Originally Posted by Rob
Pitbuller's estimates seem fairly accurate. These experimentally-derived numbers are subject to a certain amount of error, and on the MH we experimentally found that AP is less important than Pitbuller's equations say, while on the OH we found that AP is more important than Pitbuller's equations say. Therefore I'm tempted to chalk this up to random error and just go with Pitbuller's estimates.
|
I thought about this some more while I was doing something else, and it occurred to me that the reason MH DPS seems to be relatively more important than expected, and OH DPS seems to be relatively less important than expected, is that the MH will receive more WF procs due to the fact it's rolled first during a Stormstrike. So, I'm not sure which estimate is truly better: Pitbuller's {1 MH DPS = 8.48 AEP, 1 OH DPS = 4.24 AEP} or mine {1 MH DPS = 9.03 AEP, 1 OH DPS = 3.70 AEP}. Intriguingly, 9.03 - 8.48 AEP = 0.55 AEP = 4.24 - 3.70 AEP. This coincidence is too delicious for me; I'm gonna have to go with my estimates and assume that they are correctly accounting for the MH having an increased number of WF procs.
So, to summarize:
Pitbuller is theoretically correct, but does not account for the fact that Stormstrike attacks first with the main hand, giving the main hand a higher number of Windfury Weapon procs. The result of this is to increase the AEP of main hand weapon DPS and decrease the AEP of off-hand weapon DPS. The weights of MH and OH DPS are thus:
1 MH DPS = 9.03 AEP
1 OH DPS = 3.70 AEP
If someone wants to repeat my experiment with a different baseline set of stats (I used 1600 AP, 30% crit, 17% hit, 0% haste, 0 armor penetration, double Mongoose, no trinket procs, no Dragonstrike proc), they could confirm my results by reproducing them or show that my results are flawed.
An obvious shortcoming in these AEP values is that they assume a 2.6 speed weapon. Slower weapons will be worth more AEP for a given DPS. Feel free to calculate that if you try to confirm my results
