions of the body come to a vertical standstill (which would be the lowest dip point of the hips if viewing a slow motion video) the maximum force is exerted by the body through the stationary foot to the ground. This force is referred to as the 'ground reaction force' and is frequently measured and reported in units of ' x body weight' (e.g. 1.6 x Body Weight) which is really G's again in disguise. Body weight = G's x body mass. But in normal English we don't distinguish between mass and weight because there hasn't been a need (up to now anyway) because on the earth's surface they are identical by definition. On the moon, they ain't.
Ground reaction force is pretty important in running (if for no other reason than we'd like both feet to share the load, see the same forces), and it can be and has been measured at least two ways: 1. By using a treadmill equipped with a set of force sensors under the plate, or 2. By using running shoes equipped with force sensors in the sole (like Sensoria or others). Excessive ground reaction forces can lead to injury, and can to some extent be reduced by gait training.
OK, I'm getting to the point now. Ground reaction force can be estimated if we just measure the acceleration of the runner's torso, determine the maximum vertical deceleration (in G's) and multiply that by body weight. We could probably set your rS to 'heel mount', strap it to your lower back, jump up and down to make sure it started taking data, and go off running for a couple of minutes. Look at the data afterwards, approximate ground reaction force should be groundF = (Impact G''s x Body Weight).
I have not done this, and I promise to try it.
I have done it with another stand-alone accelerometer with reasonable results.