In fact, it is overexplained. There are many different explanations, sometimes overlapping, sometimes inconsistent. My OP generated a small survey of what some members here pick as their favorite explanation. It turned out to be a subset of explanations existing "on the market."

3 hours ago, studiot said:

May I suggest that electrons would be a better example than water molecules since H₂O and D₂O have some different physical properties ?

Yes, you may. Your introduction from this other 'natural kind', D₂O makes your world view more precise.

3 hours ago, Genady said:

Which regularity makes differential geometry effective in describing GR?

None, if you ask it in this way. But I would say, based on the regularities we see (perihelion precession e.g.), we know there must be at least some mathematical explanation. It is 'just' a theoretical insight that differential geometry does the trick.

On 1/6/2022 at 8:57 AM, Genady said:

In fact, it is overexplained. There are many different explanations, sometimes overlapping, sometimes inconsistent. My OP generated a small survey of what some members here pick as their favorite explanation. It turned out to be a subset of explanations existing "on the market."

Here's one you probably haven't seen;

My "theory" is that math is logic quantified and reality is logic manifest.   

The imperfect overlap creates rounding errors, constants, and a misapplication of mathematical principles and equations to a digital reality.