Getting developers to all offer recording features would be a major step past ALL of this, though. Predictable FRAPS results for the vast bulk of games, so one doesn't have to be stuck with the most-used (and therefore most-targetted-for-optimizations) games/benches for years on end.
The only problem with THAT is, of course, that the demo recorders could specifically target areas of a game that favor one IHV over another, but I suppose if you're getting down to THAT level of distrust of reviewers, there's a lot more "bad" going on anyway. (After all, they could just be making up the numbers they put down.)
There will always be a place for static benchmarks that look specifically at certain areas and abilities of cards, though, as we still want to be able TO specifically target certain capabilities, as well as sketch out a general picture of what to expect in the future, or when certain development trends follow a course that we can't see yet just by looking at current games.