For that game, the sunk-costs fallacy and the training-to-do-random-things-infinitely phenomenon may help in speculating about why so many sink and continue to sink time into it. I've noticed that people who bite the bullet and quit speak not as though they were dependent and longing to relapse into remembered joy, but rather as though horrified in retrospect at how they let themselves get used to essentially playing to work, that is doing something which in theory they enjoyed yet which in practice was itself a source of considerable stress/boredom/frustration. (Again, I have had no direct experience with the game.)
For cocaine, straightforwardly there's the expectation that it would do bad things to your receptors (as well as your nose lining...) such that you would gain dependency and require it for normality, as with caffeine and nicotine and alcohol. Your priorities would be forcibly changed to a state incompatible with your current priorities, thus it is worth avoiding. If there were a form or similar thing which in fact had no long-term neurological effects, that is one which actually gave you the high without causing any dependency (is that even theoretically possible, though, considering how the brain works? Well, if dropped to the level of most things then instead, say...), it might be worth trying in the same way that music is helpful to cheer oneself up (if it cost less as well?), or perhaps sweet foods would be a better example there.
The standard answer for sex is that it's already part of your system of priorities, and so there's little helping it. Practically speaking, it would probably be far easier if one could just turn off one's interest in that regard and focus one's energy elsewhere--particularly, in terms of the various psychological/physiological health benefits, if one already cannot experience it yet is near-futilely driven to seek it. Again though, there one more wants to turn off 'the impulse to have sex' rather than sex itself, since there are advantages if you want to have sex and do compared to if you want to and can't, and also advantages if you don't want to and don't compared to if you want to and can't.
Hm... returning to the original question wording, if one treats World of Warcraft as a potentially-addictive use of time that may truly or otherwise effectively rewire one's sytem of priorities to the point of interference with one's current priorities, then it is likely reasonable to avoid it for that reason. It's again important to note which priorities are true priorities (such as improvement of the world?) that one wishes to whole-heartedly support, and which are priorities which, when stopped to think about, don't have a particularly reason to value (such as the sex drive issue, which actually doesn't have much going for it compared to other ways of pursuing pleasure).
(Species-wide reproductive advantages are acknowledged.)
I've avoided playing World of Warcraft because many people enjoy it so much that they neglect other things in their life.
Does that make sense?
How about cocaine?
How about sex? I hear that's pretty good too.
ADDED: Lots of interesting discussion, but no one is getting at some points of particular interest to me. Most answers assume that you have important stuff to do, and you need to decide whether WoW will prevent you from getting that important stuff done. They also assume that your brain usually errs on the side of telling you to do "non-important" stuff (WoW) at the expense of "important stuff".
One question is whether there is any evidence that your brain is biased in this way. I think your reflective self greatly overestimates the probability of success at the "important stuff". I have worked very hard, twelve hours a day, 7 days a week, on "important stuff" for most of the past 30 years. The important stuff never pans out. So it appears that when my brain told me to play Freecell rather than work on that important paper on artificial intelligence that got pulled from the book the day before publication due to petty office politics, or to watch Buffy rather than do another test run of the demo I spent three months preparing for DARPA that no one from DARPA ever watched because the program officer was too busy to supervise his program, or to go hiking instead of spending another weekend working on the project for NASA that was eventually so big and successful that my boss took it over and then tried to get me fired1, or to go dancing rather than work on the natural-language processing approach that got shelved because my boss felt it emphasized the skills of mathematicians more than his own, or to LARP rather than put in another weekend on my approach using principal component analysis for early cancer detection that it turned out some guy from the FDA had already published 6 months earlier, or the technique for choosing siRNA sequences that a professor from George Mason already had a paper in press on - all those times, my brain was using a better estimate of success than my reflective self was.
Another question is why the "important stuff" is important. Fun is fun. On the surface, we are saying something like, "I have a part of my utility function that values contributions to the world, because I evolved to be altruistic." If we really believe that, then for any contribution to the world, there exists some quantity of fun that would outweigh it. And people use language like, "WoW may be fun, but it has little lasting effect." But when you contribute something to the world, if the relevant motivating factor to us is how our utility function evaluates that contribution, then that also has little lasting effect. If you do something great for the world, it may have a lasting effect on the world; but the time you spend feeling good about it is not as great - probably less time, and a less intense emotion, than if you had spent all the time accomplishing it playing WoW instead. So this question is about whether we really believe the stories we tell ourselves about our utility functions.
1. He got to award himself all of the department's yearly bonus money that wasn't awarded to his subordinates, so any obvious success by his subordinates was money out of his pocket.