Short napkin math on this issue. Hope it is right. If it is, this should be something to consider while fighting Hodir. I took your wmo of Hodir and only used the average dmg done by Shred and Rake. If Shred DPE is on average higher than Rake DPE, Shred DPE should definitely be higher while having Storm Buff.
This what i came up with from your hodir log:

Average Shred Dmg: 11806 /42 = 281,1 DPE
Average Rake Dmg: 1712 Direct Dmg + 2266 * 3 Dot Dmg + (1-0,98^3) * 11806 Shred Dmg from 2 T8 /35 = 263 DPE

Ok so it seems that even on average Rake is worse DPE than Shred (even considering 2T8) on you hodir log.I hope I did not make a logical mistake here, but it seems to me it is save to assume that considering your gear level you should not Rake with storm buff up. To maximize your dps Rake without the buff but don't use Rake with the buff.
This is only 100% true for your level of gear though! Although I am pretty sure it is the same for me and many others.

From Nightcrowler a couple pages back:

Nightcrowler, being consistently good DPS isn't something that I would have considered needing a nerf, much like rogues in TBC we felt "right"; good enough that bringing more than 1 was ok if they were good players but not better than "pures" except with swipe-favoring fights (which, as was pointed out in the FA/FI arguments, is only good for giving us big numbers, and does not significantly contribute to defeating the boss)

As for being "among the top of the melee", when comparing among my guild, I tie the Fury warrior, and lose to the rogues on nearly every fight including significant AoE fights, obviously there's variance, but that's entirely due to a particular part of the fight that favors one or the other, I take less AoE damage, my AoE is better, I run out of things faster, get full benefit from hysteria, don't have to interrupt things, and am less affected by haste. This makes me a better choice for what, Vezax? Pretty sure my extra 500 DPS over rogues on that fight makes up for the 8k they beat me on Hodir.

I'm still going to trust WMO more than Blizzard or forum qq in terms of where druids are in relation to other players. If Blizzard is telling us that feral druids may be doing too much DPS, and I can see from numbers that we're still worse than at least 2 other classes on every fight, I'm really not sure they are looking at the right numbers.

Can't really see how you can use Hodir as a reference fight? It depends alot on RNG. If you should use a fight in ulduar to compare melee dps, I would use Ignis. And I would argue that cat is about 400-500dps above where we should be, considering us being a hybrid class.

Take a look at this for example: World of Logs - Real Time Raid Analysis our last ignis. Sure I was a bit lucky, but still.

On that parse, either your rogues fail or are undergeared. Rogues do deal more damage than cats, except on fights which give us a particular advantage, such as XT, mimi, yogg or vezax.

First - azorac: single Logs aren't really useful. Nobody knows if you raid with a bunch of stupid retards or with exceptional professionals.

I agree with Boevis - WMO has a fairly great database and comparing numbers there is much more reasonable than single logs, forum qq or "general opinion".
There are, in fact, several problems when comparing the top 50 of each class:

- possible bugs while converting the combatlog data
- not clear, if the players played properly (e.g. AE on the Ignis constructs instead of single target DPS)
- Bugfixes and/or classchanges are not reflected, as the database is not resetted every time
- Some class could have specific roles (e.g. rogues interrupting on vezax)
- Chaining buffs like power infusion, TotT or Hysteria will be necessary to be in the "top50 of each class"


Nonetheless - it's an average of many logs, which is far superior than single logs can ever be. Nobody is interested, if you as a feral top the DPS meters in your guild. But it would be interesting, if many ferals do top the meters.
And according to WMO that is definitely not the case.

On the 10th of June I analyzed the data given by WMO. Obviously the top50 lists changes everyday, but the order should be roughly the same.
This is a summary about all Ulduar bosses.

Rogue: Average rank: 2.4, Average DPS: 7200
DeathKnight: Average rank: 3.9, Average DPS: 6817,2
Warlock: Average rank: 4,2, Average DPS: 6837,9
Druid: Average rank: 5.4, Average DPS: 6554,1
Mage: Average rank: 5.6, Average DPS: 6808,4
Hunter: Average rank: 5.7, Average DPS: 6365,7
Warrior Average rank: 6.6, Average DPS: 6115
Shaman: Average rank: 6.6, Average DPS: 6556.2
Paladin: Average rank: 7,2, Average DPS: 6044,4
Priest: Average rank: 7,3, Average DPS: 6209.4

Source: WoW Meter Online - Combatlog Replay

Anyone know if there's a way to use WMO to rank top 50 by spec instead of class?

I don't think that's possible.
But you can list the top 20 x for every boss. For x, you can choose all players, players of a certain class or even a certain spec.

WoW Meter Online - Combatlog Replay

I often look at the top20 ferals for example, to compare myself.

The T9 bonuses have been posted at MMO. And...they're not that good.

Rake going from 9 to 12 seconds seems like a fairly decent ability, though it's not as good as something like rip boosts were due to the ease of refreshing it. But 5% more rip and FB crits?

I don't see either of these replacing the 2pT8 bonus unless the gear is seriously well optimized. I don't think doubling clearcasting is ever going to be worth the 4pT9 bonus by itself. The 2pT9 + 2pT8 might be worth it, though.

Well, when a set bonus offers us energy in any form, it's always very superior.
Remember, we were wearing 2pT4 still in Sunwell, because the bonus granted us free 20 energy from time to time.

These kind of bonuses do stack with our DPS that's why they are so extremely important.

They'll likely nerf the T8 bonuses (like they did for DKs) if it turns out they're too good, compared to the T9 ones. I'm sure the same is going to happen for resto T8 because that rejuv ability blows the new 4 set bonus out of the water. All druids would be stuck wearing 4T8 for a LONG time without a nerf.

On another note the Barkskin 4 set bonus is freaking amazing. The growl one is pretty lackluster though (unless there's a lot of taunting encounters in the colliseum).

You have to also factor raw item power into all of this, though. It's a jump from ilvl226 to 258 (assuming your guild is into that) on 2 pieces. That's a pretty hefty jump, well over 2 traditional tiers.

(Although yes, the lesson learnt from T4 being pretty much BiS to Sunwell kinda sticks around)

I understand that single log wont provide as useful information as many. But using top50 (as you said your self), is useless. In my opinion that information is as accurate as my single report. If you take a closer look at the top50 fights for say, Ignis. Almost everyone is using aoe, how will that tell you anything at all? To compare classes you have to look for fights where the players do not use any AoE or getting buffs like ToT and hysteria. And the players you compare need to have close to the same gear as possible.

Boevis, you said that our rogues where doing something wrong. I cant find any prof of that at all, I have looked through all the top fights for ignis and for rogues with about the same gear they are doing close or above the average dps.

Well, lets look at the stat upgrade for each piece (Disregarding Stamina for cat)

Conqueror's Nightsong Headguard = 82 Agi, 75 Crit, 50 ArP, 152 AP
Runetotem's Headguard of Triumph = 137 Agi, 172 AP, 102 Crit, 86 Expertise
Gain/Loss = +55 Agi, +20 AP, -50 ArP, +27 Crit, +86 Expertise

Conqueror's Nightsong Raiments = 97 Agi, 49 Hit, 66 Crit, 162 AP
Runetotem's Raiments of Triumph = 137 Agi, 204 AP, 96 Crit, 92 Haste
Gain/Loss = +40 Agi, -49 Hit, +30 Crit, +42 AP, +92 Haste

Conqueror's Nightsong Shoulderpads = 77 Agi, 62 Crit, 112 AP, 48 Expertise
Runetotem's Shoulderpads of Triumph = 106 Agi, 152 AP, 76 Haste, 68 ArP
Gain/Loss = +29 Agi, +40 AP, +76 Haste, +68 ArP, -62 Crit, -48 Expertise

Conqueror's Nightsong Handgrips = 85 Agi, 43 ArP, 33 Haste, 130 AP
Runetotem's Handgrips of Triumph = 106 Agi, 151 AP, 76 Crit, 68 Hit
Gain/Loss = +21 Agi, +21 AP, +76 Crit, +68 Hit, -43 ArP, -33 Haste

Conqueror's Nightsong Legguards = 114 Agi, 70 Crit, 50 Haste, 114 AP
Runetotem's Legguards of Triumph = 137 Agi, 172 AP, 102 Crit, 86 Hit
Gain/Loss = +23 Agi, +62 AP, +32 Crit, +86 Hit, -50 Haste

Now, from that data comes trying to pick which two pieces of T7 you'd rather keep due to Itemization and not a big increase in stats. The head and Chest provide pretty dramatic increases in raw power stats. You're going to want to upgrade to T9 for those two pieces.

That leaves us with Gloves, Shoulders, and Legs. The hands provide a relatively small upgrade, and the keeping the ArP on the T8 helps to offset the ArP you are losing by upgrading your Helm. Now the only question is, which T8 piece, of the shoulders and legs are best to keep? I can't put into words the math and numbers swirling around in my head, partly because I don't quite comprehend it - but the Shoulders look better to keep at T8 (Which I know sucks, because they are fugly as hell). Upgrading to T9 Legs you gain a ton of hit, and if end T9 gear is anything T8, you will need all you can get on your BiS gear.

So, by leaving your Shoulders and Gloves at T8, you are missing out on the following: 40 Agi, 61 AP, 76 Crit, 68 Hit, 76 Haste, and 68 ArP.

However, you'd be keeping 62 Crit, 48 Expertise, 43 ArP, and 33 Haste (which you would otherwise lose) which leaves your net losses from leaving two pieces of T8 for the bonus at: 40 Agi, 61 AP, 14 Crit, 68 Hit, 43 Haste, and 25 ArP.

That doesn't seem terribly staggering to me. Its up to someone else to do the math on whether those stats are better/worse than having OOC proc off your bleeds. My head hurts enough already.

I'm not sure that's true, exactly. The head at least gives tons of stats, but it's tons of expertise mostly - and it loses armor pen. That might be one of the better places to keep the T8 if you're looking to do so. The shoulderpads are a clear upgrade. The legs and hands are probably a good one, though they have that ton of hit. The chest is another slight one.

Napkin math on 2T9 compared to 2T8:

Rake's previous energy cost (to get 100% uptime) is 35 energy / 9 seconds ~ 3.9 energy/s.
New energy cost (with 2T9) will be 35 energy / 12 sec ~ 2.9 energy/s.

Energy gain by using the new bonus is roughly 1 energy per second, so in 42 seconds, this allows 1 extra shred.

Turning to 2T8 bonus, in 42 seconds we will have 21 rip ticks (1 per 2 sec) and 14 rake ticks (1 per 3 sec) = 35 proc chances. Given that the 2T8 bonus has 0.02 probability to get a clearcasting proc, chance of getting a free shred in 42 seconds is 1-(0.98^35) ~ 0.51.

So the 2T9 rake bonus seems much stronger than the 2T8 clearcasting bonus. Of course, having both is nice..

You want to look at the expected number of procs in 42 seconds, not the probability of getting at least one. That said, the math is still in T9's favor: 2% * 35 = 0.7 procs per 42 seconds. Looking at it another way, 2T8 is an expected 42 energy per minute (50 bleed ticks * 2% chance = 1 expected proc per minute, in the absolute optimal case) while 2T9 is roughly 60 energy per minute (more accurately, 58.333... if you refresh Rake the instant it falls off).

You still don't need much in the way of expertise thanks to primal precision and it's not that great of a damage stat anyway. I'm just saying that using that value as a great stat to keep around is not necessarily valuable. If you're under the cap by that much, sure the expertise is great. But that's not been my experience with the itemization so far.

Why do you ignore AoE and Raid Synergy? "I think I'll ignore everything but Single Target DPS since Feral Druids are terrible at that on every fight but XT-002 and Ignis, and I get to drop the Druid's rank down to 7 or 8" My use of quotes is to indicate me being sarcastic. I really want to go through the top 20 of every spec and find where feral druids stand in terms of single target DPS now, just so the nerf looks even stupider. However, you don't get to just ignore any part of a Boss fight, or Raid Synergy, or PvP, or capability in Heroics with pre-raid itemization, as much as I'd like to, since I can make an even better case about the nerf being uncalled for if I got to remove everything but "DPS on Boss" from the equation. But all that other stuff is just as much of a factor in how Blizzard Balances the game as "How much can you do to the Boss Himself". Every Boss becomes impossible if you don't at least remotely try optimizing the use of classes, some classes do certain jobs better than others, in order to not make everything carbon copy's and the game actually interesting, Blizzard lets every class shine to some degree on different fights in the game.

A single report is extremely useless when discussing what classes are capable of. Maybe you're were just particularly awesome at Ignis that day and other classes were having a case of the stupid, maybe there's a gear difference (there is, you're better geared than me, and your rogues are worse geared than my rogues). On top of those 2 controllable factors, you're also looking at a 3+ minute fight, the RNG potential is massive, and not just with Hit/Miss/Crit, there's weapon procs, energy gains, clearcasting, trinkets, set bonuses, etc. You also have to consider upgrade strength, if asked to design Druid T9, I'd be hard pressed to come up with something other than "like T8 but better" while so many other classes (rogues and healers included) only wear tier gear because of 1. availability and 2. set bonuses, if itemized properly most classes would see massive gains in DPS. If gear is what's keeping other classes back, then gear is what needs to be fixed (and in general, won't harm pvp at all!)

When discussing "Class potential" you have to look at the upper outliers on every fight, not single Meters, not Spreadsheets, and certainly not the hearsay of someone upset they got beat by "a healer class". It's cute and all that SimulationCraft, Rawr, even FbN list Druids as doing 8k potential, and it's useful for comparing our gear. However, until Blizzard puts in Target Dummy XXL The Destoyer of Damage Meters, You have to look at how DPS gets translated into the real encounters. On how many Encounters are Feral Druids the absolute best class? None On how many Encounters are Feral Druids the absolute worst class? One. Out of 10 classes, we rank 5th (once you separate Moonkin from Feral, they boost us significantly on Vezax and Hodir) in terms of average rank per fight, and 6th in terms of average DPS per fight among the top 50 best (and luckiest?) of every class. For a highly RNG based class (though I guess everyone is now) Being as middle of the road as you can get really just doesn't seem "too good".

I also refuse to believe it's "luck" that we're seeing in the WMO Scoreboard, the majority of my argument comes from watching my guild's progress on Hodir (and other fights, just more noticeably Hodir) from "Feral Druids have it easy, you don't lose massive DPS when you have to Move. Those Casters breaking 10k are just getting lucky RNG" to "Any Caster with at least 2 storm buffs that isn't breaking 8k is getting replaced". Naxx and Malygos, even Sarth if you weren't a tank or healer, were just too damn easy, even 6 min Malygos was easy once you figured out the trick, while I don't want to say something like "people just need to l2p", that really is just the case. As I told my guild, Ulduar compounds everything that you do, when you do something wrong, it hurts everything from then on, when you do things right, everything gets easier. People don't get into the top 20 on WMO through RNG alone, they're there because they are doing something that other people aren't (even if it is just gear)

There's also the argument that maybe Druids really are just better players than everyone else, after all, we do have possibly the strongest community support and most accurate resources easily available, you can't even call it hard anymore between just using Rawr and FbN. We've also spent years surrounded by hundreds of misconceptions, and literally prevented from doing good dps by Blizzard (it was stated they didn't intend on Druids being anything but healers in raids), in my case, and I'm sure many other cases, Druids are just more likely to try that extra bit harder. Terrible argument I know, but I'm sure I'm not the only one that remembers the old MC joke where a Balance druid got in the top 5 on a boss, and then when awarding loot 2 mages 2 rogues a warlock and a hunter all apologise for being afk.

I understand that Blizzard can't exactly come out with a Press Release declaring "Feral Druids are fine, You all just suck" But seeing everything in the context of "How good are classes at the various aspects of the game" I cannot fathom how Blizzard justifies nerfing something that ranks Middle in PvE and practically dead last in PvP except as catering to bad players.

Taken out of context, but this quote from GC sums up why I think this nerf is happening, and why I'm irate about it.

What we should be doing is comparing 2pT8 to 4pT9, since those are the ones we will have to choose between.

I just wanted to thank everyone for all the information in this thread. Face Mauler has made really made this role viable for me and couldn't play without it now. That mod alone increased my DPS by almost 700 (sustaining 3600~ now) on the dummies and I've been #1 dps on almost all boss fights in my 25 man pugs. Feral is still an offspec for me, but now i feel confident that I can land in the top 3 DPS spots everytime when needed.

Feral DPS is by far the hardest class/role I've played and I now have a lot of respect for any kitties I see pulling out good numbers.

As far as bonuses go, 4T9 isn't close: FeralbyNight (Cat & Bear simulation tool)

It's possible that the difference in stats will still make 4T9 worth it, however.