Your Game Ratings Suck

Is that new game you bought rated a 7.3 or a 6.97? How much of a difference does a third of a point make anyway? If you think it’s an 8, and like it better than that other game you also rated an 8, shouldn’t you downgrade the inferior one? If there’s a truly fantastic game, do you rate it an 11… or just make 10 louder?

A small but annoyingly vocal percentage of elitist, snobby internet grumps would have you believe that ratings are useless, and nobody takes them seriously. But do people really care about board game ratings?

Yup, for sure. I’ve been in plenty of stores where employees recommend games based on the average rating. I’ve had conversations with people who have said such-and-such must be crap because it’s “only” a 6 or must be great because has a 7.7 average score (ignoring that it’s based on a just handful of votes, two of which are 10s from the publisher and designer). And yes, I’ve seen Board Game Geek ratings referenced on Kickstarter pages in direct attempts to bring in the ca$h.

So there’s no getting around it – ratings are a factor. And they are useful when taken as a part of the story.

How do things get ratings, anyway? Well, random people choose numbers based on how much they like something. Or sometimes they base the numbers on what the quality is like even if they’re not crazy about it themselves. The decision captures how they’re feeling about that particular game on that day and time, based on their mood, what they’ve eaten for breakfast, and often after one play where they don’t even know if they played correctly. A great session of a poor game might get a strong rating, and a play of an otherwise excellent design may be marred by happenstance. Some people rate games without even playing them, which is… well, weird.

Board Game Geek tries to help by adding a quasi-objective set of guidelines to this 100% subjective process. I guess that helps consistency, but the system still skews things far higher than I like. Plus it’s sabotaged by misguided Ratings Crusaders tossing angry 1’s around to offset blind 10’s from fanboys, designers’ pals, and Kickstarter backers trying to justify their purchases. And this is all before the games have been released!

So yeah, number-based, subjective rating systems are inherently sort of lame. You can’t get around subjectivity being an issue, so let’s focus on the number portions of the systems themselves. Ten-point rating systems are garbage. Five-point ones aren’t much better.

First off, what’s an “average” rating on a 1-10 scale? Five, right? Yeah, five. It’s gotta be five. Everyone knows five is the answer. Only an idiot wouldn’t kn- Wait, it’s actually 5.5? Hmm…

(10+9+8+7+6+5+4+3+2+1)/10 = 5.5

Yeah, I guess so. No biggie; a minor technicality. I’ll wait while you go and rejigger all your flawed game ratings.

So most games should rate a 5.5, and go up or down from there. That doesn’t seem to be the way we usually do it, though. It’s like when you see talent judges on TV who don’t want to hurt anyone’s feelings and give 6’s to awful performances, 7’s to so-so ones, and jump straight to 9s for the “pretty decents.” What’s the point? They’re never going to use the whole range of scores.

What’s the average rating you give out? I bet it’s more like 6.5. Maybe even higher since a lot of people don’t bother to rate games they don’t like. My average game rating is in the low 6’s, and I want to punch myself for blowing the curve on my own numbers.

But really, 1-10? Why the need for such a wide range anyway? Is a 4 ungood, a 3 plusungood, and a 2 doubleplusungood? If you put stock in ratings, you’re not going to play any of those regardless of the degree of badness. Bad = bad = bad. I don’t know anyone who says, “Well, I’ll play a crappy game, but not a terrible one…”

On a scale of 1-10, I rate the 10-point rating system itself as a 4. Today, at least. Ask me again tomorrow after lunch.

What if we subscribe to a five-point standard?

1 = Poor

2 = Below Average

3 = Average

4 = Above Average

5 = Great

That’s better than a ten-point scale. It doesn’t fix all my concerns about the 10-point system, but the smaller range of scores helps a bit. Now you know if people generally think something is good, without trying to interpret different shades of goodness or translate numbers. No more gradation between 6’s, 7’s, and 8’s – now they’re just 4’s. Suck it up, kiddos.

This shorter scale should bunch most items in the middle which is where things, by definition, belong. If you’re using ratings to rank games, this works fine as well, with plenty of differentiation once there are enough voters to average it out the decimal points.

On a scale of 1-5, I’ll rate the 5-point rating system as a 3. It gets the job done, but there’s room for improvement. I’d give it a 3.5, but half point ratings kind of undermine the whole purpose of this scale. Let’s cut it out with the half points.

I’m partial to a 4-point system myself, as it forces people to get off the fence.

1 = Poor

2 = Below Average

3 = Above Average

4 = Great

It’s something that good market researchers do when writing survey questions. You don’t learn as much from a survey if indifferent responders slap AVERAGE across the board. If you get rid of the lazy answer, people are forced to push things higher or lower, especially if you don’t allow cheating with half points (seriously, half points are the devil). But people don’t like thinking or making hard decisions, so this won’t fly.

On a scale of 1-4, I give the 4-point system a 3. It’s better than the two other systems, but still falls into the other traps: subjectivity, arbitrary 1’s and 4’s, etc.

There’s really no ideal rating system out there. The movie website Rotten Tomatoes makes a nice go at it, showing you what percentage of reviews a movie has received are positive, but even then you don’t know what it’s really saying. 75% could mean 750 positive reviews out of a thousand… or 3 out of 4. And all positive means is “not negative.” Could be an A+, could be a C+. And then it’s still a random assortment of personal feelings.

I guess what it comes down to is: what’s the purpose of a rating system?

To me, the only rating that should matter is: is a game worth playing, or not? Pass/Fail. Yes/No. Play/Don’t Play.

There are way too many games out there nowadays to futz around with ambiguity. I want cold, ruthless, short-attention-span, black-and-white deterministic answers. A friend who knows what games I enjoy playing telling me “you’ll like this” is worth far more than 1000 strangers proclaiming it’s a 6.7336G2!5.

Ratings will always be subjective. You can’t control how people make their decisions, and you can’t tell if we’re all working from the same set of rules. They can be part of your decision-making process, but shouldn’t be all of it. Factor them into the equation with other considerations, like BGG’s weirdo overall game rankings and peoples’ dumb old top ten lists, which are entirely different cans of internet worms.

Am I going to stop rating games? Nah. Even we cranky jerks still think it’s kind of fun, and a flawed system can still be useful. If it helps people, I’ll keep doing it. I’m going to make sure I’m extra thoughtful with my 6.383’s and 5.995’s. And I’m going to TRY not to be an elitist snob.