Probably not-- the above is for averaging many vectors, and a helical axis is only a single vector. A time series of the principal axis of the helix backbone atoms from an INERTIA time series would be more what you need; the angle can be computed from the unit vector Z projection.