The behavior I observe is similar to what I've rarely encountered when the pitch of the sprocket doesn't quite match that of the track/ chain
or the spacing between the sprocket halves is marginally off. It's not enough to be obvious in a no load condition but at heavier demand it causes a problem

I don't yet understand the difference mechanically, dynamically between running forward / reverse - clockwise / counterclockwise but I'm sure there's
a clue there.

I thought of sticky pins but I've run the tracks enough to hopefully rule that out and that wouldn't quite entirely explain the forward/ reverse difference.

I'm going to run the model on my bench to see if I can spot something. I can use a break from my JagdPanther.

I'm not above buying another set of tracks/ sprockets for this model as I really like it. It steers well being slow and as I mentioned the power is great.
I may upgrade it to a Clark unit; I'm using a Dave BU I assembled and it's ok but the functions are not as granular as I'd like.

Jerry