Peer review in and of itself is of limited value, and usually reveals cosmetic, as opposed by structural, errors. How many code reviews can be summarized by the letters "L", "G", "T", and "M"?
Replication matters. The echo chamber of peer review much less so.