Agree there's not a lot of logic in some of the differentials. Some are in favour and get consistently overvalued - think of ARM for instance, and some are consistently underrated. Large caps do better than small caps - there is some logic to that, but if you can pick the right undervalued small cap you'll make much more money. It all makes a little more sense if one looks forward one to two years rather than focusing on historic p/es - the market always looks forward and , in theory, those with high multiples generally have rated management, growing profits and good forward visibility of earnings.
PRO i guess is seen as having much to prove; a chequered past and a touch of the 'jam tomorrows', though at least it is looking like tomorrow rather than far out into the distance. A p/e in the high teens would a very good rating for a company like PRO - not impossible given a string of good trading updates and the likelihood of a leading share in it's market - but it's not something I would count on.