Thank you for the long answer. You ask what kind of metric I use. I base my argumentation on light traveltime because that result us the same if I interpret space as flat with refractive index or curved like normal GR. The underlying principles is that a particle needs to maintain its present coherence so that all corners of the particle remain together in the same level of simultaneity. Because every part is a component if the whole wavefunction with a rotation or spin of the particle, the density on one side has to increase. Imagine cars driving around in a big circle with one side slower allowed speed than the other side if the track. Each car will make the same loop time, but the cars are densely packed in the slow section and scarcely in the fast section. If that circle trac would be on a balance then the plane would shift... So, every component has its own variable traveltime over a section of its path. Since you understand the duality, I hope that you understand that it is easier to imagine it as variable speed.
Based on this understanding of gravity, I have to conclude that also the center of balance of a particle, the center of mass, has to wiggle around with some probability around the center if the particle. Thus the resulting gravitational configuration is a time dependant function or gravitational wave. This carries away energy and lets the particle shrink.
You ask about what radius I use. I don't care. I call it the relevant particle radius, because if outside of that radius the probability density function falls off with a steep or mild slope, that doesn't matter so much. The shrinking of the particle will scale the whole density function. So, mass linear with the diameter or radius is just a numerical factor.
For the shrinking matter universe it is important that the effective mass scales with the diameter and not with the volume. Just take the Newtonian equation of stable circular orbit.
F=GMm/r²=mV²/r
If every mass is double and the orbit radius is double, then the orbit is stable with double orbital time, no speed change. That fits with observed redshift z=1. If the mass would shrink with the cube of the radius, all would be unstable. So, my explanation of gravity fits with a shrinking matter universe.
Another issue is the mass problem with neutrinos. There are different versions of neutrinos and they can change their mass. If they have a thorous shape, like a donut, then the electric and magnetic lines can be closed while it has two diameters. One the diameter of the donut and one the thickness of the ring. So, depending on the orientation to the gravitational gradient, the effective mass of the same object is different.
Could it be calculated? The mass depends on the radius and the particle wave function. We know the mass. We can measure that. We can get some indication of the diameter. From there we can get some information about the wave function. But that becomes complex if you think that a proton is the conglomerate of 3 quarks. Each quark is ⅓ of the group but not ⅓ of the mass. This, because a quark alone has a far smaller diameter and a different set of orbital components than it has inside a proton.
Why is the gravitational force so much weaker than the other forces? Because it detects the difference in light traveltime over the distance of the proton diameter. To measure the Shapiro delay we got miliseconds for a sun fly by. That is a distance difference between earth radius and grazing just past the surface if the sun. So, yes that are small numbers and gravity has to be a weak force. But, in GR it is explained by curvature, and that is the same light traveltime. So, I don't go outside the GR equations.