Why do tonearm specifications list null points?

This happens all the time: we conflate tracking angle error with "distortion" without actual evidence that the two are perfectly correlated. And what sort of distortion might be correlated with TAE? Suffice to say it seems to be a good idea to minimize TAE, but whether that also minimizes "distortion" in a predictable or audible way is not established. Or if it is, where are the data?

Several years ago, I experienced the issue discussed by Dover and Larry with a Dynavector DV505 tonearm, probably because the DV uniquely places the vertical pivot so far forward from the horizontal pivot.  The DV was designed with headshell offset for Stevenson or something very close to it.  When I aligned a cartridge according to Baerwald, I of course had to twist the cartridge in the headshell so that its long axis was deviated toward the spindle. I was unhappy with the resulting SQ.  Reasoning that the problem was due to the necessary misalignment of the cantilever with the vertical bearing arc, I then straightened out the cartridge using Stevenson.  Whether this hypothesis for cause is valid or not, correcting the alignment of the cantilever to be in the same plane as the arc of the vertical bearing made for a big improvement in SQ.