2 BM engineers instructed me that using a double-sided disengagement ("pinch") approach "should" increase axis spring longevity, if the user is a card carrying member of the Frequent Flicker Firm.
Whereas, a third BM engineer unequivocally stated it makes no difference, asserting instead, that deeper and full compression of the spring trajectory is a more likely culprit.
As per my own personal 2 decades of experience, every axis spring breakage occurred during the course of single-sided disengagements, with one noteworthy exception for which I will attempt to locate a photograph, long ago taken.
With the one above-referenced exception, no axis springs have broken in any of my many BM axis lock folders, since switching to a double-sided ("pinch") disengagement approach, many years ago. Whether or not the correlation is sheer coincidence, it simply FEELS wrong and weird, when utilizing single-sided disengagement approach.