Did You Know You Could Predict Review Scores?

Can you predict review scores? Graham McAllister thinks so and I agree to a certain extent. Eurogamer reports that Graham Mr McAllister has been conducting research on predicting game review scores. I do agree that review scores can be predicted relatively accurately, however I don’t go down the scientific route with McAllister.

The research apparently involves ‘œbehavioural or sequential analysis’ fancy terms for, if you fall asleep, you’re not enjoying the game, or if your controller gets hurled through the screen, you’re angry, etc. John Gottman who is responsible for this particular piece of the research might disagree with my blasé analysis. Reverse engineering of old reviews, particularly those of Edge magazine, is another method used in this research. They look at the correspondence between words and phrases used in a review and the score given to the game and chart them. This is a piece of the research I will admit to wanting to read.

The people used in this research are selected from a wider audience than perhaps is usual, according to McAllister “Four or five years ago people made games for gamers: geeks making games for other geeks, basically. Now you have people making games for everybody. The problem is that the games companies don’t understand everybody,” he went on to say “We’re very, very thorough about the people we choose to test the game.” The testers are hooked up to biometrics than monitor heart rate and skin response, this gives a more accurate report of what the player is experiencing over what he actually says, to me it sounds like an anti bias filter which is great news. Now I am not scientist and I’ve probably done this research a grave injustice in my interpretation of it, for that I am sorry. I just can’t help but wonder is all this really necessary? Mr. McAllister would almost certainly say yes.

According to McAllister the games are sorted into bands specifically ‘œlow (1-4), medium (5-7), and good (8-10)’ so I decided to try a little research of my own. Modern Warfare 2, I predict will be in the good band, the next Dynasty Warriors game will be in the medium and the next game in the Imagine series on the Ds will be in the low band. This experiment took me around a minute, long enough for me to type the sentence out and take a sip of ice cold juice.

Yes you could say that it’s obvious that Modern Warfare 2 will sell like hot cakes or that games called Imagine Babyz will be shit, however is it not the same with most games? It doesn’t take pages of scientific data to predict that games like ODST and Modern Warfare 2 are going to get great reviews and sell a lot of copies. I’m just wondering why all this research is necessary? This is all besides the point that reviews are all biased in one way or another as they are all subjective opinions. If you like a game or series then a bad review shouldn’t affect you, that is, however, a debate for another time.