>>would moving the sub around relative to the fixed dimensions of the room likely change the curve much?

Absolutely -- it's just harder to predict what the results of each move will be. There are a FEW tricks you can do that are somewhat predictable, like positioning the sub X distance from a wall so that the distance to the wall and back is 1/2 wavelength of the frequency where you have a peak, but there usually aren't enough variables to fix all your response bumps.

Hey, I wonder how well an instrumented sub crawl would work ? Put the sub in your listening position, move the mike around the room and find the spot with the best response, then place your sub in that spot and take back your listening position ?

M60ti, VP180, QS8, M2ti, EP500, PC-Plus 20-39
M5HP, M40ti, Sierra-1
LFR1100 active, ADA1500-4 and -8