Hollow depth discussions are a bit off-topic, but:
For your experiment to work, you need a sharpener who checks the calibration of his dressing jig and who dresses his wheel carefully. I grew suspicious of one guy, so I bought a Hollow Depth Indicator (HDI) gauge. I always asked for a 7/16" ROH, but what I got ranged from 7/16" to 9/16". I dropped him. My current guy is more consistent, but to get 7/16", I need to ask for 3/8". Also, some sharpeners hand finish the hollow with a cylindrical or conical honing stone to get rid of chatter marks from power grinding; hand finishing can lead to variations in the final effective ROH.
Most hand finishing (and any good skate tech does hand finish) is to do one or more of:
1. Repoint the sideways pointing sharpening burr into a downwards pointing foil edge.
2. Polish the sides of the edge - especially if #1 is done, because the wheel only polishes one side.
3. Get rid of the sharpest portion of the edge, because many people don't like skates too sharp, and it is too fragile to walk on off-ice, even on rubber, without blade guards. The sharpest part of the edge is very close to the sides of the (chrome relief section of the) blade.
4. Straighten out the edge.
5. Complete deburr the edge.
I haven't seen anyone touch up the hollow. There will be no visible chatter marks if you sharpen with a fine grain wheel, and you lubricate the blade with an oil or polishing fluid first.
Of course, on steel there will be microscopic roughness on many scales, including the fact that the edge height is always quite ragged - you can see the initial raggedness, before hand finishing, with a 50-100x optical microscope, on a size scale not much smaller than the mean difference in height between the sides and center, even when the sharpening is done by an expert skate tech. I have wondered whether that raggedness isn't a part of what makes sharpened steel skate blades cut into the ice, but don't know how to tell.
Smaller scale roughness cannot be eliminated in sharpened steel, as you can see if you look at
electron micrographs of sharpened razor blades.
(Properly stropped razor blades get rid of most of the edge raggedness - but that nothing that subtle is done to skate blades. And electron micrographs show that smaller scale roughness remains even after stropping. Maybe someone like you who understands material science can explain why to me - maybe it is due to a property of steel, or to the way sharpening and polishing is done by abrasives?)
If you use a coarse wheel, or you sharpen much too quickly, or don't use an oil or fluid, maybe you can't avoid visible chatter marks? But that would just indicate a poor skate tech, or one pushed by management to work too quickly.
If you do #1 and/or #3, and possibly #5, or don't do #4, I don't think a simple depth-based indicator can possibly give a correct reading of hollow radius, because I think it is designed to estimate hollow radius - and indirectly to estimate edge angle (which many believe to be the most important characteristic) from that, simply by measuring the height difference between the sides and more or less the middle of the hollow. But side height won't be "right" if anything is done to the edge that makes it not be the shape of an intersection between a cylindrical hollow and a vertical side wall. The edge angle will also be estimated wrong if the blade has vertical side honing - and if there is horizontal side-honing, edge angle and hollow depth will (and should) vary along the blade, because blade thickness varies.
Incidentally, hollow radius might also be estimated wrong by depth-based gauges if the sharpening isn't new - because skating also grinds off the sharpest part of the edge.
I think you have said before that you have seen inconsistent hollow radii on different portions of the blade. Maybe it is really a difference in how the blade was hand finished? And that could apply to what you are talking about now too?
The only way a height-based depth gauge could consistently estimate hollow radius right is if it ignores the sides of the blade, where hand finishing and wear affect shape, and just looks at points very close to the center. Does your depth-based gauge do that?