You're forgetting all the biases (& fallacies) which may or may not have affected the experience or the memory leading to the anecdote. These biases (& fallacies) are detectable or the lack of them is verifiable, with the scientific method.

I'm not denying an anecdote is not valuable to the individual. It is, for the individual who made the anecdote; not the rest of us, not the readers.

I'm also will admit I, too, use anecdotal evidence. But that does not mean one has strong evidence when they do; they don't, and neither do I.