Time saved vs used is a pretty good argument against.
My previous printer (Anycubic Mono 6k) allowed you to change lift height and speed on the printer during a print, so I could just lower it until just above hearing the separation. My current printer (Saturn 3) doesn’t have this capability, so it would have to be multiple prints, which would take a couple hours.
What might be a better solution, is to do a similar test, again with “worst case scenario” (ie a cup with no suction cup release hole, or a very small one), increase the result by a large margin of error, and go with that until I need to replace the FEP or prints start failing.
All very good points. I think if I want to continue entertaining this idea I need to adjust my sights to a more general number than anything exact. Another user pointed out that time spent testing vs time saved would be a pretty close cut. I might devise a test to find a general minimum, increase it by a fair margin, and call it close enough.