State of Rogues - Page 11

shegil · 12/18/08, 6:07 AM

Originally Posted by songster

Tricks should not affect your damage at all, since it has zero energy cost.

Tricks of the Trade costs 15 energy, so you will lose personal dps by using it =). The raid will however gain dps, especially if you time it with BL on a mage when the target is below 35% health.

rozetta · 12/18/08, 9:12 AM

To preface this, here's my character:

The World of Warcraft Armory

I rerolled from Paladin to Rogue a couple of weeks prior to WotlK's arrival.

I believe this thread is warranted. As GC has stated, they're keeping a close eye on rogue damage, however he's been careful to also state that _if_ it's been found to be low, it will be buffed. The fact that he hasn't posted any more on this issue either says that they're still evaluating the situation or lends credit to the idea that, in fact, rogues are where the "baseline" should be. He further mentioned that they have some fantastic rogue math guys/theorycrafters at their disposal, and this says to me that they have the tools they need to analyze the situation. However, we do know that they likely pay attention to these forums, and a several page thread with meaningful discussion about the state of this class has got to grab their attention, especially on a forum know for it's complete lack of "QQ". The mere fact we're discussing this suggest there is something to look at.

As far as my personal experience goes, I have a few things I can say:
- we have 3 rogues who are all regular raiders
- the other two are slightly better geared than myself, having gotten to 80 a couple of weeks before I did. We've all had about an equal amount of participation in 25-man content and our guild has cleared all but 2 and 3 add sartharion (our server is very laggy and we have never had the proper opportunity to try it)
- I often see the other rogues in the top 5 (all of us are muti spec)
- one of the rogues was initially combat, but switched when we started to overtake him in damage in 25-man raids
- myself and one other rogue tried HaT last night in Naxx. However, it didn't perform the way we'd seen it advertised. It could be that we're not at the optimal gear level for it or it could be due to the often 4-second response times we were getting on some fights last night. Either way, we've pretty much written the spec off as gimmick and will not run it again.
- we are simply useless on 90% of trash

I do believe that a slight buff is in order and that we should put out slightly more damage than other classes on longer single-target fights. This is something I'm still not seeing, but as you can see, our guild is still not in best-of-best. I'm hoping things level out as the situation approaches that. I also have a feeling that we'll start seeing better scaling come Ulduar, and I certainly hope that the trash is more rogue-friendly and not so much of an aoe-fest.

I personally would like to see:
- some change to HfB to not be so manual and not so punishing if it drops
- a slight buff to our single-target DPS capabilities (one thing that comes to mind is to change the poison hit cap to be the same as the yellow hit cap)
- some sort of meaningful raid buff, if possible, to prevent aforementioned scenarios where taking a rogue forces the omission of a raid buff in 10-man content (something more along the lines of bloodlust sounded nice and also flavorful)

I'm glad that this thread has kept to meaningful discussion and anecdotal evidence. I'm very interested in hearing more of what people have to say.

songster · 12/18/08, 10:21 AM

Question: how is the 1.5x multiplier for FoK with daggers applied? Presumably it checks each weapon individually, but it would be nice if someone with PTR access could test this. This is mainly interesting to Fist/dagger combat specs, of course. On the other hand, if it only looks at the main hand, there would be some silly fun to be had swapping to dagger/fist :-)

Ghostwolf · 12/18/08, 10:33 AM

I am currently specced mut and am not hit capped, here are just a few 25 man naxx runs, not full clears but from a few nights.

With aoe we may be able to move up there in the dps charts as currently almost every pull in naxx is aoe'd down. Every other class has a viable aoe, with only whirlwind being a debatable one.

Wow Web Stats

Wow Web Stats

Wow Web Stats (The other rogue is combat specced and I spammed Tricks of the Trade on him for increased dps on the Loatheb fight.)

Klyse · 12/18/08, 10:56 AM

This is just some fanciful dreaming, but do you think it would be overpowered to award Fan of Knives a combo point? I could see it being a combo point builder in those huge AoE fights. And more unlikely, I think it would be neat to have an AoE finisher of some kind, either damage or utility (AoE stun?).

Tinwhisker · 12/18/08, 11:14 AM

Originally Posted by Klyse

This is just some fanciful dreaming, but do you think it would be overpowered to award Fan of Knives a combo point? I could see it being a combo point builder in those huge AoE fights. And more unlikely, I think it would be neat to have an AoE finisher of some kind, either damage or utility (AoE stun?).

In a world where CPs are attached to your target (a single target) this doesn't make much sense.

Couple this with the fact that the reason we are getting the cooldown removed is because we currently don't contribute a lot to AoE pulls. As much as I like the idea of having some meaningful contribution to an AoE pull I do not want to become a fully capable AoE class.

Leto · 12/18/08, 11:15 AM

Originally Posted by Klyse

This is just some fanciful dreaming, but do you think it would be overpowered to award Fan of Knives a combo point? I could see it being a combo point builder in those huge AoE fights. And more unlikely, I think it would be neat to have an AoE finisher of some kind, either damage or utility (AoE stun?).

Blizzard has stated they do not want FoK to be used in a rotation, so I highly doubt they would have it generate a combo point.

Klyse · 12/18/08, 11:26 AM

Originally Posted by Tinwhisker

In a world where CPs are attached to your target (a single target) this doesn't make much sense.

Couple this with the fact that the reason we are getting the cooldown removed is because we currently don't contribute a lot to AoE pulls. As much as I like the idea of having some meaningful contribution to an AoE pull I do not want to become a fully capable AoE class.

You're absolutely correct, I didn't think about the combo points being attached to a specific target. I suppose our role in those AoE situations is to roam around and pick off the trouble makers in the AoE crowd, like the casters, etc.

So yeah, scratch the combo point idea

I'll settle with some opening AoE.

ekval · 12/18/08, 11:30 AM

FoK is pretty much fine now and even greater when cooldown is removed. FoK with Throwing Specialization/poisons is good utility in certain AoE packs and it does pretty significant chunk of our total damage in situations where it's used. And everyone should know how good utility ToTT + FoK is (even more after cooldown removal change).

Don't think anyone wants us to become AoE class, therefore FoK is pretty balanced utility/damage skill from my perspective. FA or AR combined with FoK spam without cooldown is going to be enough, and making FoK to generate combo points is out of the window in my opinion, even as idea.

Mindflayer · 12/18/08, 11:47 AM

I don't think individual WWS parses are going to prove to disprove the perceived state of rogues. An individual parse doesn't take into account all the myriad of factors that affect DPS over the course of a particular fight.

If we want something conclusive, the best way is to go thru *ALL* of the WWS parses for WotLK raids, and start compiling/graphing the data until the bell curve for each class starts to show. (OLAP/data mining anyone?). Then you will have a comparison that could be considered valid.

So has anybody tried pulling down all that data from WWS? Can you even pull it down? Let's figure out how we can *prove* our assumptions, rather than assujming there's a problem and speculating on solutions.

Garren · 12/18/08, 1:01 PM

Originally Posted by Mindflayer

I don't think individual WWS parses are going to prove to disprove the perceived state of rogues. An individual parse doesn't take into account all the myriad of factors that affect DPS over the course of a particular fight.

If we want something conclusive, the best way is to go thru *ALL* of the WWS parses for WotLK raids, and start compiling/graphing the data until the bell curve for each class starts to show. (OLAP/data mining anyone?). Then you will have a comparison that could be considered valid.

So has anybody tried pulling down all that data from WWS? Can you even pull it down? Let's figure out how we can *prove* our assumptions, rather than assujming there's a problem and speculating on solutions.

The problem is that every parse is not considered equal, especially this soon after launch. Every rogue has different gear, every rogue plays at a different skill level, every rogue plays with a different raid composition, and every rogue plays with members at a different skill level. And to top it all off, it sounds like many of the top end servers experience very serious lag in Naxx right now. The situation is too muddy right now. Some people post parses where they top the meters, some people don't. There is kind of a hint that there might be a problem at the top end, but it can be difficult to separate the signal from the noise

I will say that in my experience, my DPS more or less places the same as it did in BC. Maybe it has fallen a couple of places, but I'm still near the top in pretty much every fight. And frankly, I'm willing to admit that maybe I'm not the best player in my guild.

Mindflayer · 12/18/08, 2:04 PM

Given all those factors you just mentioned, if the sum of those reports means nothing, then individual reports mean even less. That's the way statistics work.

Take your own example. "my DPS more or less places the same as it did in BC". Is everyone at exactly the same skill level that they were in BC? Nobody learned anything in the xpac? Is everybody in exactly the same gear? Probably not. Are Raid comps exactly the same as they were in BC? Maybe. Nobody got a new computer? Nobody has different rotations/latency/distractions/whatever? Given that everything has changed, its not really a valid comparison to say where you rate on the DPS meter, much less rate rogues as a whole.

Only when there is enough data do the individual factors start to even out. With a good sample size, it is a fair assumption that for every geared rogue, there is a geared hunter, mage, warrior, pally, etc. With a good sample size, you can assume for every high/low-latency class x, there is a high/low latency class y. And so on. It all averages out.

So by my own statements, we probably don't need to go thru all the WWS reports, but enough to get that 95% confidence. Anybody remember how to calculate standard deviation/error on the WWS report total so we can see how many we need to pull down and analyze?

Monistatus · 12/18/08, 2:47 PM

Originally Posted by Garren

The problem is that every parse is not considered equal, especially this soon after launch. Every rogue has different gear, every rogue plays at a different skill level, every rogue plays with a different raid composition, and every rogue plays with members at a different skill level.

And that's *exactly* why we should be bell curving all this data - we can't just look at the performance of the creme de la creme of rogues. We have to look at every raiding rogue, regardless of skill level or gear level.

I don't doubt Blizzard does this... as I've stated previously in this thread, this expansion (so far) was tailored for the more "casual" player; players of exceptional skill level are only a small piece of the pie now.

MasterDinadan · 12/18/08, 2:48 PM

He's pretty much right... The best conclusion we can come up with is one that is based on the most data that we have available. While taking a large sample size means we are likely to end up with a lot of rogues who have terrible gear or simply don't know how to play their class, the same would be true for all classes that we get data on.

We could at least determine whether or not there IS a difference in average DPS among classes. Based on that conclusion alone, we could not say that it is due to one class being better than another, but it at least shows some support for that hypothesis. Further analysis would be needed to done on the gear level of all samples taken (if all the warlocks you looked at had better gear on average than the rogues, this could explain a damage difference) as well as differences in skill level. Unfortunately, the latter is difficult to quantify. However, given a sufficient sample size, we feel that average gear and skill levels should not vary much between classes.

Liminality · 12/18/08, 2:58 PM

"There are lies, damn lies, and statistics"

You can analyze every WWS report, but its an exercise in futility. You have to make several huge assumptions (like gear evening out) is one example. You can't make leaps of faith like that if you want evidence with credibility.

You want to limit the number of factors that come into play. I would think you would be better off looking at good guilds WWS reports of players who are around equal gear level and skill and anaylizing that data. With the same skill + gear + server, you knock out some assumptions (not all).

Overall, I think the rogue community can objectively look at their own DPS and recognize that its not where it should be. I think the FoK buff will definately help for overall damage meters, but something small is in order to help with single target DPS.

I also completely agree with people that think that the rogue rotation for muti needs to be less clunky.