#17: A scale of one to 10
Scores are just the start of a review, as much as aggregators want to make them the final word.
What’s the average on a scale of one to 10? Most people assume it’s just the tall end halved. In this case, five. The mathematical midpoint is actually 5.51 — but this is also wrong.
What if I rephrased the question: What is the average response on a scale of one to 10? Anecdotally, it’s somewhere between 7 and 8. In fact, Metacritic’s cutoff for a game to be labeled “mixed or average” is exactly 74. (Which means that the “most average” game is Dead or Alive 5: Last Round, the best reviewed mixed or average game.)
To me, this feels intuitive. I’d watch a seven, but not a six. It’s the “friendly version of average.” But this is partly a question of taste: Is averageness a negative quality? Would you watch a movie that you knew had exactly nothing going for it? Some would, some wouldn’t.
Any rating is ambiguous. Scale ratings are as open to interpretation, misperception, and bias as anything else, despite their appearance of authority. Just as soon as we corner something into objectivity, feeling prevails. Math can say nothing about the midpoint on a scale of one to 10. It really, really depends.
In my days at the literary magazine, we rated every piece on a scale of one to 10, and this was a real problem. What exactly is the difference between a poem that’s a seven and a poem that’s an eight? People agonized.
I learned that a senior editor called Dave — one of those people I hardly knew, but who had an outsized impact on my worldview — only gave three scores: one, five, and 10. A poem either sucked, or it did nothing, or it soared. Thinking about it now, I’m not even sure if he included five!
Some critical publications use scored reviewing — IGN, Roger Ebert — but I think of this as a kind of tradition. Modern reviewers publish full-length reviews with no scale. But this, too, is a question of taste.
Scoring publications prize efficiency for the end user. Writing shows the work of the critic, like writing out the formulas on a math test, but the score is the final answer. Readers need not scroll through sevens, or even eights or nines, until they’ve checked out all the 10s. These reviews are functional.2
Conversely, a critical review includes no score. Audiences must read the whole text in order to know how the critic felt. They aren’t for people just looking for something to watch on Netflix: they’re for people who already watched it, and who want a critic to describe their feelings back to them. They want an interpretation of the work.
These two reviews — critical and functional — don’t exist on a binary. The smaller the scale, the more interpretation is needed, and the more critical the review becomes.
The gaming site Polygon thoughtfully implemented a two-point scale: recommend, or not. Calling scores a “verbal plaque,” editor-in-chief Chris Plante wrote, “The soul of the reviews program is healthy and strong, but now it’s free to inhabit an endless variety of bodies.”
In other words, quality can mean many different things. Tetris and The Last of Us can’t both simply be called 10s.
The reverse is also true. No human person — no friend or critic you seek the opinion of — would kibitz over a 74 and a 75 the way Metacritic does. Metacritic, and its movie counterpart Rotten Tomatoes, are ur-functional. Numbers in a void of context.
Hyperfunctional reviewing is extremely seductive. But they also flatten something totally ambiguous — how good is x — to a single figure. (Remember New Year’s resolutions and value capture?)
This is possible because of aggregation en masse. Rotten Tomatoes presents reviewers with a two-point scale: “Did you like the movie?” It then spits out the proportion of 👍’s to 👎’s. That means that a movie which simply manages not to bother anybody can get an insanely high score.
I like Thor: Ragnarok, which sits at 93%. But is Ragnarok exactly as good as No Country for Old Men? Is it miles better than Scream at 83%? Maybe it is for you! But I think those numbers conceal a lot, rather than clarify.
And when big games publishers tie financial awards to scores on Metacritic, as they often do, rather than any kind of critical quality? For me, that’s where scoring goes awry.
We seek reviews to better understand art. Metacritic conjures the feeling of standing before a crowd of strangers, hollering the titles of movies, cupping one ear to catch their response. The crowd’s volume, whatever that volume could mean.
Conversely, I think the perfect review is the moment when you leave the theater, look side to side, and then breathlessly exchange all the things you couldn’t say in the movie’s quiet.
Or maybe it’s Marianna Starke’s 1820 travel guide, among the first recorded uses of scores in reviewing, used exclamation points to evaluate pieces. I like this approach for two reasons.
One: because enthusiasm rolls off the consecutive exclamation point. When I first saw John Wick, I felt “!!!!”.
Two: because Starke’s exclamation point rating comes with no upper limit. It doesn’t assume some maximum of enjoyment.
That’s what I want in a scaled review. I want to know what you loved or hated, your enthusiasm and expected enthusiasm. The “scale of one to 10” is an idiom: the number doesn’t mean anything if it doesn’t tell me how you feel. Tell me seven, tell me 10, tell me a billion, tell me a color.
Don’t sweat if you’re wrong: the question has no right answer.
Thanks for reading!
I recommend CDSK, a really very good trivia board game that asks you how much you know about each topic on a scale of one to 10. It’s like an object lesson in the fallibility of scoring, which is to say that I usually lose due to incredible hubris! (Also you get to choose whatever as your token, which is a nice touch.)
DR
If you need convincing: there are five numbers between one and five: 1, 2, 3, 4, 5. And there are five numbers between six and ten: 6, 7, 8, 9, 10. So the mathematical midpoint is between these two halves. Thus, 5.5.
I’m borrowing this “functional” vs “critical” language from the critic Claire Dederer.
remember when youtube used to have you rate videos from 1-5 stars? iirc they switched to like/dislike buttons (and later removed the dislike button because they were salty about the reception to that one youtube rewind but that's a different story) because with the original 1-5 star system, most people would only ever give one star or five stars, so the middle options were meaningless
I’ve always used the zero to ten scale for ranking albums. However, I hate how 7 is seen as average. I prefer my scale to be anything above 5 means I at least like it and now I’m scrutinizing how much I like it. If it’s less than 5, I don’t like it and the 0-5 is for ranking how much I didn’t like it. To me, I really notice a difference of an album I gave a 6 compared to an album I gave a 6.5. I honestly love being so nit-picky on albums :) something I don’t do in normal conversations 😂 (btw, my average would be a 5.5. This would be an extremely inoffensive album that didn’t make me feel much, but I wouldn’t say it’s bad. Example: SUPERBLOOM by Misterwives) I also change my scores if I need to. Art can be interpreted in a different way at different points in your life!