How Ele shaman and Enhance stack up VS other specs per WoW Progress DPS Sims

Classes Shaman
1

Using the WoW Progress website https://www.wowprogress.com/ you can see simmed dps https://www.wowprogress.com/simdps/us based off of the simulationcraft website http://simulationcraft.org/ . This link will actually take you to a page that explains how it actually works and how they get their information https://www.wowprogress.com/post/138_WoWProgress_launches_DPS_Rankings_based_on_SimulationCraft

So now lets breakdown some class/spec info. How I will attempt to do so is to show the dps variance between the top 100 sims and what makes up the most common spec of each and then relatively speaking how far behind other specs are from the top contender.

If there are certain specs that are overshadowed by a certain spec of the same class; I will go deeper into the list to determine where that spec picks up and starts to shine. In doing so, maybe we can shed some light on the current state of the two shaman dps specs and determine if one happens to be as bad as what most people claim.

Notes

  • These are Sims not actual dps.

  • What i consider consistent representation is at least 3-4 appearances per page which is roughly 10-13% for 3 consecutive pages. I do this because if there is one spec that is outperforming other spec(s) then the number of people playing that spec will cause inconsistencies in the representation of the other (lesser played) spec. I can alter this is someone else has a better solution; as wow progress doesn’t break down specs individually.

  • Melee has a tendency to fall behind at the late xpac stage due to scaling. However they are typically buffed late for compensation. Historically melee start xpac strong as soon as they get weapons and trinket upgrades. Afterward they plateau off as ranged continue to gain ground due to scaling.

DK

25507.70-23939
Predominantly UnH by 75-80% for most of what I can tell (checked top 1000 and it was pretty consistent)

DH

27875.30-26812.60
One DPS spec

Druid

30847.20-29558.60
No feral in top 170 and they don’t become consistently present until <25k.
There was one feral at 29299.60

Hunter

28937.10-27832.40
Predominantly Marks and Beast with the majority being Marks. Survival doesn’t become consistently present and I checked down to 24k dps.

Mage

28517.10-26885.30
Top 100 all arcane except 4 frost. Frost is seen more prevalent <26k dps. Fire isn’t even seen consistently until <24k dps and then it still isn’t a lot.

Monk

26462.00-25491.10
One DPS spec

Paladin

27482.50-26240.00
One DPS spec

Priest

28295.80-27033.30
One DPS spec

Rogue

28067.60-27493.80
2 outlaw in top 100 the rest Assassination. 3 sub in top 500. Outlaw becomes more consistent at <25.6k dps. Sub not consistently seen above 23k dps.

Shaman

30554.20-28676.70
All Ele in top 100. 3 enhance in top 500 with dps 28600.00-27888.60. Enhance not really seen consistently until 26.4k dps

Warlock

28403.80-26761.70
2 Affliction in top 100 the rest Demonology. Affliction not consistently seen until <26k dps. Destruction not seen until <25k DPS and then it isn’t consistent (probably due to the popularity of the other two specs).

Warrior

29030.00-27582.30
Fury only top 1379. First Arms Warrior 1380 with 25667.80 DPS. Arms doesn’t become consistent until <23k DPS.

Ranged DPS

Boomy
30847.20-29558.60

Ele
30554.20-28676.70

Beast/Marks
28937.10-27832.40

Demonology
28403.80-26761.70

Arcane
28517.10-26885.30

Shadow
28295.80-27033.30

Frost
26.5k

Affliction
26k

Destro 25k

Melee DPS

Fury
29030.00-27582.30

Assassination
28067.60-27493.80

DH
27875.30-26812.60

Ret
27482.50-26240.00

WW
26462.00-25491.10

Enhance
26.4k

Combat
25.6k

UnH DK
25507.70-23939

Feral
25k

FDK
<24726.40

Survival
24k

Subtlety
23k

Arms
23k

Discussion

  • Ele is in a really good spot as well as Boomy. I llok for both to get nerfed in the next xpac. Ele especially as some of the new azerite may push Ele even further.

  • Enhance “on paper” doesn’t seem to be as bad off as everyone claims.

_/pops popcorn :popcorn: and grabs a very large drink :tropical_drink: and a nice comfy chair :couch_and_lamp: _ :grin::grin:

2

Wowprogress sims are not simulationcraft sims. They’re a bad joke.

Because it’s single target damage was pretty good, the problem was getting nerfed twice.

So are you aware you basically just spend thirty minutes making a list that literally has no actual meaning.

3 Likes
3

Wow someone is salty as my :popcorn: so which is it? Not reliable or patchwerk fights? For someone who doesnt play shaman and has zero clue besides “interrpretting data” you sure got a lot of opinion.

4

Sure.

Sure.

You’re using wowprogress sims to back what you call data.

Case closed.

1 Like
5

I’ve been saying this (pve wise). Not that it’s top tier, but Enh isn’t holding back your raid. I get why top 10 guilds aren’t using them, sure. For the vast majority of guilds, Enh can be a contributor to your progress. Sure, Ele is in a better state at the moment, but I’m garbage at the spec, so I stick with Enh.

PVP wise, I think Enh definitely needs some defensive buffs, but that’s off-topic,

6

So this is a case of; you have zero proof to counter my points and it goes in direct conflict of what you “feel” so it is wrong?

Typically when someone gives info you have to counter with actual proof that said info is false or inaccurate. Where do you see this is bad info?

Sort of like saying hey look at all theses photos of the round earth and you saying come on all that info is wrong the planet is flat.

7

The sims don’t actually match what the people actually sim for?

Tell you what. Go into any of the class discords. Literally any one. Ask about wowprogress sims.

Get laughed at.

Every single theorycrafter and apl writer for Github publicly condemns the wowprog sims?

8

Im asking for a source? Just because all the fanbois dont think its cool doesnt mean its wrong.

9

Seeing as how I do not write sim apl, don’t ask me. Ask the experts who do it for a living.

10

tell that to the m+ group leader that asks if you are ele then denies you because you arnt.the meta is the reason to blame aside from bad class design.

11

Troll does what troll does

1 Like
12

yep most people in m+ are toxic i agree.

13

Not saying wow progress is better or more accurate than Raidbots; just using a comparison for validity.

Dimira Hyjal

Raidbots quick sim
29,431 DPS
https://www.raidbots.com/simbot/report/rBteyFPRJyZWRCU59DP1HB

vs

Wow progress sim
28600.00
https://www.wowprogress.com/simdps/us/class.shaman

vs

Warcraft logs Mythic BoD
20,723.6-47,330.5
https://www.warcraftlogs.com/character/us/hyjal/dimira?mode=detailed&zone=21#

Revze Mal’Ganis

Raidbots
28,997
https://www.raidbots.com/simbot/report/eajs4P5puq9mLxwHQtpiBx

vs

Wow Progress
27888.60
https://www.wowprogress.com/simdps/us/class.shaman

vs

Warcraft logs Mythic BoD
18,214.0 -42,666.8
https://www.warcraftlogs.com/character/us/malganis/revze?mode=detailed&zone=21#spec=Enhancement

Sturr Spinebreaker

Raidbots
28,700
https://www.raidbots.com/simbot/report/5iSpf6WCPAgsJ7UtQ4okAp

vs

Wow Progress
27492.10
https://www.wowprogress.com/simdps/us/class.shaman

vs

Warcraft logs Heroic BoD
15,175.9 -39,601.0
https://www.warcraftlogs.com/character/us/spinebreaker/sturr?mode=detailed&zone=21#

Gelato Nagrand

Raidbots
27,428
https://www.raidbots.com/simbot/report/7kseVpqvC4yM9eKzYzYVA9

vs

Wow Progress
26617.80
https://www.wowprogress.com/simdps/us/class.shaman

vs

Warcraft Logs Heroic BoD
13,457.1-40,216.4

Still adding more and the data doesnt seem any different from my claim. Yet it is amazing no one is actually adding ANY data to dispute all of this

14

You can use me as a source if you like. I am an Enhancement Theorycrafter and have done extensive work on the APL for years and develop and maintain the Enhancement module in SimulationCraft.

@Madalynn:
WowProgress sims are low iterations (1000) which results in a high error percentage. This impacts some specs much greater than others. It also includes many characters with different gearing strategies and differing class populations. From a purely population perspective, if class A is competitive to top dps but the least players play it, you’ll see a disproportionate number of spec A compared to everyone else, and will likely draw the wrong conclusions. You also end up with bloated population lists thanks to gear RNG on higher population specs.

Here’s the thing with the data set you’re trying to compare. When looking at simulated data and comparing specs/classes, you should try to eliminate as many variables as possible to get a more “fair” data set. By introducting player agency, you’re introducing a ton of extra things that taint any kind of comparisons you want to make. That’s why the Simulationcraft stacks always use a set of rules to create a baseline that’s the most fair for every spec. If you’re looking for “real-world” data, WarcraftLogs takes into account both player agency, player execution and encounter styles to create ranked averages by purely looking at the end result. This works because its not attempting to mix simulated data with player agency.

And on top of it all, cherry picking datasets the way that you are does not actually create any meaningful data. You try claiming that:

In Statistics, there’s something called a “null hypothesis”, which states that “there is no relationship between two measured phenomena, or no association among groups.” What this means is that there’s a burden upon you to disprove the null hypothesis and establish that there is a correlation here, which you have not even remotely done.

TL;DR - Wowprogress is a very poor source of data for measuring accurate trends in character / spec / class performance, you have not actually created a hypothesis (I don’t even see an attempt beyond “enhance and elemental both exist in a dataset that I made up”) and I don’t see any attempts at a conclusion.

6 Likes
15

I appreciate the assist, and the in depth look. If this comes up in the future, as I know it will in some forum, I’ll keep you tabbed. Unless, of course, the site decides to improve it’s sims, in which case it’d be a moot point.

16

The burden of proof still lies on you since you are making the claim not Drez. If someone questions the validity of the source of your data then you should be able to provide the information that it is accurate. Without doing this, or being able to, it puts into question the justifications and methods you used for coming to your conclusions and undermines your argument.

17

From the looks of comparing Wow progress sims to Raid bots sims they seem to be pretty similar.

This attitude of burden of proof is exactly why there is a measles epidemic. There is literally tons of data supporting vaccines. However, one moron writes a paper that has been disproved repeatedly and supposedly educated people stop vaccinating their kids and now measles (which wasn’t even considered as a concern due to vaccines) is now a major health threat. The guy in question didn’t have to prove his position and yet countless scientists/medical professionals provided data (factual evidence) that vaccines don’t cause autism and are crucial to preventing disease spread in society. And yet like here people casually denied the provided data as false without actually providing any real proof said data is false.

And I love the fact someone went to a fanboi site to recruit help and said help actually provided nothing of the sort but some proclaimed unproven status followed by some dribble he/she probably found on Wikipedia. Please keep said person on tab. Hell that is probably just your alt shaman you claim not to play.

My hypothesis is that enhance dps numbers don’t suck. The null hypothesis would be that enhance dps numbers suck. All the data that I have provided is that in fact enhance dps numbers does not IN FACT not suck. This is proven by sim dps similarities between wowprogress and raidbots. to further prove said validity I present the following link to warcraft logs since in fact that appears to be valid according to our self proclaimed enhance theorycraft expert above.

https://www.warcraftlogs.com/zone/statistics/21#difficulty=4

BoD Normal Standings current 99th percentile
24,399.99-27,481.81

BoD Heroic Standings current 99th percentile
28,056.33-34,865.86 dps

BoD Mythic Standings current 99th percentile
30,846.58-32,657.25

CoS Normal Standings current 99th percentile
18,094.43-26,820.10

CoS Heroic Standings current 99th percentile
22,545.14-32,025.80

CoS Mythic Standings current 99th percentile
34,904.90-34,904.90

I couldn’t get warcraftlogs to give me dps numbers for mythic+

The only place where what data I was presenting and what Raidbots and/or warcraftlogs differs is Mythic CoS where enhance is currently ahead of only Boomy and Marks. However that seems pretty skewed from all the other data.

I still stand by what I have said this post. All of the other standing has enhance right where the sims says it is. Enhance isn’t as bad off as what most people want you to believe.

18

No one ever argued that. Perception of the uninformed public will always be inaccurate.

19

What? No, look, you can’t try to rationalize what complete morons think and do. There’s not a measles epidemic because of a burden of proof, but there is quite literally all of math and science thanks to that “attitude”. The difference with your “data” is that we questioned your source with completely reasonable concerns. You’re pushing a conclusion that is entirely on you to prove, it’s not healthy to accept your data as fact. Ironically, that’s why there’s vaccine deniers, who can’t think critically for themselves by accepting a conclusion without questioning it, either internally or externally.

I did not come here bidden by anyone. I saw this thread that contained some concerning attempts at suggesting a conclusion and decided to weigh in with a more detailed response as to why you’re seeing some resistance from people. I am noone’s alt shaman but my own.

You misunderstand the concept of a null hypothesis. A null hypothesis isn’t a hypothesis it all, it’s that the default position of all statistics is that there’s no correlation unless explicitly proven otherwise. And great, you gave us your hypothesis, “Enhance dps numbers don’t suck”.

Wowprogress and raidbots will have have similar outputs because over a very large population (millions of users) there will be enough geared players to create a large enough data set to mimic a simulationcraft “stack”. And both wowprogress and raidbots use Simulationcraft to run their numbers. This doesn’t prove anything - it’s not 2 datasets correlating a trend, it’s the same dataset (simulated data from Simulationcraft). All it proves that on paper, on pure Single Target, classes stack up a specific way. The issue is that this data (wowprogress/raidbots/simcraft sims) does not correlate with in-game performance since the amount of Single Target content in the game is effectively zero. Mythic+ contains loads of AoE. Every single current encounter in BfD contains either multiple targets, forced downtime, inconsistent movement, mechanical responsibility or target switching, all of which are factors that cause SimulationCraft to not be the “end-all be-all” of what is good and bad. There’s also the issue that there is a large gulf in effort between maintaining spec profiles in Simulationcraft sample profiles - It is not a joke that some specs contributors just don’t put as much effort in, whether done so intentionally or not.

You linked aggregate spec scores, then pulled some numbers from a completely different place. But lets look at some other aggregated data:

https://www.warcraftlogs.com/zone/statistics/21#difficulty=4&amp%3Bclass=Shaman&amp%3Bspec=Enhancement&amp%3Bdataset=99&class=Shaman&spec=Enhancement

This is a comparison of Enhancement’s aggregated score at different percentage parse levels. Considering that Enhancement has one of the highest upper variance potential, lets ignore the 99% and take a look at 95%, a much healthier population size. As player skill (across all specs) increases, Enhancement’s aggregated score decreases compared to the aggregate score of all other 95%+ parsing specs. As player skill increases, Enhancement does worse than other specs. This is a very concerning indication of a downward trend as players reach their skill potential.

Let’s address your “hypothesis”:

I disagree with this conclusion. Enhancement actual looks reasonably decent on Simulationcraft “stacks”, but fares extremely poor on WarcraftLogs aggregated statistics (conclusions that come from mountains of actual data). Shadow Priests look extremely poor in Simcraft, but then perform absurdly well when applied to difficult content in-game. I absolutely think that Enhancement’s resulting performance is poor overall, and I will back that assertion up with WarcraftLogs’ aggregated statistics which is a more reliable outcome of real performance than Simulationcraft (though its not without its faults, as every dataset inevitably is).

Enhancement is exactly as bad off as what most people believe, and the issues don’t entirely stem from just DPS. There are a lot more factors that go into desirability in content like survivability, mobility, and utility that contains little to no overlaps. When taking additional factors into account, it has to compete with its other DPS spec, Elemental, which is currently worlds ahead on DPS, has better survivability and identical utility and happens to be range to boot.

That isn’t to say it’s unplayable, because you can certainly walk into raid with it, there’s just no compelling benefit to playing it over nearly every other DPS in the game, and you still do less DPS on average in the process.

I think your issue is that you made a conclusion first (Enhancement has good DPS) and then tried to find data to support that conclusion and never bothered to evaluate the data. That’s not how statistics or the scientific process works. You’re supposed to make a hypothesis, conduct iterative tests (you didn’t do that, you picked data you wanted to see), see if the data supports your hypothesis and then make a conclusion. You skipped straight to conclusion and then tried to find data to support it.

5 Likes
20

You don’t even need to go into the actual dps numbers that far, you can just look the number mythic parses for BoD and see that Enh is 3rd from the bottom with ~4k appearances, while Ele has ~38k appearances. That right there tells you something. Not only that, if you look at CoS mythic there are ONLY 2 Enh logs… Strictly using simulation numbers is never going to show you the whole picture (especially if said simulation data can’t reliably mimic the actual events they represent). As was said earlier, there are also other factors that simulations don’t factor in, like Utility, Survivability, Encounter Mechanics, and Varying Fight lengths because of RNG.