My thoughts on this might surprise some but my answer is generally not to bother at all unless you have one of the pivoted tangential trackers that are VERY loose in the fore-aft direction (I am referring to pistonic action of the arm wand along its axis). Other than those cases, the mathematics just don’t support the effort of chasing the tiny changes in overhang due to suspension settling.
Using dynamic and static testing on the lab bench, one of our engineers determined that my cantilever alignment accuracy is comfortably +/-0.5°. I can’t be sure of getting any more accurate than this on my best day, with the best illumination and the most exposed cantilever. On average, a 0.1 mm drop in cartridge height due to suspension settling will result in a 1° change in cantilever angle. While I am not saying this is impossible, I have yet to see this amount of a drop that has not been due to damage to the cartridge. However, for the sake of this argument let’s just pretend the cartridge has dropped to 0.2 mm and the cantilever angle is ridiculously high at a 45° angle which would mean that the overhang would’ve also changed by 0.2 mm. (BTW: I have seen a 40° VTA before! Instead of returning it to the factory, I wish I would’ve kept it to measure the distortions it causes. VTA is not as important as SRA but it becomes very important on vertical modulations.)
So what does 0.2mm change in overhang result in? This is easily calculable, but I’m going to give you figures by memory so forgive me if they are slightly wrong. At the inner null the angular error that would’ve been introduced is 0.17° and at the outer null area is only 0.09°.
Remember, my cantilever alignment accuracy is only +/-0.5°, so it doesn’t make sense for me to try to correct this as, statistically speaking, I could possibly even make it worse!
The stacked tolerance errors inherent to alignment protractors such as yours that use multiple parts fastened together to assure the geometrical relationship between the pivot point, the spindle and the stylus are far more likely to be greater than the change in angular error resulting from a suspension having settled after break in.