SleepDoc_PDX said:

The boring version of this is the one that works, and the boring version is: measure a baseline, change one variable, wait, measure again under the…

Correct, and worth adding that it is reversible. Nothing being described here is a one-way door, which changes how cautious it is reasonable to be.

Happy to go further on any of that.

SleepDoc_PDX said:

The boring version of this is the one that works, and the boring version is: measure a baseline, change one variable, wait, measure again under the…

Bookmarking. The distinction being drawn above is the one nobody else makes.

From the other side of the consultation, briefly. Whatever the answer turns out to be, the method for getting there is the same: state the assumption, do the arithmetic in public, and invite the correction. That is slower than asserting, and it is the only version that survives being wrong.

julia.endo said:

Whatever the answer turns out to be, the method for getting there is the same: state the assumption, do the arithmetic in public, and invite the…

Can confirm. Same sequence, different timescale. The detail I would add is minor and it is already implied above.

Closing the loop on my own question.

Closing this out. What actually changed my mind was the point about waiting for a trend instead of reading a single measurement, which is embarrassing in hindsight and worth saying anyway.