Okay so I need some help here. I am absolutely unconvinced after my own testing that mastery is always greater than stacking hit to high levels. Below I will be explaining what it is that I am seeing and why this is inconsistent with what all current theorycrafting has come up with. Understand that my point with this post is to try and understand the discrepancy I am seeing, and not to just try and argue that my way is better than some other way. Honestly, after doing alot of digging, it makes sense on a whole lot of levels that mastery is the superior stat over stacking hit for fury warriors. Im just not able to confirm it and I would like to know why.
I performed 18 tests on a boss mob level target dummy. 9 tests were done with an emphasis on stacking hit to as high as I could with my gear. 9 tests were done with emphasis on stacking mastery as high as I could with my gear. The gearsets DID NOT change, only gems and reforging (mostly reforging).
Now, I completely understand that a target dummy is not an accurate or even a good representation of actual warrior dps. The lack of raid buffs on me and lack of raid buffs on the boss are significant to the actual damage we do. My testing was merely an attempt to understand the theory that with the changes in 4.0.6, mastery is a better stat to stack on your gear than hit is beyond 8%. The argument is that with a greater amount of hit you are able to heroic strike more and your white swings will do more damage by missing less. With mastery, you maintain your rotation and continue to heroic strike when you have rage but your Raging Blow will hit harder all the time and your Deathwish will be up more often. The reason why I feel my testing is valid is because during a raid environment there is absolutely nothing that will increase your mastery level and cause you to do more damage from Raging Blow that wouldnt already otherwise be applied to a gearset that stacks hit. Conversely with these tests, there are several elements in a raid that can improve your hit further then current levels making testing it on a target dummy a less accurate representation than a raid environment. By this thinking, stacking hit should show itself to be lower by a significant margin than stacking mastery. This is not what I have seen from my testing.
Below are my test result summaries. Please keep in mind that they are summaries of what I wrote down (on paper) and not combat log parses or specific numbers from each test. They are the averages that I concluded from the information I took down because I wasn't trying to go too far overboard:
armory: http://us.battle.net/wow/en/characte...zzinoth/simple
spec: http://wowtal.com/#k=BeB4M2iT.aei.warrior.RlfZEK
(note the spec is 2/3 incite, this is on purpose to compare gearing and not specs)
Stacking Hit
Stats with Hit (stats inlcude Battle Shout)
7.83% Haste
19.19% Hit
19.87% crit
8.96 Mastery Rating
10493 Attack power
Test Summary with Hit
Each test ran for 7 minutes exactly from behind the target dummy. All cooldowns including Deathwish, Recklessness, and Colossus Smash were popped at the start of the fight and the third Deathwish was popped when Recklessness came off of cooldown to get maximum effect out of the ability.
Priority rotation for hit stacking was as follows:
Colossus Smash > Bloodthirst > Raging Blow > Slam Proc > Heroic Strike
Understand that I used Heroic Strike very liberally without ever breaking this priority list. If I was able to Heroic strike and Bloodthirst and still have enough rage for Raging Blow, then I did it. A test was stopped if at any point in time I broke my priority list by rage starving myself and its results discarded for another attempt.
Also know that with the new mechanics of Inner Rage I was able to pop this and occasionally successfully pull out an extra heroic strike when rage was permitting. This was saved for every other Colossus Smash or during Deathwish/Recklessness burn periods.
Average Statistics across all 9 tests:
Overall DPS: 15.2k
Damage Done to target from abilities from largest to smallest: BT (28.6%) > Melee (23.5%) > Raging Blow Main Hand (12.6%) > Heroic Strike (11.1%) > Raging Blow Off Hand (8.0%) > Slam (6.8%) > Deep Wounds (6.6%) > Colossus Smash (2.9%)
Number of Heroic Strikes: 68
Stacking Mastery
Stats with Mastery (stats inlcude Battle Shout)
7.83% Haste
11.19% Hit
19.60% crit
14.59 Mastery Rating
10493 Attack power
Test Summary with Mastery
Each test ran for 7 minutes exactly from behind the target dummy. All cooldowns including Deathwish, Recklessness, and Colossus Smash were popped at the start of the fight and the third Deathwish was popped when Recklessness came off of cooldown to get maximum effect out of the ability.
Priority rotation for mastery stacking was as follows:
Colossus Smash > Bloodthirst > Raging Blow > Slam Proc > Heroic Strike
For mastery, even though raging blow is a higher damage ability (and indeed the ability I am stacking to improve) it is still most efficient to use Bloodthirst first for the Battle Trance proc as well as the increased chance on Slam. Also note that if a slam was already up and BT came off cooldown, I stuck to the priority list and used BT instead of the slam. This is because BT does SIGNIFICANTLY more damage than slam and can proc the deadly calm. I did however attempt to use every slam I could.
For mastery I also did not always save Battle Trance for a heroic strike and instead incorporated it into my rotation if I was low on rage to ensure that I would be able to complete my rotation at all times.
Understand that I used Heroic Strike when I could without ever breaking this priority list. If I was able to Heroic strike and Bloodthirst and still have enough rage for Raging Blow, then I did it. A test was stopped if at any point in time I broke my priority list by rage starving myself and its results discarded for another attempt.
Also know that with the new mechanics of Inner Rage I was able to pop this and occasionally successfully pull out an extra heroic strike when rage was permitting. This was saved for every other Colossus Smash or during Deathwish/Recklessness burn periods.
Average Statistics across all 9 tests:
Overall DPS: 14.4k
Damage Done to target from abilities from largest to smallest: BT > Melee > Raging Blow Main Hand > Raging Blow Offand > Heroic Strike > Slam > Deep Wounds > Colossus Smash
I do not have average % because I didnt think to write it down until I did my hit tests :-/ however there was significant difference in overall Raging Blow damage between hit and mastery. I dont have the %'s to say definitively but Raging blow did roughly 9% more overall damage with the mastery gear than the hit gear.
Number of Heroic Strikes: 54
Conclusion:
With these tests under the guidelines I mentioned above there was almost 1k dps increase on average between stacking hit to my 19% level and stacking mastery. This is in opposition of the theorycraft model that is being accepted with 4.0.6. I understand that there is a possibility that the mastery superiority has more to do with its reliability in raids, its consistency across all types of mobs and fights, and that perhaps the mastery idea is based off certain gear minimums. I have been unable to find any minimums with reference to these two stats when compared side by side. The general assumption being that regardless of what you are doing, stacking mastery will improve your damage significantly over that of stacking hit to various levels. There is no mention anywhere that I have seen that says your mastery should be at X level before this becomes viable.
Note that with these results, 1k dps does not come from the increased number of heroic strikes and more than likely comes from the increased amount of white swing damage. However I foolishly didn't collect my damage %'s and cannot confirm that at this time. I would like to believe that the top DPS parses in the world from WoL are reliable enough to know that Mastery > Hit with 4.0.6, but I am not seeing that for my gear level under this controlled environment. I also know (as should any fury warrior who has been playing for a while) that being unlucky in a fight can be the difference of about 5k+ dps from one attempt on a boss to another. This means that even if Landsoul has the world top DPS parse on some fight with mastery, he may have just had the stars align absolutely perfectly that one occasion. It is very hard to use that information definitively.
Even still, I am not naive enough to think that going mastery isnt a good thing. It doesnt require nearly the skill or luck that stacking hit has with heroic strike, but I would like to be able to confirm if there is a point where the two specs cross and one becomes superior to another. Or if there is a situation where the truly max dps on a specific encounter can only be achieved by one spec over another because of some reason.
I am hoping with this post that people will sit down and actually try out both ways of doing things. Trying to eliminate as many variables as possible and see if they come up with something consistent with my findings. Try raiding one week with mastery as your stat priority and another with hit stacked as high as you can. Help me by looking at the information I have included here and point out anything I could have done wrong. Or anything I could tweak to get a better more consistent test result. Ultimately I want to know FOR ABSOLUTELY SURE that hit stacking is an inferior way to play as a fury warrior with 4.0.6 regardless of gear level. If it isn't, then where does the gearset change and why?
Thanks alot for any and all help you can contribute to this!
Magnuss
