Bob,
First, let's make sure that I properly understand the situation. I'm assuming that you have the elevators connected to the elevator pushrod, so they are effectively moving as a single unit. Thus the total angular movement from full down to full up is the same on each side.
You say that the up deflection differs a bit from side to side, and that you might have some HS twist. I interpret this as saying that if you put one elevator in trail with that side of the HS, you get a different amount of angular displacement to full up than you get on the other side. The difference in up travel is because you have a different starting neutral position on each side, due to HS and/or elevator twist.
Assuming all the above is true:
I would average the results from the left and right sides and see how it is. If it is inside the recommended limits, put away the digital level, and never check it again.
If the down travel is a bit more than the recommended max, personally I wouldn't worry about it. I can't see how that will get into any trouble.
If up travel is a bit more than the recommended limit, I would investigate how to reduce it. Too much up elevator could contribute to a too high AOA condition someday, so it is probably something to be avoided.
If either up or down travel is less than the recommended minimum, then you should do whatever is required to fix it.