Adding the clinical framing, because it changes how the question reads. The honest answer is that the effect is real, the magnitude is contested, and the individual variation is larger than either. Those three things can all be true at once, and most arguments here are two people holding different parts of that.

That is the short version; the long version is somebody else's post.

roxy_nash said:

The honest answer is that the effect is real, the magnitude is contested, and the individual variation is larger than either.

Can confirm. Same sequence, different timescale.

roxy_nash said:

The honest answer is that the effect is real, the magnitude is contested, and the individual variation is larger than either.

There is a second half to this that has not been said yet. 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.

Adding the numbers, since they settle part of this. If you are going to change something, change one thing and give it long enough to express itself. Four weeks is the usual minimum for anything pharmacological on this board, and two weeks of data has told you almost nothing.

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.