In any case, this doesn't change anything about the segmented arcs I'm getting on this part. I tried exact stop this afternoon, and it didn't make a difference. Still getting the same behavior.
I think the answer lies right there, in the numbers of your test. X axis is pretty much obvious, heaving an absolute deviation of 0.002", which is enough for those segments to be seen. I would guess that if you did the test over a longer distance you would see a distinct repetitive nature of the numbers. Entering the numbers in Excel would make it easier to visualize.
Since you're not using a belt drive, you have to start looking elsewhere in the driver mechanism. For instance, what is the motor? Could be a stepper motor with poor accuracy - unevenly spaced poles or magnets on the rotor. Since your Y axis is much better than the X, you could try swapping the motors.