I think their evaluation method and need for repeat visits means they can't have very good coverage anywhere (there are what, 20-30k in NYC alone?). I doubt a

I think their evaluation method and need for repeat visits means they can't have very good coverage anywhere (there are what, 20-30k in NYC alone?). I doubt any region/city has more than a handful of reviewers. If they were uniformly sampling this would be a disaster, but presumably they have a (probably opaque) system for funnelling down to a manageable number.