Let's see if I get this.
The fellow measures something.
He flips the cable.
The measurements change.
He calls it significant.
He guesses that it is a difference in attachment quality between the two end connectors.
He imagines that the difference is a change in impedance. AND
That difference causes a change in the magnitude of reflection. AND
Dynamically changing standing waves are induced.
He does not measure the cables.
He does not consider the possibility that unplugging and reconnecting the cables is itself a factor
But he does note that the difference is also a function of the device being measured. The greater the device error, the greater the impact of the cable direction.
Oh, and it is also a function of cable cost.
This is an example of why I no longer am a subscriber to Stereophile. It looks like science, but it wanders into wine tasting. Kal Robinson and some few others excepted.