Another reason I miss Bob Lemke. He himself had an interest in variation cards and wrote about them often in his column and his blog. When he was editor of SCD I found him always willing to discuss variants and whether they should be considered variations in SCD.
He actually started removing some variants, particularly border gaps, from the Catalog lists of variations because he came to think recurring print defects ought not to be classified as variations. He seemed to believe there had to be some indication the card had been intentionally changed. That left a gray area as to cards where it is virtually impossible to tell if a difference in cards was intentional or not...which is often.
Since there is not an accepted universal hobby definition as to what constitutes a variation, I think the recognition process by PSA or catalogers will always be haphazard.
But if PSA can recognize an errant green smudge in the baseball on the back of the 61 Ron Fairly card as a variation, anything may be possible
Persistence and documentation does seem to be they key