Most of us are probably familiar with GearScore -- at this point in the life of WoW, it's pretty unavoidable -- a simple four-digit score that supposedly tells you all you need to know about your self-worth. At 5502, you are two whole points better than someone at 5500. What that means in real-world terms is ... you know, nothing. But it's cool to assign numbers to things. I did it all the time when I was 8. "This stick is worth 25 points. I have 25 more points because I have this stick. Your stick is worth maybe 10 points at best. I am better than you."
As if to highlight how arbitrary these scores are, there are a couple of different scores out there to measure yourself. Take my shaman, for instance -- she is a 1374, while simultaneously being a 2609 and a 301. These numbers alone certainly don't provide a whole lot of information. They're not entirely useless -- I could tell you that someone with a GearScore (as calculated by the "popular" add on or by wtfismygearscore.com) of 2609 is woefully unprepared for ICC -- but beyond that, I'd be guessing if you asked me what kind of DPS I should be doing based on that number alone. It's just not calculable, because all gear scores are not created equal.
Raid leaders who use it are guessing, too. That's where the problem lies. There's nothing wrong with me assigning a value of 25 points to a stick that I found on the ground. The problem comes when I start using those 25 points to decide that a kid with a 23-point stick can't climb on the jungle gym with me.
What is GearScore, anyway?
We can compromise and say that since we're playing pirate here, you need a stick sword. We can clearly tell that a branch with a few leaves still on it -- a 4-point stick at best -- is not appropriate for playing pirate on this jungle gym. But if you've got a stick and it at least looks like a sword, there are far better ways to determine whether or not you should be playing with me than some number I invented. (For example: Being a gnome sympathizer always disqualifies you, regardless of stick quality.)
Now, just like an 8-year-old Fox, GearScore and its contemporaries assign numbers to the various pointy sticks we have in our inventory. GearScore is calculated based on item level, which is itself based on a item's stat budget. Every level 264 glove that drops in 25-man ICC has the same basic stat budget.
A hypothetical piece of gear can have 250 points worth of stats to distribute. Most of the basic stats for shadow priests -- hit, crit, haste, spirit and intellect -- have a basic relation of one to one. (Spellpower is worth six-sevenths, and stamina is worth two-thirds.) That means one glove could have 100 intellect and 150 hit while the other could have 100 intellect and 150 spirit, and both simplified pieces would be viewed the same when it comes to determining GearScore.
Different class, different benefit, different balance
And therein lies the problem. Different classes get different benefit out of different stats. So long as we're still under the cap, shadow priests view hit to be far superior to spirit. In reality, one point of hit is worth approximately three points of spirit. In the world of GearScore, the two stats are equal. That person with the 4800 GearScore could actually have far better, far more appropriate gear than the guy with 5100 that just died in a fire.
That doesn't even begin to address skill, of course. There's no way for GearScore to measure how well you'll use the equipment that you have. As a lone arbiter of who you play with or as a predictor of DPS potential, it's pretty lame. After all, that kid who had the 23-point stick could have a stuffed parrot at home that he can get. And he could also do these crazy backflips. And maybe he also just got Mega Man 2 for Christmas. Man, is that game sweet! But because you didn't let him play with you on the jungle gym, you'll miss out on going to his house later. And did I mention his mom will make Chef Boyardee when you get there?