Kijanki, agreed on all counts, aside from the last sentence :-)
I would add, with respect to the mention of "a little of black magic in this hobby and placebo effect," that there are probably also many cases in which effects are correctly perceived but are not attributed to the correct variable. With the correct variable often being unrecognized due to some combination of inadequately thorough methodology in the evaluation, and/or technical misconception. That said, though, as I'm sure you'll agree a substantial body of anecdotal evidence, taken with a modest quantity of grains of salt, can often provide significant value.
Best regards,
-- Al