Dave, thanks for the insights. FWIW, the method I’ve settled on is basically the first method you described. For one thing, my lathe is small and the bed is too short to fit a barrel between centers and chamber in a steady rest anyway.
I use an outboard spider and either a four jaw with a copper ground wire ring in the jaws or a D&T’d faceplate with brass tipped screws at the inboard end. I use a combination of Gritter’s Rods and a long stem Mitutoyo to dial in the breech end as close as I can and let the muzzle run where it ends up. I try to start where the throat will be located and go forward from there as I don’t care what the bore is doing toward the tailstock as it will become chamber.
I then put the reamer in my Manson holder (being careful to ensure the floating surfaces are in full contact) and then set the compound to follow the body taper on a reamer flute. I then run in a twist drill to hog material and follow that with a carbide mini boring bar to establish the chamber taper with at least four passes to account for flex. I then run the reamer in to complete the chamber.
My lathe is small, light, and, frankly, not very precise or rigid. A heavy nudge on the chuck will upset the dialing-in process although it does seem to gravitate toward a consistent center. This is where my pondering upon pilots comes from and why I use a floating reamer holder. I agree, the reamer will want to cut concentric and only does otherwise if forced to do so, but I can’t help but wish there was a simple way to ensure perfect alignment every time.