He also wasn't commenting about trans people as far as I know. His last words, in response to a question about mass shootings, were something like "depends on if you exclude gang shootings"

How do we tell/measure that "disorder" is up? Are we talking about certain classes of property crimes? Coming from the Canadian context (where we didn't see rioting in summer 2020), public sentiment seems similar but I wonder if it's tied to similar objective measures of "disorder".

I found the paper "Explaining Justificatory Asymmetries between Statistical and Individualized Evidence" by Renee Jorgensen to be pretty convincing (though I find rational settling confusing). Paper explains the problem with "naked" stats and gives reasons for being more skeptical of eye witnesses.