yes.

have you ever done A/B comparisons of multiple stems with everything else constant? or multiple bars?

for me, doing actual testing made some things clear that previously i might not have felt mattered. what was eye-opening to me years ago was when i had a bar/stem setup where i thought the bars might not be stiff, so i swapped them out for something definitely stiffer. the exercise actually made me realize that the stem was a weak link, despite others saying that model was 'plenty stiff.'

ymmv, of course, and perhaps you might not be sensitive to changes if you have tested them or might not care. however, to suggest that others cannot tell the difference because there is too much else that is going on in the system is not fair.