> Turns out the people actually doing the work are usually pretty sensible.

The replication crisis suggests otherwise; they’re often delusional (believing in things that aren’t there) or outright frauds (publishing results they know are false for personal gain).