> I'm sure you could run the same experiment with e.g. prestigious math journals. There is probably a significant overlap between authors and reviewers. These

> I'm sure you could run the same experiment with e.g. prestigious math journals. There is probably a significant overlap between authors and reviewers. These are completely different matters. Any idiot can have an opinion on just about any non-academic literature. Whether any idiot can be taken seriously is up to debate. Whereas for math papers, there are very few people outside the pool of authors who can understand enough of it to even have an opinion.