Dr.KarenChen 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. It helps to say which part of this you are uncertain about. A precise question gets a precise answer; a general one gets everybody’s favourite anecdote.

One concrete data point for the thread. For anyone reading later: the numbers in this thread are worth checking against a primary source before you act on them, including mine. Half the figures circulating in this community trace back to a secondary summary that dropped a qualifier.

One thing that is still open after Dr.KarenChen’s answer:

What would you measure differently if you were starting again?

DataDave said:

For anyone reading later: the numbers in this thread are worth checking against a primary source before you act on them, including mine.

Careful with the causal language. Two things moving together in a period when five things changed is not a mechanism, and stating it as one makes the thread less useful to somebody reading it later.

OP back with an update, since a thread like this is useless without one.

Late follow-up. I did not do what was suggested, I did something adjacent, and it worked; posting the difference in case it matters to somebody.