Average Class TOC DPS List

#0 - Oct. 22, 2009, 1:26 p.m.

Due to suggestions I made more detailed rankings
Looking at WoW Meter Online, taking the top 20 sample of every class for just heroic TOC 25

Class Average 25man TOC DPS
DK Unholy 10643
Mage Arcane 9698
Warrior Fury 9069
Mage Fire 9000
Rogue Combat 8977
Hunter Marks 8862
Paladin 8734
Feral 8377
DK Blood 8192
Warlock Destr 8026
Rogue Assas 7953
Hunter Surv 7903
DK Frost 7807
Elemental 7772
Warrior Arms 7703
Moonkin 7672
Priest 7624
Enhance 7493
Warlock Demon 6989
Rogue Sub 6857
Warlock Afflic 6486
Hunter Beast 4830
Mage Frost 3295

So, as you can see. Considering they are a hybrid class, DK's are probably over 2000dps overpowered
Warlocks are probably a little under powered.
Enhance Shammies, Shadow Priest, Moonkin, and Ele Shammies appear a bit underpowered, although their raid buffs may balance this out a bit
Warriors may be a little over powered, rogues maybe a little under

Just as 1 more factiod, if you take out Faction Champions for DKs their dps is still:
10268

Original Post:

Looking at WoW Meter Online, taking the top 20 sample of every class for regular and heroic TOC 25

Class Average 25man TOC DPS
DK 10744.335
Mage 9863.53
Warrior 9139.19
Rogue 9003.075
Hunter 8984.69
Paladin 8772.475
Warlock 8440.065
Feral 8342.22
Elemental 7916.775
Moonkin 7793.135
Priest 7667.815
Enhance 7570.69

Pure DPS 9072.84
Hybrid DPS 8493.329375

So, as you can see. Considering they are a hybrid class, DK's are probably over 2000dps overpowered
Warlocks are probably a little under powered.
Enhance Shammies, Shadow Priest, Moonkin, and Ele Shammies appear a bit underpowered, although their raid buffs may balance this out a bit
Warriors may be a little over powered

Ghostcrawler

#216 - Oct. 23, 2009, 2:08 a.m.

Q u o t e:
Threads like these are the PvE equivalent of quoting Arena representation.

Yeah, it's a bit of sticky wicket for us. I try to say "posting with numbers makes your argument stronger," and that's true. Some players then just pop out a link to Simulationcraft or Wowmeteronline or whatever. *Nothing* is wrong with *either* of those tools when used correctly. The problem is too many players use them as the true and accurate measurement of what someone's dps really is. What someone's max dps really is is essentially unknowable. It varies a lot depending on the encounter, gear, raid synergy, the person behind the keyboard, latency, whether someone was trying to inflate the numbers, and a host of other factors. It is almost certainly *not* what *your* dps is. If you are not topping the charts, it is rarely because of the factors measured by tools such as these.

So on the one hand, it feels unfair for us to say "post numbers" and then reply "no, not those numbers" when players try it. But we do ask that you consider what impact your post is going to have. There are several widely available tools, spreadsheet, simulators and websites of raid logs. Most likely the other players you are talking to, and particularly Blizzard, are familiar with them all. Ask yourself if you are really sharing new information when your post consists of a very brief "See? We are low. Here is my quote or link." At the very least try and provide some analysis if you can.

Not everyone is going to be an expert theorycrafter or mathematics expert, and we're not expecting you to be one in order to participate in these forum discussions. But also be honest about the fact that linking to someone else's work doesn't necessarily turn you into one either.

Was that too harsh?

Ghostcrawler

#218 - Oct. 23, 2009, 2:10 a.m.

Q u o t e:
Too harsh.

Aw. :(

Ghostcrawler

#310 - Oct. 23, 2009, 4:38 p.m.

Q u o t e:
For those that do understand the limitations etc. however I believe it's a pretty useful tool when used correctly.

Yes, it's absolutely a useful tool and I'm sure it will just get better over time as it gets further refined. The way I would use it would be to ask "What happens if I swap out this talent or piece of gear or used this spell?" or "With the 3.3 changes to Shadow priests, what happens to my dps?" The way I would not use it would be to say "Simulationcraft shows Demo warlocks at 11,000 dps and Desto at 9000 dps. Please buff me."

Ghostcrawler

#314 - Oct. 23, 2009, 4:45 p.m.

Q u o t e:
So much to say. So little space. If you dont like feedback from posters, why do you encourage us to post here? Just so that we can be sniped at like above? If the source we use are not good, please build a tool to measure DPS/healing/other such stats in the blizzard UI.

We do like feedback from posters, or I wouldn't spend so much time here. "Buff me" is generally not useful feedback.

I'm not sure it would be a good idea to build in some kind of damage meters into the game. While we could probably reduce your add-on lag quite a bit, it also feels like we would be endorsing the meters as a source of competition instead of grouping for instances promoting cooperation. We would also see even more posts of "Here is a screenshot of my dps. Please buff."

"Please buff," doesn't give us much to work with. We aren't asking you guys to balance the game for us. We balance the game. We do like getting feedback from players. "This talent is hard to fit into my build," is good feedback. "I have trouble with rogues in PvP," is useful feedback. "I run out of mana on long fights," is useful feedback. "This is a fun spell," is even useful feedback. "Buff me," rarely is without a whole lot of extra context.

I continually struggle with being able to communicate the difference between "We like to get your feedback on WoW classes," and "We'll make whatever changes you guys want." A lot of posters think the two are the same.

Ghostcrawler

#316 - Oct. 23, 2009, 4:48 p.m.

Q u o t e:
Thread decays over the weeks and sometime down the line we are told the data isn't meaningful because the fight is based on either a gimmick, an issue that supports X classes abilities more than Y classes, because it was a fluke, because of differing player skill, or because the fix for said classes issues are in the pipeline, but not expected until the next major patch/expansion. The very existence of this board masquerading as a place for us to go and raise legitimate class issues infuriates me because I know its all nonsense. Its overrun by trolls, barely moderated, and likely mocked at the company water cooler daily.

This is going to sound harsh, but given the tone of your post I'm going to assume you're comfortable with harsh. This basically reads to me as "Because you didn't make the changes I wanted, this forum is a farce."

Ghostcrawler

#319 - Oct. 23, 2009, 4:53 p.m.

Q u o t e:
This is the TOP the very best of every class in the current raid dungeon.

No, it's actually not. It's "Who got the biggest numbers on a boss encounter." Those are pretty different things. You're equating skill with big numbers. There are a lot of way to get big numbers on a fight that are largely irrelevant to whether you are contributing to your raid's attempts to down a boss. Consider that the biggest numbers are rarely ever going to come from the most challenging encounters or first attempts. They are going to come from when things are overgeared or on farm, when every possible variable lined up perfectly to give someone unusually high dps (for it iwasn't unusual, it wouldn't break records), or when people are trying to pad the meters. Go try Living Bomb on Emalon adds or Ignis constructs. They make for pretty, and pretty meaningless, numbers.

Again, I am emphatically not saying that no number can be trusted. I'm saying don't use lists of who has ever done the most damage to say your individual dps is low in your individual raids.

Ghostcrawler

#321 - Oct. 23, 2009, 4:57 p.m.

Q u o t e:
A sim can give you a pretty solid estimate of performance in a specific situation. The more likely it is that a player can actually perform as predicted in that situation -- meaning mainly, the simpler the encounter -- the more accurate the sim is likely to be.

But it's only a starting point. From there, you need to mess with what the sim can do to simulate variant scenarios, and check that against your own reasoning, and accept that these estimates are going to drift further from reality as the encounters get more complex.

If a simulator shows that Spec A is 30% behind Spec B in a simple encounter, Spec A probably really does have a problem (particularly if multiple simulators agree that it's pretty close to 30%). If a simulator shows that Spec A is 10% behind Spec B in a simple encounter, that's a much more difficult thing. If Spec A gains on Spec B as you introduce movement or burst phases or something like that, then maybe it's actually in a pretty decent place. If Spec B gains on Spec A in cases like that, then Spec A probably still has problems. But figuring that part out requires a lot of discussion (and hopefully testing) to hash things out, and the sims can only go so far to help you with that aspect of the analysis.

Quoting Lhivera.