The biggest reason Druids are so uncommon is that only one race per faction can be Druids.

As for the active population of a class being tied to its perceived power level, I dont buy into that. The most popular classes in any MMO are always Warrior and anything else that is fun to play. Healers are never popular because most people dont like to play a healer.

I think a lot of people actually like to heal. But outside raids, there is barely anything to heal :p.

As opposed to the current regime, where Blizzard applies huge nerfs/buffs in a frequently unpredictable pattern to the various classes? A class-heavy guild can still get "crushed" in the same way, but we learn to cope. Whether relative power changes are market driven or the result of a planned economy, annoying changes to game balance will result. The advantage of a market solution is that it would respond dynamically but in a highly granular way.

I think one of the biggest reasons that people aren't re-rolling Druids and Warlocks en masse is that, while those classes are certainly powerful, they aren't so much more so that people are willing to give up their significant investment in their current class. By and large classes are balanced, and most can farm reasonably well and are competitive in PvP (resto and prot specs being the notable exceptions). So what's the incentive to take up to a month off to start over? Getting your arena gear slightly faster? Moving up a couple of spots on the raid DPS charts? Compared to what you stand to lose - your raid spot, PvP team spot (and rank), and falling farther behind guild/friends in rep and gear - it's not surprising that people don't switch to the flavor of the month class all that often these days.

That is exactly the problem, you have to make a fairly major investment for a reroll to "pay off" in the end. A mage in SSC can't exactly just choose to play a warlock and pick up where he left off with 10% more DPS, he has to level to 70, reattune, regear, rerep, etc., and all for what? By the time he gets his warlock to where his mage used to be, months will have passed and he will be behind everyone else in the guild anyways - worse off than if he had just stuck with the mage. And for all we know the warlock class is going to get nerfed, or the mage class buffed in the meantime, and all that effort would have been wasted. It simply isn't worth it for most people to reroll.

I'm not going to address the issues related to class balance, but I'd point out that the assumption made in Economics that people act rationally to maximize their own benefit is not always true. If you don't believe me, I suggest you look up information about the Traveler's Dilemma. It's been demonstrated time and again that players in this game choose an option that does not correspond with the Nash Number (Nash Numbers correspond to options or decisions from which deviations can only produce lower possible returns, roughly speaking). Research has shown that people routinely deviate from the strategy that maximizes personal return without respect to age, nationality, religion, sex, etc.

So your argument is based on an inherently false assumption. Just thought I'd point this out.

http://en.wikipedia.org/wiki/Traveler%27s_dilemma

Summarizing, the principle is that you pick a number from 2-100, and another player does too (no collaboration). Both players get the second highest value written down, with two dollars exchanged punishing the higher bidder, rewarding the lower bidder. (A: 51, B: 50 results in A getting 49, and B getting 52).

The article summary (at my time of post) suggests that the TD is a critique of game theory, because the Nash Number (ie., "let's solve a game for the optimal play") is the least desireable outcome (a bid of 2 is the NN, but getting 4 doesn't compare with both players always stating 100 irrationally, thus performing 25 times better then the NN prediction).

The point being TD's suggests that player behavior is rational, actually, which is in opposition to what you said.

For what it's worth, my earlier post and link are actually in support of your suggestion - people are not rational agents. But TD is not a supporting case of that premise.

However, it is a far cry from "do not rationally maximize" to "make no attempt to maximize". Just because other factors come into play does not mean you can dismiss outright theories that assume resource maximization. Thus, although some players undoubtedly pick classes based on guild needs, sex appeal, Legolas worship, favorite D&D toon from the 80's, or whatever, there are also undeniably some people who either pick a first class or roll their first toon based on current perceptions of most powerful class.

I don't want to get into a big argument that derails this thread, but I strongly disagree. While it's possible to argue against the weight given to Nash Numbers in terms of choosing a strategy, it would be illogical to argue that NOT choosing a Nash Number would always yield higher returns. By definition any deviation from the Nash Number has a higher potential for reduced returns; hence, most would consider the Nash Number to be the rational choice if one aims to maximize personal benefit. This is why arguing that consumers (in this case, players choosing classes) always act to maximize personal benefit is a dubious claim; in TD players are not acting rationally. Rather, they act on an assumption that other players will naturally pick the largest number in order to maximize mutual benefit.

There are a number of cases where IF players acted in a concerted fashion they would reap higher benefits (such is the case in TD; there are similar cases of Prisoner's Dilemma in which if both actors refused to testify, they would mutually benefit more, but the Nash Number reflects a course of action to maximize personal benefit in TD). However, IF you are assuming that players maximize PERSONAL benefit (which typifies the economic assumption that consumers are 'rational'), this is not always true. Obviously, this is one of the strongest criticisms of Game Theory (as you pointed out) but arguing beyond this point would both put me out of my depth of knowledge and probably derail this thread further.

I'm not attacking the merits of whether or not people actually are choosing classes for this reason, I'm stating that assuming people are doing so because in economics class you learned that people act rationally is a dicey proposition. The inherent argument here is that people choose classes because they are overpowered and players are maximizing their own benefit; the OP concludes that we can see a reflection of this in the population statistics. I'm just challenging the basic assumption, that's all. I didn't (and won't) touch on class balance. I didn't intend people to get upset about my argument, but I strongly believe that if you assume players are acting to maximize personal benefit, you'll go down the wrong path in your analysis.

Just want to second this. I've leveled every class to at least 60 (most are mid-60s now, and two 70s) except shaman, and a feral druid was by FAR the fastest to level. I hit 70 in under 7.5 days /played without even trying. I did quests, instances, maxxed tradeskills, etc on my way to 70 -- I wasn't just mindlessly grinding, either.

Cat DPS rivals any other class once you get mangle at 40. Combine that with zero downtime (heal yourself every 4-5 mobs; your mana regens to full while you're meleeing the next 4-5) and your killing speed is incredible. Against green-con mobs you don't even have to stop to heal as ILOTP is usually enough to keep you topped-off permanently.

Plus you have the ability to solo most group quests/elites as a bear.

Even solo mage AOE grinding doesn't really seem that much faster to me, and there are very limited places where you can do that efficiently.

Bah, double post.

This is an interesting point. I wonder how the class balance would look if we removed race from the equation? How many people would play warlocks if dwarves/nelfs/draeni could play them? How many people would be priests/paladins if gnomes could play them? Etc.

This, of course, may be my own personal bias because I like gnomes and thing that a gnomish druid or shaman would be the most hilarious thing ever.

Is there any way we can remove faction/race issues when analyzing class population balance?

I would have rolled a druid if I could be a dwarf or gnome still.

Also, dwarf warlocks/mages would be awesome. I would totally be a dwarf mage. Gnomes are good enough though.

If they allowed gnomes to be druids, would their feral forms be teddy bear and pussy cat though?

There used to be dwarf mages =).

For my part, I think the widespread belief that there is a sizable number of "flavor of the month" rerollers is so wrong that it defies common sense. Unless you're Joanna/Mancow, rerolling and decently gearing a level 70 is going to take you many months. Very few people have the patience or even the willpower to abandon a preexisting character to do that.

My first experience with it was about ~3 weeks after the expansion, when apparently all druids were FOTM rerollers. I must admit that I was pretty impressed. After all, these people must be pretty bad if they deserve the title of "FOTM reroller." And yet these supposedly bad players leveled 1->70 and geared up in the time it took most normal players to hit 70. Truly an amazing feat.

Calling someone a FOTM reroller is just a lazy ad-hominem used to flame someone on the forum by establishing a non-existent caste system.

My personal suspicion is that rather than a significant number of rerolls boosting the standard population of FOTM classes it's actually alts being shoved forward as mains/dusted off and played after long periods of inactivity. By the end of Vanilla WOW a lot of people had multiple 60s and from personal observation a fair number were trying to decide whether to switch mains, TBC being the perfect opportunity after all. Consequently classes previously a bit lacking (Druid in some respects being a good example) get a substantial boost, not in the number of high level druids existing, but in how much play time those characters get.

Yes, probably. I always found that absurd too. Warlocks are the best farming class.

Hunters and rogues have better "get out of jail free" cards -- this has been my guess why they are better farming classes. Corpse runs kill your gold/hour rate more than anything else, and a warlock only has wipe protection every 30 minutes. Plus, soulshards are more than at least some want to deal with on a "fun" alt.

Soulshard farming must kill off some of allure of pvp FOTM and easy levelling of warlocks.

The thing of Alliance having enormous amounts of pallies in the early days and that going away was partly that they discovered that they weren't melee DPS in end-game after all, and partly the Seal of the Crusader bug fix that made them not actually be decent DPS even while levelling. With that fixed, levelling pallies became much more of a pain and the wellspring dried up. Levelling is the mindkiller. I wanted a blood elf pally in TBC, and he's currently idling at 41 or 42. I loved the new zones 1-20, got out, and promptly HATED the slow grind in Hillsbrad etc. It did get better admittedly once I got into the late 30s and then at 40 getting the mount, I'd probably finish getting him at least to Outland if I had more playtime (the last month I've hardly had a chance to log in other than for raids, one or two heroics, and a bare minimum of farming to keep consumable supplies up).

Or, they think they have the patience, but then it evaporates long before the rerolled character is levelled. I know a lot of people in my (Alliance) guild were talking the Shaman talk in the leadup to TBC's release. But look at the guild roster now, and you'll see 2 level 70 shamans.. and about 50 level 10-20.

I didn't play WoW at launch because I could not be a gnome priest.