My thinking is that the torque applied is equivalent, but the harmonic frequency of the system changes because of the length of the moment arm. If the weight is greater and higher up on the shaft, the harmonic frequency of the system is higher and will (probably) settle out faster than when the weight is hanging way out on the end of the counterweight shaft. This seems to be a better configuration overall. There are other factors involved including the correction rate for the tracking and the magnitude of your non-periodic (random) error in the tracking rate.
My two cents.
Director Electrical Engineering
Explore Scientific, LLC.