One thing that has tended to bother me for a while has been game review scores. It isn't so much that I don't agree with the scores certain people give games (although many times that's the case), but rather the entire rating system itself. Most game publications give games a score on a 10 point scale. In actuality however, it's more like a 100 point scale, since everyone seems to award points in tenth of a point increments instead of the full point. This has the nice side effect of allowing a greater range of scores and as such allows the reader to see really where a game falls in that grey area from 0 to 10.

Movie reviews tend not to have as much grey area. Most movie reviews are on a four star scale, or in the case of Ebert and Roeper, a boolean scale, either thumb up or down. One problem associated with not having a lot of grey area in your reviews is that it's harder to differentiate certain items from each other. Ebert can give two movies a thumbs up rating, but which one does he like more? Which one should I really see?

These types of decisions are even more important with video games for the simple fact that you are parting with $50 of your hard earned money. Good thing they use that 100 point scale in gaming then, right? Ahh, but that's where the skewed math comes in. Instead of review scores being evenly distributed over the 100 point range, they are weighted more like a grading curve. Meaning "A" games are perhaps a 9.2+, and "F" games are a 6.5-.

This is where the problems start to creep in. Very great care seems to be taken in numerical positions of games in the 8.0+ range, yet in the 5.0-0.0 range, a range over twice as large as from 8.0 on up, the numbers seem to just be randomly thrown out on the table. When a 5.0 is already a failing grade, what's the difference if a game instead recieves a 2.3? In reality, there is none. What's basically happened is that this nice big grey area game reviewers had to score games with and show some actual differences between games, has now been reduced to a four star system. A 10.0 score is now four stars, a 6.5 is half a star. Anything less than that, no stars.

What I'd like to see is for game reviewers to use a scoring format that actually corresponds to their scores. If you're only going to use 6.5-10.0 for your real scores, then use a four star or A-F grading system. If you're intent on keeping the 0-10.0 scoring system, then use it properly. Average games would get a 5.0 score, the same games getting an 8.0 now. Personally I'd like such a system, but I'd rather just have one in place where a reviewer felt they could actually use the full range of scoring, regardless of what type it was. Until then I guess I'll just have to live with 40% of games getting a score in the 8's.