I am indeed. I tried looking up the changes and just went “uuuuh what?” so I guess it will make more sense once I know what all the talents do currently.
1 Like
At the current moment it is monk and specifically brewmaster spec.
Any data set that isn’t literally complete will have biases. The challenge is identifying what they are and whether they are important for the study.
And my point is one of the biases is far, far less relevant for the discussion for the other:
- It’s easy to prove correlation between class power and class choice. It happens all the time and sometimes it’s even forced upon the player
- It’s basically impossible to prove correlation between using Data for Azeroth, or having run into anybody who did at any point (it also uploads info on players you encounter, even if they don’t use DfA), and class choice.
I sincerely doubt correlation 2 even exists. I’d be extremely surprised if it did.
wow. i switched to race and pandaren are 0%!
No, Pandaren neutral are at 0%. Check Pandaren A and Pandaren H, add them. 
We’re one of the more popular races.
1 Like
i beg your pardon. true 
DfA are the first to admit
All statistics are computed across the full Data for Azeroth database, which may not be an accurate reflection of the overall WoW population.
They’re in stealth you fool.
1 Like
I want to play the Adventurer class 
1 Like
Thou shalt not play this class.
This is actually a real class. It’s class 0, the non-class. It exists as a debugging tool. It’s existed since forever - since before the game had classes.
It’s the class equivalent to “Which graveyard should I respawn at? Uhh… Westfall I guess.”
Except this is much simpler so that never happens.
dont see many Aug evokers anymore
Ide say your data set has a few really big issues
1 it has a duplicate class problem. For example I have 6 paladins and 1 evoker. The evoker is by far my main and the one I play the most but in this data set my paladins would have a 6x bigger impact
2 evokers are currently limited to one person realm and you need to have a level 50 of another class on the realm to be able to make one.
It’s this combo that really makes the population sizes meaningless. Really we need a way to view it at and account by account level at best you could use snap shots of this data to see how populations change over time relative to eachother.
Take my example. I do not play my 6 paladins I couldn’t even tell you what spec they arem they form part of an alt army who’s only job is to do weekly transmutes. They spend all day in valdrakken and likely encounter a lot of players using the add-on so are included in your data set
Now if we look at the other data set based on doing season content. You will see 1 entry for my evoker and 1 entry for my holy paladin. The two classes I actually play.
In this instance the dataset provided by raider.io based on blizzards own leader boards.more accurately represents what I am playing.
So in the context of what is the least played class this is more accurate.
You could make the arguement that all 6 of my paladins should be included because I would have had to play them to get them to max level. But the reality is most of my 30 or so level 70sbgot to max level in an hour tagging random mobs in elemental storms. And I would say that dosent count as playing a class for an expansion
Really the first thing we should do is define what counts as playing a class for an expansion. And what do we mean when we say least played class. Are we interested in the class that has the least people log into it including transmogs runners and alt army’s or are we examining it though the lense of seasonal content.
For me personally I would say it should include any character engaging in PvE and PvP content.
At which point data sets like easier io form around 1/3rd of the picture. The remain consisting of raiders. Pvpers and people who PvE but not in keys like m0 only players
Sure, although it should be noted that since it looks at what people are actually playing, if you make 6 ret paladins and don’t play any of them then they won’t show up.
But true, Evokers are an issue here. That’s probably why their popularity is so low.
But I would expect the data to be biased equally in favour of all other classes, so besides Evokers it’s not a big deal.
That’ll have the same problem. If anything it’s even worse because now we see all the alts that aren’t being played instead of just the ones that have been playing at least once or twice.
So yes, I am making that argument.
But yeah I agree that we need to define what we mean by “playing”. Are we talking about what people’s mains are? Are we talking about characters being played at least X amount of hours? A week? A month? A year?
Well, some people don’t even have a main. But I think we can pretty reliably make the call that most of the characters we will see are the characters most people play most of the time.
oh my a reasonable response on the forums. isnt that against the TOS
joking aside I think we can probably safely state that depending on the lenses, the brand new class evoker. rogues or monks are probably some of the least played classes.
1 Like
Feels like it sometimes not gonna lie…
Yeah, I think you’re right about that. It’s not 100% clear - there are biases and here, but I think it’s pretty close if not the truth.
All my paladins show up on Data for Azeroth, because I pull their info every time I refresh.
I’ll save you a full screenshot of the whole lot, it gives you the idea
You probably deserve a world first.
Holy is currently more suited in this dungeon rotation more than Disc.
Disc struggles to heal single-target dot encounters, such as the ones in Uldaman and some other bosses are super annoying to deal with as Disc.
Druid or warrior probably.
Just behind you.