The State of DPS - A WoL Parsing Site

[Sepirion]

In an effort to elevate concerns regarding DPS from generalized moaning to precise moaning based on empirical evidence, I've put together a little website that will hopefully be of interest to my fellow players.

http://www.stateofdps.com

http://search.stateofdps.com

Basically the site takes the average DPS for each class and spec from the top 120 parses on World of Logs. I know that this is a very small data set and contains some of the most hardcore players in the world; that said, it also is a list of players who are Doing It Right™ and pushing the limits of what is possible for their class.

What I hope to show in this list, both in it's initial form and in future iterations, is not only how DPS stacks up against each other for the purpose of general curiosity, but also (over time) give us an idea of how DPS grows with scaling gear. The main feature I'll be adding after this is a weekly snapshot of DPS numbers, along with the ability to compare and graph DPS changes through time.

I hope this page will be of use to my fellow players here, it will be updated once each morning. The next step I plan to add is a graphing utility to visualize how classes scale through time. Any feedback or suggestions would be greatly appreciated!

[Simca]

Yeah, I saw your site a few days ago and loved it. WoL doesn't present 'by spec' DPS information in a good way at all. It's nice to see all DPS specs like that.

The healing information is less useful, but still - now it is there for people to make bad comparisons!

It would be cool if you could find a way to separate 1H and 2H Frost and 1H and 2H Fury, though - just so people could see those separately. Blizzard supposedly wants to keep both of those viable, so would be interesting.

[Sepirion]

I'd love to be able to provide a bit more data like that; unfortunately WoL is quite difficult to pull data from. The parse already takes about an hour to complete just from looking at the top DPS lists; attempting to gather individual ability use would take considerably more time. That said, it's an idea I'd like to flesh out at some point, if they ever open up their data a bit more.

[Sunseeker]

Really goes to show how low Ret sits on the DPS charts. 2 specs of 3 different classes show up before them. That's sad....

Great data though.

[B@nj0 P3do]

Awesome site, mon!

[skylla05]

Good concept, but please fix your methodology for analysis. Skimming the top 25 is a stupid way to form a sample. Analyzing "potential" means nothing when you have to consider that Blizzard balances around averages, not just the top end. I don't mean averages between multiple fights, I mean averages between high, mid and low end ranges.

I'm sure your intent is to provide a list of DPS, expecting people to use common sense to analyze it. The problem in practice is that's not how most people look at it. I think this is a good start, but I'd be wary of promoting this as a "State of DPS", because it's extremely biased and indicative of nothing but what guilds with the luxury of extreme min/maxxing and catering can pull. This isn't including things like damage variance, because you are sampling off a top end damage range that someone may have seen *once*, but it's logged on WoL, hence your parse will take it into account.

Good, but it's rather inaccurate. I hope people understand this.

Just to clarify, I'm not posting this as an attempt to ridicule your work. I think you have something really great going, I just really, really hope that people look at it correctly because I don't expect people to be doing so with how it is promoted.

[tacosalad]

Cool site. One filter you might consider adding is one to hide pure pvp specs from the listing. While I realize that there is some subjectivity in determining which are and aren't, there certainly are some that historically and currently really only see use in a pvp setting. Arms warriors, frost mages, sub rogues, etc would fall into this category. Part of the reason why it would be a nice feature is just to clean up the view to better see which pure raid specs are successful and which aren't. Additionally though I question how useful the data would be on specs that the community largely chooses not to use in the raid setting. The best 120 players of a high performing dps spec are probably going to be better players that the best 120 of a low performing dps spec of the same class, given that the best players optimize in any way possible. Granted some of these specs can have powerful utility rolls in certain encounters, but the numbers would generally not be favorable to them in such cases anyway.

[SilkforCalde]

Really goes to show how low Ret sits on the DPS charts. 2 specs of 3 different classes show up before them. That's sad....

Great data though.

Paladins are currently BY FAR the best healer and the best tank.

It makes no sense to be ret. I laugh at ret pallies that try to apply to my guild. I laugh at ret pallies period, actually. Love seeing them in BGs. Free HK for everyone!

[Sepirion]

Certainly agree with the comments posted thus far, and also am well aware of the limitations of the data. Unfortunately, WoL only keeps the top 200 parses, so garnering a true average sampling would be very difficult. I am going to bump up my scrape to 200 samples, and possibly chop off the top and bottom 5% to remove some margin of error.

That said, I raid in a top-end guild, and enjoy competing with the top parsers in the world. As much as I want this tool to be used by all, I personally don't care too much about the average-joe dps. I want to know the numbers put up by people who are excelling at their job in raids, and how they compare between the classes. I'm also interested in how DPS changes and scales through time, a feature I'll be implementing soon.

Thanks for the comments though, it's always good to get other perspectives.

[Sunseeker]

Paladins are currently BY FAR the best healer and the best tank.

It makes no sense to be ret. I laugh at ret pallies that try to apply to my guild. I laugh at ret pallies period, actually. Love seeing them in BGs. Free HK for everyone!

I don't really give a crap if they're the best healers or the best tanks(neither of which I agree with IMO, priests and shammies tend to do a much better job IME), the point is their DPS spec is broken and poorly designed. it's not their fault that their output sucks.

Certainly agree with the comments posted thus far, and also am well aware of the limitations of the data. Unfortunately, WoL only keeps the top 200 parses, so garnering a true average sampling would be very difficult. I am going to bump up my scrape to 200 samples, and possibly chop off the top and bottom 5% to remove some margin of error.

That said, I raid in a top-end guild, and enjoy competing with the top parsers in the world. As much as I want this tool to be used by all, I personally don't care too much about the average-joe dps. I want to know the numbers put up by people who are excelling at their job in raids, and how they compare between the classes. I'm also interested in how DPS changes and scales through time, a feature I'll be implementing soon.

Thanks for the comments though, it's always good to get other perspectives.

Even though I am far from great at the game, I prefer to see what the "best" are doing as opposed to what joe-blow or myself is doing. I look at those numbers as what I CAN do if I put my mind to it.

[Sepirion]

Paladins are currently BY FAR the best healer and the best tank.

It makes no sense to be ret. I laugh at ret pallies that try to apply to my guild. I laugh at ret pallies period, actually. Love seeing them in BGs. Free HK for everyone!

For someone that doesn't do much raiding, you have a lot of baseless assumptions! H Pals are great, sure, but they're not blowing healers out of the water or anything. All the tanks seem to be around equal right now.

As for ret, sure it's low, but it's the highest 3% damage buff class in the game right now. Does your guild have logs? It'd be interesting to see how my 'laughable' ret damage stacks up.

[Khevis]

Cool site. One filter you might consider adding is one to hide pure pvp specs from the listing. While I realize that there is some subjectivity in determining which are and aren't, there certainly are some that historically and currently really only see use in a pvp setting. Arms warriors, frost mages, sub rogues, etc would fall into this category. Part of the reason why it would be a nice feature is just to clean up the view to better see which pure raid specs are successful and which aren't. Additionally though I question how useful the data would be on specs that the community largely chooses not to use in the raid setting. The best 120 players of a high performing dps spec are probably going to be better players that the best 120 of a low performing dps spec of the same class, given that the best players optimize in any way possible. Granted some of these specs can have powerful utility rolls in certain encounters, but the numbers would generally not be favorable to them in such cases anyway.

Except that Blizz is making them viable PvE specs, or trying to. Sub rogue is the only underrepresented spec here, and all this data is from PvE sources, so I would imagine that this data is still, in fact, viable and useful information. "Removing" specs that are perfectly acceptable in raiding is stupid - Blizz wants them represented and has said as such. Just because people still want to see them as PVP only doesn't mean the data is useless.

That said, everyone represented on that site pretty much echos a lot of what GC said in his post about "adjustments" coming. Most of the specs dragging behind in DPS (aka a large margin behind the top) and the top two (who look to be getting a little nerfed) are there, and so is the "very low representation" of Sub rogues...

That said, I'm pretty sure Blizz looks at data like this when making judgments on classes. The "average" player will often try to emulate the best anyway. Blizz works on making things work under "optimal" conditions - within reason. Doing a reasonably proper rotation, etc. If you're under performing against everyone else when you could be doing more (in same gear) Blizz probably won't consider an adjustment. If you're playing by the book and under decent conditions and still way behind (WAY behind... 9k is a HUGE gap) then yes, something is wrong.

3~4k behind isn't the end of the world. Someone will always end up better on average, not everyone can be #1 all the time. As long as there isn't large disparity, Blizz is happy. And sometimes this means nerfing some people while buffing others. I'm curious to see the long-term playout as well, especially after class adjustments, etc. come into factor, and then new content.

[Deleted]

nice site!

[sicness]

For someone that doesn't do much raiding, you have a lot of baseless assumptions! H Pals are great, sure, but they're not blowing healers out of the water or anything. All the tanks seem to be around equal right now.

As for ret, sure it's low, but it's the highest 3% damage buff class in the game right now. Does your guild have logs? It'd be interesting to see how my 'laughable' ret damage stacks up.

Really? You're trying to argue that paly healing doesn't blow others out of the water when heroic mode guilds are taking up to 3 holy paladins atm?

He makes a valid point, the class is viable in 2 specs and one spec is somewhat lacking. Not to the point where they can't raid but lacking. Therefor the priority of their dps being increased should be low.

[Simca]

Really? You're trying to argue that paly healing doesn't blow others out of the water when heroic mode guilds are taking up to 3 holy paladins atm?

He makes a valid point, the class is viable in 2 specs and one spec is somewhat lacking. Not to the point where they can't raid but lacking. Therefor the priority of their dps being increased should be low.

That's because top end guilds would stack anything that was slightly better than anything else. Regardless, Holy Priests are performing very similarly to Holy Paladins on 25-mans, if not flat out better on quite a few fights. It's only in 10-mans where Holy Paladins really 'blow things out of the water'.

And Ret IS the most viable source of 3% damage - he's right. The guy he was quoting was "laughing" at Ret applications... sounds like he's missing a raid buff that is pretty dang vital in 25s.

As for Prot, it's definitely viable and is a great tanking spec, but not the 'best'. I'd argue that DKs are the best tanking class right now with the other tanks very slightly behind. Overall, tanking is pretty balanced.

[Sepirion]

Three of a single class isn't really stacking hard, not like shamans in Sunwell or anything. The same could be said of Disc priests for other encounters.

Regardless, just because one aspect of a class is working well doesn't mean the others should suffer. That's very vanillia-esque thinking.

[sicness]

Three of a single class isn't really stacking hard, not like shamans in Sunwell or anything. The same could be said of Disc priests for other encounters.

Regardless, just because one aspect of a class is working well doesn't mean the others should suffer. That's very vanillia-esque thinking.

I'm not saying it should suffer, and it isn't, ret is completely viable at the moment. I'm simply saying that with the other two specs being perfectly fine and one spec being slightly behind other dps I think their priority on fixing ret should be low, as it probably is.

[Simca]

I'm not saying it should suffer, and it isn't, ret is completely viable at the moment. I'm simply saying that with the other two specs being perfectly fine and one spec being slightly behind other dps I think their priority on fixing ret should be low, as it probably is.

Did you see the PTR patch notes? That's a massive buff for Ret - I'd say it isn't a low priority at all. The new DP is a DPS boost, the new mastery is a massive DPS boost (especially if Inquisition works on it, fingers crossed), and now mastery will be a desired secondary stat.

In plus, I'd say Ret is always a pretty high priority for them because last I checked, there are more players playing Retribution than any other spec in the entire game.

[Sepirion]

The fixes are already on PTR that brings it more in line with ~Warriors. It was on this site!

[sicness]

High priority to me is them hotfixing something to fix it. Low priority is waiting until a major patch, which they're doing.