WoW realm status - WoW servers population (wowanalytica.com)
For those wanting to fight over who is more popular.
What hard data are you looking for? At least he provided some form of factual data.
Itâs not? Youâd just assume one class to decrease as difficulty increases - at a faster rate, than anotherâŚfor no apparent reason?
I wouldnât. Iâd assume that is how statistics is supposed to workâŚunless thereâs some other reason for the decrease â such as inability to survive or the inability to put out meaningful damage became more apparent as the difficulty got higher. That would account for such a drastic absence in one class over another as that difficulty increased.
The fact is - yes - that is how statistics are supposed to work. If 20% of the people of one spec, play another spec - youâd expect that percentage to remain consistent throughout the difficulties - unless another reason was present⌠which is the case here.
I can put it into other termsâŚif that will help you understand.
At key levels 20 and up:
- Fire Mages represent 4.8% of the dps population
At key levels 25 and up:
- Fire mages represent 8.9% of the dps population
At key level 30 and up:
- Fire mages represent 22.2% of the population.
Whereas Elemental basically falls from 1% to less than 1% to even less than that as the key goes from 20, to 25 to 30.
That is showing statistics (the practice or science of collecting and analyzing numerical data in large quantities, especially for the purpose of inferring proportions in a whole from those in a representative sample.). What summation you derive from it - is up to you. Iâm providing accurate numbers, where to me the reason is obvious. Sorry if you disagree (but that doesnât make you right and the data wrong [lol]).
- Former data scientist for the US Military, and current data scientist for the government (And yes, I do understand how statistics and data work) and Iâm not blind to the reasons for an otherwise unexplained difference between two members in that set of data. (PS. Not a brag - just a point that - I definitely, 100% understand how statistics, data, and probability work).
4 Likes
If the relative representation of ALL classes goes down as difficulty goes up, then what are people playing?
4 Likes
Imagine doing the equivalent of âHigher ice cream sales led to higher murder ratesâ type statements and saying youâre a data scientist.
1 Like
You should be embarrassed for admitting this. You have no grasp on statistics and are sitting here literally drawing false equivalencies. You should be placed on a PIP and booted out of the government with this level of ineptitude. Itâs people like you that give government workers a bad name.
I love it that people are discrediting someoneâs interpretation of the data without providing their own interpretation. Kind of hard to take that seriously.
6 Likes
Itâs one thing to provide an opinion based on descriptive data that is prefaced with it being an opinion. Unless you collated these data points into a chart with proper notation of demographic data because all of this is nested you would need to identify the origin point of each data point to do any type of linear modeling.
Any interpretation of the data presented is simply an opinion devoid of a causal explanation that is meaningful because you canât test a hypothesis in a meaningful way to determine a causal relationship. It again simply the equivalent of looking at patterns of ice cream sales and patterns of murder frequency and stating that ice cream sales are related to murders because they spike at the same time. And the situation he is discussing is similar to that thought experiment where ice cream sales and murder rates go up in a similar fashion because both of those things tend to increase during summer seasons. He fails to provide that level of nuance to his âinterpretationsâ.
You combined this with him derailing other threads because he has some vendetta about Ret being more popular than Shaman DPS for some reason. And to add on top that his opinion is that of an expert is why he deserves to be discredited because if was one, he wouldnât even be attempting to make the point he is making. Any point he makes may qualitatively have some relevance but to say that it is indicative of a causal and meaningful relationship based on data makes it a quantitative matter. To which he is off the mark.
Except any rational person can tell that ice cream sales vs murders is a ridiculous comparison to make against class representation vs class performance. This isnât murder or some complicated real world study. Itâs class representation. There are only so many ways you could possibly look in to this.
4 Likes
The ice cream vs murder example is an example of correlation =/= causation. AKA two variables that on a descriptive level are related on a data point level in terms of frequency and direction.
His point that ret vs sham representation shows something wrong is only looking at this from this expansion. If he wanted to do a proper analysis, he would look at every season since BFA and track changes in class representation. There is a whole historically based explanation to his points that he completely misses. The major one is that Ret generally has more representation than Shamans due to reasons unrelated to balance and gameplay. Some people get a hard-on for the divine justice RP and âDEUS VULTâ memes.
Class representation in a patch is the cause of 100 variables with multiple moderating and mediating factors.
So if the drastic representation differences arenât class performance based, what is a better explanation? We know the holy warrior archetype is popular. So letâs shave off two thirds of Ret representation in this data sample. Why are they still clearly out pacing Enhance here? Thereâs only so many reasons. 1) Ret is OP, 2) Enhance is UP, 3) combination of both, 4) Ret is mechanically more fun to the player base at large. I mean obviously there are other variables like if mages are OP then hero wonât be needed, whether the healer spec of paladin/shaman is taken a lot etc. but you get the point, I hope.
Itâs easy to discredit others without providing a counter argument.
5 Likes
For the third time since reading is still hard for some.
Haha, youâre joking right? Your explanation after all the name calling to the other poster is âitâs always been that wayâ?
6 Likes
Oh come now, Antaris- you know that every great debate throughout history has been settled with ad hominems, repeating things hoping they come true, false equivalences, refusing to answer the question, and a well-timed ânuh uhâ.
5 Likes
Personally I just canât be bothered with Enhance/Elemental in its current âMetaâ iteration, bring back the Storm.
This is why Iâm praying Stormbringer is my saving grace for both specs.
But this is just a personal opinion on why I stopped in S3 and S4 moving forward.
Iâm confused by the people arguing against the numbers.
If you assume at difficulty level 0 class A and class B each make up %50, youâd say with fair certainty that at that level they are for all purposes âequalâ
However, if you move to level 1 and there is a shift in favor of 1 of the classes, you could imply that it is because they are no longer âequalâ as the difficulty goes up.
If that shift continues to a point where itâs 90-10 at the top end, you can conclude with certainty that 1 class is far superior.
This applies in pve and pvp alike, there is something called the meta for a reason.
You people who want to argue with the numbers because âwe canât prove why anyone plays anything at any levelâ are the same people who think that all the ret players got good at pvp magically during retpocalypse or that the reason demon hunters far out number enhancement shamans at the top of the ladder is because demon hunters are better players, couldnât possibly be their class.
8 Likes
I donât understand why youâre so mad at me for posting numbers and then an observation. You do realize that my observation is basically âEnhance doesnât perform as well at high keys as Retâ - I mean itâs literally just fact anyway.
4 Likes
Youâre literally picking the meta classes and saying thatâs the reason enh isnât played as much. No sh#$ sherlock.
You also continue to ignore the fact that enh and shaman in general have always been played way less.
You also ignore the fact that as you go up, far fewer people participate. At the same time as you go up, far more of those people who do participate will run the current meta classes. Watching the few of you posting these numbers and not understanding meta classes will see higher rep in higher keys is dumbfounding.
Youâre drawing false equivalencies because you donât understand or are willfully ignoring the reasons the rep doesnât stay static. The fact that enh has just had the best xpac its probably ever had regarding performance and yet you still canât grasp this says a lot.
What âfalse equivalencyâ did I make?
Are you saying that a meta class such as Fire Mage or Ret Paladin doesnât perform better at the highest end than Enhance?
Are you saying that Enhance doesnât have inherent problems that make it not meta?
Iâm just confused on what this âfalse equivalencyâ isâŚlike what did I post, exactly that is making you this upset with me?
The two classes that Iâve posted numbers for (Ret and Fire), their representation goes up as the difficulty increases. Enhanceâs goes down. Iâm looking for the reason. Iâm not posting some insane theoryâŚother than the fact that Enhance has problems lowering its effectiveness at those higher difficulties when compared to a meta spec. Thatâs literally why theyâre meta lol.
Stay mad I guess.
4 Likes
Iâm saying youâre not too bright for picking meta classes and then complaining that youâre spec isnât repped as much as a current meta class. Why donât you go back and start linking retâs numbers before the rework⌠Go on, Iâll wait.
The irony from someone unable to grasp something so simple is astounding. Iâm not mad because you lack basic logic.