What should impact a game’s score?

This week’s Challenges of Game Reviewing is going to be a bit different. Instead of an organized article, we’re going to go on a roaming magical thought experiment in an attempt to answer the question: what should a game review’s numerical score actually represent? Along that way, I’m going to propose ideas, state why they’re good, and then throw a situation where that idea breaks down. It’s entirely possible, and downright likely, that you’ll disagree with some of my rebuttals or suggestions, and that’s fine. What I hope, though, is that this will demonstrate the complexity of the problem and the reason why there is no simple solution. Let’s get started.

  • Review scores should focus on how much fun the game is. Nothing else should matter.That’s probably the default place most people would start when figuring out what a review score should mean. And doesn’t it make sense? Games are meant to be fun, so you rate how much fun it is. Simply, elegant, and powerful.But wait! The problem here is simple. How do you compare a fantastically fun game that takes two hours to beat to a really fun game that takes fifty hours to beat? Or a pretty fun game that has a multiplayer mode that you’ll spend three hundred hour playing? Where does play time and replay value come into this equation?
  • Ok, fine, review scores should be fun times time. A reasonable suggestion — consider how long you’re having fun when considering how fun the game is. If you’re having a blast for 2 hours, doesn’t that compare favorably to having a good time for 20 hours? That sounds like a fair comparison.But wait! That doesn’t quite work when you compare games with vastly different scopes, and, importantly, prices. One of my favorite iPad games is Space Miner, an innovative take off on games like Asteroids, Gradius and Space Trader. Would I say I have as much playing it as Batman: Arkham City? Definitely not. But I played both for the same amount of time, and Space Miner costs less than 5% of Batman: Arkham City. How do you compare those?
  • Fun times time divided by price? That’d be a reasonable compromise: take the amount of time you spend playing it, multiply it by how much fun you have playing it, and divide by how much it cost to get it. That’d give you a measurement of Fun-Hours Per Dollar, FHPD, which would seem to be a pretty good rating of the game’s quality, wouldn’t it?But wait! That’s a pretty reasonable measure, and this might be the point where you think this line of reasoning can stop. But how does this deal with rehashes, rereleases, ports, and other things of that nature? Take Madden, for example. In a given Madden game, if you like that genre, you’re going to have a lot of FHPD. But should the rating ignore the fact that the game is basically the same thing that was released the year before, typically with minimal updates? Do you analyze every game solely on its internal merits and not against the general climate of the industry? In my opinion, no, that information has to be included.
  • So, take the same measure, but scale it in some way according to how similar an experience you could get for less money. That makes a lot of sense, actually. Take the game, calculate its FHPD, and then look at the FHPD of other games of similar genres and scale its ultimate rating accordingly. If Madden NFL 11 gives 50 FHPD while Madden NFL 10 gives 75 FHPD, then Madden NFL 11 gets a lower rating than it would if there wasn’t a similar experience available. (Of course, this introduces one of the issues with including price as a major criteria — prices change over time.)

    But wait! What about innovation? Take a game like Portal. Portal would certainly have a pretty high FHPD, and it was pretty dissimilar from other games, which means it wouldn’t have anything to scale it down. But does that alone describe Portal‘s appeal? I don’t think so. A lot of Portal‘s appeal came not just from what a fun game it was, but also from how revolutionary it was. Nothing quite like Portal existed before. Shouldn’t it deserve some extra credit for that?

  • So make the scale go both ways: scale down when there are similar games out there, scale up when it’s pretty different. A good compromise — identify a game’s FHPD, then scale it back and forth according to whether or not you could get the same experience elsewhere. If it’s a rehash like Madden, it loses some credit, whereas if it’s completely new, it games some credit. Maybe imagine multiplying the FHPD rating by a factor scale from 0.5 to 1.5, giving the necessary extra credit.

    But wait! So far, this entire analysis has rested on fun being the operative measure for a game. But where about other qualities, like artistic merit or societal reflections? What about a game that makes a good point, or paints a beautiful plot, or creates an engaging world, all without being necessarily in service of the game’s “fun”? How do you rate those?

  • Maybe abstract out fun a bit? Instead of fun, be more generic and just call it quality? That would make a lot of sense; instead of just asking how fun a game is per hour, you ask how good it is overall, which subsumes its artistic merit and other key considerations. The best game is the game that gives you the most quality.

    But wait! What happens when the different forms of quality collide? Take, for example, ICO. Please don’t dismiss this point because I’m insulting ICO, but in my opinion, that game just wasn’t fun. It was artistic and immersive, sure, but it wasn’t fun to actually play. There are other games like that out there, too — games that aimed high on some goal besides fun, but still aren’t enjoyable to play. Some might say Final Fantasy 13 is like that, a game that aimed high on story and narrative but forgot to be fun moment-to-moment. How do you then rate those?

I’m going to stop there because, frankly, I don’t have a good answer to that last question. I don’t think it’s effective solely to rate games based on how “fun” they are because games like ICO, Shadow of the Colossus, and even the Assassin’s Creed series clearly aim for appeal that goes beyond simply being fun. But at the same time, if a game isn’t fun, who is going to want to play it?

I think (hope) this illustrates what makes review ratings so difficult — there are so many different criteria that people will argue ought to go into a review rating. I’m analytical by nature, so I’d love to have a formula by which I could plug in ratings and come out with an objective score, but I don’t think that’s quite possible. Yet, if review ratings don’t reflect some underlying truth, what good are they?

3 Responses to “What should impact a game’s score?”

  1. And to make it even more complex: what’s fun for you does not necessarily equate to what is fun for others. Fun is not objective. So as you can see, there is no possible equation.

    What good are review ratings? The answer is pretty simple: they are not good nor bad. They are just out there because people “want” to see them.

    • I almost covered that, too, but with me, the thing on that is that there are ways to solve that problem — either pay attention to reviewers with the same taste as you, or pay attention to aggregate scores instead of individual reviews. But both those only work if these above problems are solved, too.

      • In my opinion, the more professional a reviewer “tries” to be, the worse his/her reviews are going to be. I think professional reviews focus too much attention on technical aspects of the game and forget about the “overall picture”. As for aggregate scores, the problem is that they are just numbers, and numbers in reviews are insignificant in measure compared to prose.

        A good review to me is one that is not influenced by others, one that the reviewer knows what he/she likes, one where the reviewer honestly explains why he/she likes or dislike certain features to the point that it becomes understandable.

        But keep in mind one thing: there are no absolutes. The fact that one review is good does not imply that the opposite is bad. Two very different reviews of the same game can both be good. They can coexist together and none of them is better than the other.

Leave a Reply

Your email address will not be published. Required fields are marked *