If I can, I try to indicate about where the neck will be, or as close to it as possible. Not possible on big things like 338 Lapua. Chads method with the probe and wieght probably better for that, but have had good success with getting close.
On the irregular stuff, either move forward or aft a little big, seems to help a lot, kinda tells me it is a localized issue, BUT .... not something I want in the throat! If one groove is a bit deeper, on that 4 groove for instance, I just spread it out between the two sides. For instance, I number my adjustment screws 1-4, 1 and 3 have equal runout, 2-4 have equal runout. All four may not be the same, but that would be the preferred endpoint.
If the grooves are off a bit in clocking, same thing, move in a bit sometimes makes it much less so. Rifles seem to shoot OK even with grooves out of precise clock position anyway, this may not be an issue. I'd be more than a little curious to hear others thoughts on that, I am thinking to get a borescope to check out the local conditions next time I see this Might just be a burr, or irregularity, rather than a clocking issue, knowing how cut rifled barrels are made.
Doing coupling alignment on big power turbines for years, was a very good training ground for this. It's one reason I use this method,
it is simple for me to remember. If you can do rim and face readings on a coupling, this is easy.