Armswing#

Tue, 16 Jan 2007 14:08:54 +0000

Being a summary of the effective speedskating armswing, and some flailing around to attempt to explain how it works. Let's start with the conclusion.

• An effective skating armswing is approximately front to back. The arm at its foremost position should have bent elbow and hand somewhere in front of nose: it then travels backwards and outwards until the arm is stretched out straight somewhere behind you

• there should not be so much lateral movement that the shoulders rotate: this is tiring and doesn't achieve much

• the arm should brush the torso as it passes

• swinging arms faster will tend to make the legs go faster (this is often stated as "the arms drive the legs") - good for starts, accelerations, anything else where change of cadence is required

• swinging arms whn you don't need to is tiring.

• armswing on uphills may be a bit more lateral (but the shoulders caveat still applies)

This much is standard, and empirically we know it works, but the interesting question is "why?". If you disagree with the conclusion you are unlikely to be interested in the speculation which now follows.

Let's get the standard rant out of the way first: Newton's Third Law won't help. Over the course of a whole stride, throwing your arms in one direction will not make you go faster in the other direction unless in doing so you sever them. Otherwise, you have to bring them back towards you: the negation of the move you previously made, which at that time will make you go slower in the other direction.

Then what? A simple mechanical explanation, also, does not suffice. We are swinging our arms with a largely fore-aft motion apparently to counterbalance a lateral motion in the legs? Does not compute. I have speculated in the past that the small lateral component helps somewhat, and indeed it must have some kind of effect, but if that was all there is to it then a casual gentle armswing would be just as effective as a full-on sprint armswing - after all, the lateral motion is more or less the same - and we know that's not the case. It also doesn't explain the connection between armswing speed and leg cadence.

So I think we have to look at THE BRAIN. And the supposition is that as we evolved from quadrapeds we have vestigial stuff in our neural system to enable four-legged locomotion without falling over: specifically, that the timing and power used in a foreleg stride will affect and be affected by the timing and power in a hind leg stride. So when your front feet - er, arms - break into a trot, your back feet will follow.

One possible mechanism for this is the Central pattern generator - neural networks in the spnal cord that produce "rhythmic patterned outputs". They control the flight of the locust and the wiggle of the lamprey, and are claimed to be present in humans too. Another is good old-fashioned muscle memory - which basically says that if you practice it for long enough you can eventually stop thinking about it (in the cerebellum) and rely on your motor cortex to fire the right signals at the right time. Not being any kind of expert I don't know which is right, but I prefer the first explanation because it's cooler. Neural networks in our spinal columns Zombies! Er, ahem.

Anyway, that's basically it. The three most important things about armswing, in my book, are

1. don't swing side-to-side, it makes you look like a gumbie
2. swing faster to accelerate
3. armswing during crossovers makes foot placement a whole lot simpler. But that may be a topic for another time