We have been in 4.3.2 for two weeks now, meaning that raidbot‘s two-week sample period is full of 4.3.2 data. What are some of the trends in hunter DPS since the patch?
The table below titled “All Parses” presents information drawn from all parses over the last two weeks at raidbots. The information is presented by raid type and contains the following variables:
- SS: This is a spec’s spec score. It represents the spec’s median dps divided by the top spec’s median dps. It thereby represents how close to the best spec a spec is.
- FS: This is a spec’s Frostheim score, a name I’m giving the score since Frostheim is the one who promoted its use. It is like spec score except that instead of the top spec, it is a spec’s median dps as a percentage of the median spec’s median dps. Basically, it measures how far above or below the middle a spec is.
- Sample: This is how many parses that raidbots has received for the spec over the last two weeks. Please take note of how small the sample sizes are for some specs and adjust your reliance on the data accordingly.
All Parses | |||
25H | SS | FS | Sample |
SV | 96.90% | 106.67% | 17,047 |
BM | 90.30% | 99.40% | 709 |
MM | 75.20% | 82.75% | 1,088 |
25N | SS | FS | Sample |
SV | 96.80% | 105.74% | 31,458 |
BM | 97.03% | 97.03% | 1,757 |
MM | 86.62% | 86.62% | 6,864 |
10H | SS | FS | Sample |
SV | 96.40% | 108.74% | 28,758 |
BM | 86.20% | 97.32% | 1,548 |
MM | 75.90% | 85.70% | 2,176 |
10N | SS | FS | Sample |
SV | 96.70% | 108.41% | 82,889 |
BM | 89.40% | 100.25% | 5,047 |
MM | 82% | 91.93% | 20,806 |
This “All Parses” table represents the median hunter by raid and spec. In other words, it represents the middle-most raiding hunter, which is by nature not the best raiding hunter in terms of DPS. What it shows is that for the middle-most raiding hunter, SV is ascendant in more than one way. First, the buffs to SV have sent it over median spec to be within striking distance of the top spec. The buffs have also prompted a sweeping realignment; we hunters are switching over to SV en masse. SV is simple to play, is not hobbled in some of the ways that MM and BM are, and does very high dps. This popular switch over to SV often leaves BM and MM below raidbot’s threshold for statistical reliability in terms of representation. This is one of the main reasons that I’m reporting on figures related to median dps here: MM and BM simply don’t make the cut for percentile calculations.
The “All Parses” table also shows BM doing rather well, often sitting right on top of the median spec. MM, however, appears to be in a state of relative decline, falling below both SV and BM and only hitting the 75% on SS scores on hard modes.
SV does, of coure, makes the cut for 90th percentile calculations and some of these numbers are represented in the table below.
SV on Ultraxion | ||
90th % | Rank | |
25H | 45,008 | 4 |
25N | 41,240 | 3 |
10H | 42,552 | 3 |
10N | 39,218 | 2 |
This table represents the 90th percentile along the distribution of SV parses from Ultraxion by raid type. It uses parses from Ultraxion under the assumption that this fight is the best way to judge raw DPS(e) numbers owing to it being a simple tank and spank fight with no DPS modifiers. The table thereby represents high-performing SV hunters and suggests that what holds true for the median hunter also holds true for the high-performing hunter: SV is doing very well these days.
I think it’s unfortunate that they were only able to hit their stated goal of “don’t make everyone need to switch specs” in 1 out of 3 raid tiers this expansion.
– MM (along with BM) was buffed and SV was nerfed heavily during tier 11
– MM remained on top for tier 12, although not by a huge amount
– SV was buffed back on top in tier 13
“We heard from many players how frustrating that was — to wake up one day and have to learn to replay their class because the anointed highest dps spec was now a different tree completely.” — Ghostcrawler
(source: the old, decommissioned WoW forums. Here’s a link to a news article with that quote in it: http://wow.joystiq.com/2010/11/02/ghostcrawler-shares-dps-spec-design-philosophy-for-cataclysm/2)
I think they should have been less conservative in using the new mastery stat and spec bonuses to keep specs more balanced during Cataclysm.
Blizzard has a tendency to over-nerf and over-buff.
My brother has speculated – seriously, not humorously – that Blizzard deliberately gives classes and specs their “turn” at the top of the meters. The primary reason I give some consideration to his speculation is that these are NUMBERS. Blizz can procedurally mass-test many kinds of things, and have to know most of what’s going to happen (nearly always) when they buff X or nerf Y. It feels to me like they are not throwing enough ideas at the wall to see what works.
For example, what possible ramifications could it have to buff MM mastery? Buff the proc chance or buff the shot damage, and do so deliberately and carefully. That will easily reward good play, because more shots = more Wild Quiver procs. Or buff Piercing Shots, so that the dot is a higher percent of damage, so much so that keeping it rolling is as important as keeping Rip up is for kitty dps. Buff the duration of ISS so that we can take more advantage of the 2pc and 4pc T13 bonuses.
For BM, it might be a mistake to buff Master of Beasts for PVP reasons, so what else could we buff? Buff Ferocious Inspiration so that it says 3% for all party/raid and 6% for hunter and pet. Buff Killing streak so it does 50% more damage, or so that the next TWO Kill Commands have the 20% buff. Buff Invigoration to the point where Cobra Shot only needs to be cast (during a boss fight) to keep Serpent Sting up. Buff Serpent Sting with some other deep BM talent.
I tend to believe that Blizzard’s live team is understaffed, and the above discussion is fairly indicative of why I feel that way. There are a dozen nobs to turn for each spec and class. Some – perhaps many – will have ramifications that echo all over the game, but if Blizzard really has all the tools to test these things that Ghostcrawler claims, then it should be fairly easy to run through a few dozen iterations of Frost Mage/BM Hunter/etc specs to figure out some likely candidates that could have gone up on the PTR with 4.3.
that was, of course, the main reason for having mastery. To be able to tweak the percentages to balance things out.
Oddly, that’s exactly what blizzard haven’t done when it comes to balancing specs and classes out.. they’ve deliberately steer clear of it and instead tweaked the numbers on core abilities. I can’t help but think why – as long as the they dont over do it to affect the stat priority it should be easy. e.g. bringing mastery down on a spec where it’s already lowest stat prio isn’t going to affect the prio.
For me the evidence of Cataclysm is that Blizzard do not understand their own game. The only way they know how classes will compare is make the changes in live, see what happens, then react.
This is understandable given how complex the game is I suppose but what I find disappointing is that:
– They fail to fix simple things that would make the situation a lot easier to manage. e.g. why did the fixes for BA and L&L have to wait until the end of the expansion? They should have been sorted in at the start or 4.0.2 latest.
– They don’t engage with their customer base, some of whom undoubtedly know more about their specialist areas than they do. For example, why not actively involve people like Zeharah or Kruf?
– Their design process seems a bit haphazard. Mastery is clear example of this, it was a brand new stat for Cat and should have been ideal for balancing. But, as mentioned above, its all over the place, for some specs its top, for others its next to useless. All in all its just made things worse.
Using mastery as a balancing stat doesn’t ensure that mastery has the same importance to all specs and classes; in fact, I suspect we’d see mastery working out just like it is now anyway, and this is why:
If Spec X is a little weak, they buff the impact of mastery for Spec X. If Spec Y is hitting too strong, then they nerf the impact of mastery for Spec Y. Eventually, the value for mastery will be wildly different depending on who lied where on the theoretical balance curve sans mastery. Those who were higher would end up having mastery be a poor stat for them, and those who weren’t would value it highly. You’d still have wildly different valuations for mastery per Spec/Class.
Yes. If it had been implemented in this way it would have been fine. Then you would typically have seen, for a class mastery being good for some DPS specs and bad for others.
You would also have seen some sort of pattern during the expansion. Sites like this and others would have noted that mastery had changed as a result of a patch and recommended reforging into/out of it as a result.
For hunters this pretty much hasn’t happened at all.
You do realize that hunters have been at (or very near) the top of DPS charts for the entire expansion, other than being “only” a smidge above median at the beginning of 4.3? You cried about DPS then; now that SV is in the top 3-5 according to raidbots, you’re complaining about Mastery?
I don’t recall hunters being at the top of DPS charts since SV was nerfed at the start of Cataclysm. And as for 4.3 well hunters were in a terrible place at the most important stage of the whole expansion. So terrible that hunters the class of Kruf were excluded from groups. Listen to his podcast and you have a clear sense of how he feels. Being buffed now is too little too late.
My comments are in the context of the others that Blizzard have not delivered on their stated objectives and are not to be able to manage class balance in a proactive way. Rather, as I have said, they seem to make changes in the live environment then react.
My complaint about mastery is that it was new to Cataclysm and if intelligently implemented would have helped them address this issue. This didn’t happen, rather it has just become another complicating factor.
BM single target is being underestimated by a lot. SV is probably stronger yes, but by much less than stats suggest. The pet reliabilty and related bugs, and the lack of good AoE (switching pets in mid fight is bad for your dps) are the real problems there. I don’t understand why they would let mastery be that bad during a whole expansion.
If BM is suffering in-game from effects not represented in models, then it isn’t being underestimated, it is being overestimated.