What score do you give a game when you like it, and you think it's interesting and important, but you have to admit to yourself that it just isn't very good? It's a seven out of ten, isn't it? And that sucks.
I was at BAFTA event  last night about games journalism, alongside Naomi Alderman, Kieron Gillen, Rob Fahey, Keza McDonald and Keith Stuart, and we kicked off by talking about Metacritic, the site which collects reviews - and delivers an average score based on them.
McDonald wrote a great piece asking whether Metacritic is "ruining the games industry" (albeit inadvertently). Games with higher Metacritic scores sell better, and sometimes a development team will only receive a bonus if they get over, say, 85.
Here's the key paragraph:
A Metacritic average undermines the whole concept of what a review is supposed to be: an experienced critic’s informed and entertaining opinion. Instead it turns reviews into a crowd-sourced number, an average. You can’t average out opinions. If you adore the new Muse album and your Radiohead-loving friend hates it, that doesn’t make it an average album. And yet this is exactly how Metacritic scores are treated by publishers. It punishes divisive games – and honestly, most interesting things are at least a bit divisive.
You can easily apply this to some of the most interesting games of the last few years. Take LA Noire, which made huge strides forward in several really exciting ways (story, motion capture, aesthetics, music), but yet somehow contrived to be less than the sum of its parts. I wrote a short-ish review of it, noting exactly that, then breathed a sigh of pure relief that I didn't have to give it a rating out of ten. (The New Statesman doesn't do 'em, being all highbrow and that.)
But look at Metacritic. LA Noire scores 83 (in a range from 92 to 60 ), and I honestly can't say what I think that means. Halo 3: ODST  gets that score, too, and that game was a perfectly pleasant (but sliiiiiightly unoriginal) iteration of an established series. Heavy Rain is another example where the Metacritic score seems oddly meaningless: its average of 87 comes from a range that goes between GameCritic's 40 and BoomTown's 100, both of which I rather uncharitably suspect were trolling for hits on a popular game by awarding an outlier mark. (A similar thing sometimes happens when writers are filing to a newspaper: they know that only a 0-star or 5-star review will make the front page; they know that a real hatchet job will get the hits rolling in. It's hard to resist that kind of pressure.)
The NS's film critic, Ryan Gilbey, is one of many who hates the five-star system on movie reviews - and who says that trying to choose between 92% and 93% would make him go nuts. Here's what he had to say about numerical ratings when I emailed him:
Scoring, whether you're awarding stars or the somehow-even-more-irritating and pompous grades (C+, A-, etc) which are becoming prevalent online, does such a disservice to the complexities and amorphousness of film - of any art form, in fact. Its sole function is as a consumer guide, a short-cut for readers too lazy to discern for themselves what a writer thinks about the subject in hand.
While allowing exemption for those writers who are forced by their paymasters to award ratings (we've all been there), it seems baffling to me that people who write their own blogs or put grades and ratings in their tweets are doing so out of choice - don't they realise it demeans them and the film? It makes it appear that they don't place any value on a film other than a commercial one. It also encourages the reader and writer to play into the whole star-rating pantomime - the reader inevitably relishes the 1-star review, and the writer performs accordingly, while the 3-star review is given only a cursory read. It's a way of pureeing the review in advance for the reader, warning him or her whether the water will be hot, cold or lukewarm.
I don't want to be read by anyone who needs their hand held through a review, or thinks films can be broken so easily into good, bad and ugly. Ratings are full of anomalies and shortcomings anyway. If Citizen Kane and Andrei Rublev and Some Like It Hot are 5-star movies, how can anything else even be worthy of 3 or 4 stars, let alone 5? Maybe we should invent a sixth star for the canon - and therein lies the insanity of Spinal Tap and the dial that goes up to 11.
The star rating system has commercial currency, no doubt - why else would distributors plaster their posters with quote-less stars ("***** - Daily Mail") which make every advertising hoarding resemble a clear summer night? But challenging it is down to the people who profess to care about the art form they're writing about.
Of course, there are many people willing to defend the score system as a service to the hard-pressed reader: if you only have limited time, then what greater shorthand can there be than a number? At one point in the Bafta debate, McDonald pointed out that IGN's readers really like scores - and Rob Fahey reminded her that it was more that the vocal ones do. (Only around 1 per cent of IGN readers are commenters, so it's impossible to guess the feelings of the community at large.)
My hope, however, is that the mania for scores is just because we're so used to them - and, actually, we wouldn't really miss them if they went. In the short-term, one single step would make games reviews more interesting: BAN THE NUMBER SEVEN (or anything in the 70s, if you're reviewing out of 100).
Most of the panel agreed last night that 7 is shorthand for "this was interesting but had flaws", and therefore was even more meaningless than other arbitrary numbers. There's also some weird voodoo where, even if a score is nominally out of 10, people still regard 5 as a "bad" mark rather than the average. Seven is thus the new 5.
Oh, and while we're at it, maybe ban 0 and 10 as well, tarty little attention-seekers that they are.
Once we've got used to that, maybe then we can start on 1, 2, 3, 4, 5, 6, 8 and 9.