Seems to me that a good rule of thumb in the replication crisis in Psychology was that if an articles claim seemed kind of “obvious” (where you would like the OP almost wonder why they bothered researching it), then the experiment would replicate. Conversely, if the result was really “interesting” (as in surprising or unexpected), chances were that it would fail to replicate.

I guess that isn’t so surprising as most of us deal with people constantly and so our intuitive understanding of human psychology is actually pretty good.

But it does prove the value of constructing the scientific theory painstakingly and carefully out of tiny claims rather than trying to do something bombastic.