RT's Tomatometer or Metacritic's Metascore are not measures of film worth as a kind of "atomic weight." Tomatometer, in particular, is a statement about how man

RT's Tomatometer or Metacritic's Metascore are not measures of film worth as a kind of "atomic weight." Tomatometer, in particular, is a statement about how many reviewers gave a positive review, without regard to each reviewer's depth of feeling. Imagine three movies: First: 90% of reviewers gave 5/5, 10% of reviewers gave 1/5. Second: 90% of reviewers gave 3/5, 10% of reviewers gave 2/5. Third: 80% of reviewers gave 5/5, 20% of reviewers gave 1/5. First movie: Almost all reviewers raved about the movie - it's probably good. Second movie: Not a single reviewer loved or hated it; it's probably pretty bland. Third movie: vast majority of reviewers loved it, and the rest hated it. It's not as sure a bet as the first, but all of the reviewers reacted strongly to it. Would you rather see the second or third movie? I'd rather see the third. For example, is The Avengers (92%)[1] really a better executed, more worthwhile film than Cloud Atlas (66%)[2]? Metacritic handles this a little better, because the reviewer score is weighted, not just a +1 or -1. However, it still has the problem of 90%/10% and 60%/40% both averaging to 50%. No number-crunching, but I have a suspicion that studios are getting better and better about shotgunning out enough chum to different types of reviewers to effectively game those numbers, which is how you arrive at RT scores like those you present. [1] http://www.rottentomatoes.com/m/marvels_the_avengers/ http://www.rottentomatoes.com/m/marvels_the_avengers/ [2] http://www.rottentomatoes.com/m/cloud_atlas_2012/ http://www.rottentomatoes.com/m/cloud_atlas_2012/ Roger Ebert wrote about Cloud Atlas: "Surely this is one of the most ambitious films ever made. ... Anywhere you go where movie people gather, it will be discussed."