The methodology for this has always bothered me. When it comes to multiplatform games, the 360 version generally gets the most reviews, followed by the PS3 version, and with the PC version receiving the fewest reviews. This often results in the metacritic scores varying by quite a bit. That variance has nothing to do with the quality of the different versions; it's simply a result of having more opinions thrown into the mix. In some cases, the superior version can end up with a substantially lower metacritic rating, compared to the inferior version.
There's no easy way to fix this while using a hard (and arbitrary) cutoff point, like 85. An improvement would be to include all versions of multiplatform games as long as at least one version scored 85+, and as long as there's no major differences between the versions. That's really just a band-aid fix, though. The real issue is that Metacritic is a tool to assist in buying decisions; a somewhat flawed tool, but still a useful one. It really doesn't work very well when applied to "list wars" competitions.
