I don't think this response should be getting downvoted; I'll upvote it for the third paragraph, which I largely agree with. We can agree to disagree on the ex

I don't think this response should be getting downvoted; I'll upvote it for the third paragraph, which I largely agree with. We can agree to disagree on the exact definition of "virtually impossible", but in the broad strokes I think we're both saying some factor analysis can be sometimes productive. In particular, I definitely agree that 1) an "actual understanding of underlying causal factors" is imperative, and 2) an approach based on trends (and trend following) is naive.