Latest Armory population figures from Okoloth
Okoloth has dropped an update to his Armory analysis, featuring the latest and greatest information on about half of the World of Warcraft's denizens. He surveyed 4.6 million characters on the Armory, and while that sounds like a lot (it sounds like half of WoW's population, except that Blizzard's 9 million figure is supposed to be players, not characters), it's not actually that much of a representative sample. Still, compared to the table scraps that Blizzard gives us, it's something, so let's get what we can from it.He finds that the biggest majority of players are at level 70 compared to the other levels, but there are still only about 40% of the characters there (adding fuel to the fire on both sides of creating midlevel and endgame content). Mages and Warriors are the standouts on the class breakdown at level 70 (with 13.5% and 14.3% respectively-- what tank problem?), while Shaman are the biggest losers-- only about 7% of level 70s he surveyed were Shaman. Sounds about right. Across all levels, Warriors still have the biggest percentage, while Hunters follow them up. And on the low end, it's Druids, Pallys, and at rock bottom, Shamans. People just don't like playing the totem class.
He's also got new stats on realm balance, but remember that these numbers are not much more than guesses. They're pretty close, though, even for that. Drysc told us that Agamaggan's Alliance/Horde balance was about 1.1:1, and Oko's figures have it at about 1.09:1 (by my math), which is pretty darn close. Big ups to Oko for putting these numbers together, always interesting to see the figures on how and where people are playing in Azeroth.
[ via WoR ]
Filed under: Shaman, Warrior, Realm News, Polls, Analysis / Opinion, Fan stuff, Blizzard, News items






Reader Comments (Page 1 of 1)
Matthew Nov 5th 2007 6:24PM
Wait, "not actually that much of a representative sample"? I'm not seeing where that conclusion comes from. Even assuming all 9 million players have 3 or 4 characters that they actively play through, polling 4.6 million characters is still very much a representative sample, I would think. Especially considering most nationwide political or issues polls only contact a thousand people or fewer.
jr Nov 5th 2007 6:36PM
It may be or it may not be. We can't say, I don't think, without know his methodology for choosing who to include in his sample.
Charlie Taylor Nov 5th 2007 6:48PM
Ha! Take that you "everyone and their mother plays a rogue" people!!
...
Ok, so not many people complain about that, but some do!!
...
ok, like 5 do. Throw me a bone people!
Goldwolf Nov 5th 2007 6:48PM
@1
Well, we might consider this a representative sample if this many characters were chosen at random from a pool of all characters.
It would NOT be a representative sample if he took characters with names starting at the first part of the alphabet and kept going until his data collecting script broke down. ;)
prudychick Nov 5th 2007 7:27PM
Over 616,000 warriors and you can never get a tank.
Nearly 500,000 priests and you can never get a healer.
SirCasey Nov 5th 2007 7:36PM
"what tank problem?"
Um, just because it's a Warrior doesn't mean it's a tank.
I think on his site he said he's working on massaging the data to produce class and talent breakdowns in the coming week or two, I recommend waiting until then to make comments like that...
Green Armadillo Nov 5th 2007 7:57PM
@6: In fact, the previous iteration of the data looked at Warriors and found that about half are Prot specced, so I'd expect similar analysis this time out. At any rate, the "tank problem" is generally not just whether there exists a competent player with an appropriately specced/geared toon, but whether there is one willing to tank for your pug, so that isn't a question you can get at by analyzing this data set.
RE: Mike's comment on only 40% of characters being level 70 as evidence for the need for more midlevel content, remember that this is characters, not players. The real question is what proportion of players have 60+ characters, what proportion of those have one or more 70 characters, and how many of those consider the addition of more low level content as the deciding factor in whether they're going to level more characters. We can't get at these numbers via the Armory either, but I suspect that Blizzard's decision not to include more mid-level content in Wrath is indicative - players who entered TBC with one level 60 main missed out on four of the expansion's eleven adventuring zones, representing a significant amount of Blizzard's dev time. I guess the second question is, once the patch day rush to check out Dustwallow wears off, whether the 2.3 changes have a sustained effect on the rate with which people re-roll/play low level characters. Again, time will tell, and again, we won't have access to the data that really matters. :(
Bunkai Nov 5th 2007 8:14PM
"...Shaman are the biggest losers-- only about 7% of level 70s he surveyed were Shaman. Sounds about right."
"People just don't like playing the totem class."
I don't think it's that we don't like to play the _totem_ class, I think it's that we don't like to play the class that's gotten mauled by the NERFBAT every patch since 2.0 (Sept-ish '06).
I still play my Level 70, non-raiding, non-arena-ing, Enhancement Shaman, even though he's not even keyed for a single Heroic 5-man. I do my fair share of bitching about how f'ed up the class has gotten since it's glorious days of OP BGs(remember the Unbreakable video from YouTube?...no?...look it up, it's worth the viewing.), but still. It was my first, and is my only, level 70 toon... and I'm loyal to him. There'd be far more players picking up their DW axes if we could magically be restored to our 2.0 PTR(pre-WF and SR nerfs) greatness.
Note: I'm not one of the QQ'ers that made a "." post. But still happy to hear that ES won't be added to the DR system. :-D
Hargrim Nov 5th 2007 10:44PM
One thing to consider on the lack of shaman is the ratio of horde to alliance and how many horde started out playing shaman vs how many alliance started out as paladin. Checking the stats, paladin is the second least played class at 70. I'm guessing this has more to do with shaman and paladin being faction specific up until BC. I'd be willing to bet not too many people wanted to level a new class to 70 when they had other chars that were already 60.
Of course, any theories you make about these statistics completely rely on how valid this data set is. Since we don't know the method for obtaining the data and it's obviously not a complete set, it definitely brings into question it's accuracy.
GamerJunkie Nov 5th 2007 11:21PM
Its expected that DPS classes are more popular than Healing or hybrids.
Warriors now dominate Arena PVP, while people think that Warlocks are OP and people would roll them but turns out not the case, its not everyone's cup of tea.
You see the classes with the lowest DPS or CC at the bottom.
Druid, Pally, Shammy.
Jason Nov 6th 2007 12:02AM
Tank problem is because tanking is stressful enough, add in dumb puggers that open dps before the tank has it, and don't know how to cc, and you got a nightmare. So most tanks stick to guild runs.
@9 I've seen other stats showing horde/ally differences and the difference between Ally Shammies and Horde Pallies is huge. Ally have hardly touched the class, even as a low level alt while horde has a ton of pallies, most are still low level alts but there's still a ton.
honkhonk Nov 6th 2007 6:29AM
"the class that's gotten mauled by the NERFBAT every patch since 2.0 (Sept-ish '06)."
You mean warlocks?
Darbad Nov 7th 2007 3:56PM
I agree that this isn't a very good sample, as far as I know the sample wasn't taken at random and furthermore I would guess that the average WoW player that has a level 70 character has at minimum 2 or 3 other characters above level 12, with an average of about 4-6 alts not including bankers. Everyone's been a little curious about what other classes or races are like, a lot of people changed mains before hitting 60, more still changed servers and still have characters left over.
Additionally I think his statistics include people who no longer play. With relevance to class balance this is the most relevant shortcoming since I would argue that tanks and gear based classes have the highest burnout rate (though I know healers burnout too).