I use a cleaning rod, collets and a bullet being pushed into the neck to get a starting point. Re insert the bullet and Measure that 5 or 10 times and you get some variance. Try it with different bullets and you will get even more variance. Might not be much, but you almost always have to take an average to get touch. Take that average to start with.
Then take your bolt and strip the firing pin so there is no tension at all when you close the bolt. It will just fall.
Then you can either remove the ejector (if the action has one) or take a dremell tool and carve out the case head/base so the ejector doesn’t touch anywhere. Make sure you size the case. Set the case in the bolt and slide it in the chamber. You will feel absolutely no resistance, it will just fall.
Seat a bullet in that sacrificed case to the the average length from the collet measurement and again set it in the bolt and run it in the chamber. 99% of the time it will not fall freely with that measurement, which means you are actually in the lands. If the bolt closing is real sticky, keep seating 2 or even 3 thou deeper till you can barely feel a tick or resistance. Then go half or 1 thou at a time. Measure between it depth and document it. Once you don’t feel any resistance on the fall or the rise of the bolt, the last measurement you took is touch. Just take that sacrificed case and throw it in you die box for the next time.
Yes it takes extra time, but it’s very accurate and repeatable every time. 5 years down the road when your rig doesn’t shoot as well as it used to, do this test again and see how much your throat has eroded. If it’s eroded 5 thou then seat you loads 5 thou longer than original and you are back in the money.
Like I said, Alex Wheeler has a great video explaining the process