FYI, this is not my work (author is chad sapieha). My comments are below the article. I did not post the full article (just first page).
Every year, upon surviving another holiday season in which I do my best to play through and review as many as 50 games in just a few months, I sit back, reflect on my job, and pour my thoughts into a lightly philosophical post. In years past I’ve written about such things as difficult publishers who restrict access to their wares and how, contrary to popular belief, being a game reviewer rarely means you get to play games during your workday.
This year I’m going to focus on a hot-button topic in the world of video game journalism: Review scores.
I began my career 15 years ago adamantly opposed to applying scores to my reviews. Assigning a simple number or letter grade to a review of any form of media seems to me to be a clumsy, inelegant, and unfair way of judging an artistic work. Taken on its own – and this happens frequently – a score lacks any sort of context and fails to do justice to games that, more often than not, large teams of people have slaved over for months or years.
That said, I’ve come to view review scores as a necessary evil.
Viewed in practical business terms, scores are essential to growing a game review Web site. Without them, reviews would never be picked up on aggregation sites, rarely be included in meta-reviews posted on popular blogs, and have less of a chance of being used as source material on sites like Wikipedia. Put plainly, they drive traffic in a variety of ways.
More importantly, they serve a purpose to readers. Not everyone is willing to devote time to reading thoughtful critiques. All they want to know is if a game is good or bad. In this sense, assigning a very high or very low score provides value to readers in a rush. They instantly know whether to invest or avoid, and – at least in the case of the former – can discover for themselves why a particular work is worthy of their time and money.
Where this argument breaks down is in games that receive mediocre scores. Someone might see my review of a game like Rage, for example, note that I assigned it a middling score of six out of ten, and move on without reading why. If they dive into the text, however, they might discover that Rage has just the sort of things they crave in a game – great graphics and polished action – and that the reason I gave it a lower score is because it failed to meet my expectations in terms of storytelling and innovation, elements that may turn out to be of minor importance to our theoretical reader. It could, in fact, be just the sort of game she’s looking for, but, thanks to the score, she may never know.
Scores also tend to bring out the worst in fanatics. It seems to me that many people who browse reviews are simply in search of a story that validates their opinion. It doesn’t matter if they’ve played the game or not; they just want to read something that conforms to and confirms their beliefs.
This is easily illustrated in the growing war between fans of Battlefield and Call of Duty. Having sworn loyalty to one series or the other (and acting under the misguided assumption that most other gamers have done so, too), these readers see review scores as evidence betraying the camp in which the author resides. They’re not interested in reading an opinion that might make them reflect on their own views; they just want the reviewer to be on their side. Giant Bomb has a nice post on this phenomenon – formally known as confirmation bias – written after Eurogamer's honest review of Uncharted 3: Drake's Deception caused a minor ruckus among fans...
I have seen this topic heavily debated on countless gaming websites from time to time. When a hotly anticipated game comes out, it's merit and acclaim is directly correlated to it's aggregate review score. The game is not worth your time. Fans use these scores to demean games, and as ammo against other forum members.
I find that this article delves into what makes a review score, what his thought process is in giving a score, and why it seems the ten point scale is really 6-10 scale. He gives points that I never considered, and I feel that chad's explanation makes intuitive sense.
This topic has been debated on these forums, and this question will be asked again. What do you think of review scores? What is wrong with them, and how should they change, if at all?
