Hm, am I missing something? On the spreadsheet, the Moonkin has 0.7 or something crits per second, but looking at the WWS, it seems that the Survival Hunter had much more crits than the Moonkin, and the Survival Hunter's crits/second is 0.5 on the spreadsheet.

Quick question and maybe I missed it, but does HAT proc for each crit tick of Explosive Shot or just for the initial?

for additional reference, here's the yellow crits per second from our last 25 man Naxx run (mostly compiled from Patchwerk, a couple from Maexxena and Rasvious)


Hunter (BM)
0.55

Hunter (Survival)
0.42

Shaman (Elem)
0.39

Paladin (Ret)
0.34

Hunter (Survival)
0.31

DK (Unholy)
0.30

Hunter (Survival)
0.30

Paladin (Holy)
0.28

Warrior (Prot)
0.27

Rogue (Muti)
0.26

DK (Blood)
0.26

Mage (Fire)
0.26

Rogue (Combat)
0.24

Warrior (Fury)
0.23

Priest (Shadow)
0.21

Druid (Balance)
0.19

Warrior (Fury)
0.17

Warrior (Fury)
0.14

Warlock (Demo)
0.13

All of them are eligible for HAT procs.

Read carefully, the trade-off is not for SS but for Ruthlessness and Blood Spatter, because you actually use Rupture in the rotation.

Thats what i meant. Blood Splatter+Ruthlessness are better then aggression then? Because from what ive heard earlier there seemed to be talk about Evi outdamaging glyphed/talented rupture when you hit about 45% crit.

Not a chance. In a HaT spec, where the bulk of your combo points come in from outside, the main remaining limitation is the energy cost of the finisher. With 5/5 Relentless, the energy cost of a Rupture is zero. It is, quite literally, free damage, at least until you get to the point where you have so many combo points coming in that you're GCD-locked spamming Eviscerate. That was the case for bugged HaT, but isn't the case for solo HaT.

I might just be understanding what you are saying, but you seem to be missing the biggest benefit from hunters/DKs. Their pet attacks are on a different 1s cooldown then any other class, meaning they can basically give 2 combo points a second. I'd assume simply adding 4 more member slots in your inputs and allowing the user to input pets individually, would be the easiest way to do this.

I would recommend trying to average several parses, with many different players/guilds. In my own parses I'm noticing for some classes, many more crits/sec than you have reported. My other suggestion, would try using other classes' modeling/simulations to develop your crit/sec numbers.

Don't get me wrong, I totally understand what you talking about with pets. You can see it from my calculations. I know they are on a separate cooldown timer, that's why the crit/sec numbers for hunters are almost twice higher than any other class. The 0.52 crit/sec for the survival hunter in that parse already includes his pet, so the number is much more modest than what you alluding to by saying "they can basically give 2 combo points a second." I know BM hunters can have higher crit rates, but I think they are of the order of 0.65 in top gear, including pets (just remember parsing a WWS before the hunter respecced to survival, but don't take my word for it). Also do not expect high crit/sec rates from DK's. It's true they have pets, but their own crit rates are low. You can see in the table I have that a DK and his ghoul together did 0.28 crits per second, and that's a very well geared unholy DK (granted blood-specced DK's can do better, but I do not see many around anymore).

I agree with everyone who is suggesting that using crit/sec numbers from several parses would be more reliable. However the point of my work is not to establish what the most applicable crit rates for each class are. We have a large community that can help me establish these numbers. Arguably if we are to reach any concensus, we need to hold the gear level, pet types and specs constant (and hope players' skill and rotations are close enough).

Just like with any other DPS spreadsheets, there are two main uses for this one. One is to figure out what's the best gear, rotation, and spec for the gear and group that you currently have, or can have soon. For that you need to collect the data from your own raids, from the group you are typically running with, then plug in you current gear and calculate your DPS using the spreadsheet to determine the best spec, dps rules and immediate gear upgrades. It's informative to some extent, but not very interesting to everyone else from the standpoint of theorycrafting and comparing with other specs.

The other use of the spreadsheet is to determine "ideal" dps. For that you want to plug in best in slot items, and determine the crit/sec rates of best group members (arguably four BM hunters) who are also in their best in slot items and using best suited pets. Another good approach is to determine "reasonable" dps, which will take into account that it is unlikely to have 4 BM spec hunters in your group, first of all because your guild may not even have 4 hunters in best in slot gear to put in your group, second because BM spec is no longer dominant, so you have to live with the crit rates of survival hunters, and third because you might have to leave a spot for another HaT rogue. "Reasonable" group is up for dispute, but I want to focus my efforts on developing a good tool to measure DPS for when we have proper crit rates to plug in.

Path411 does raise a good point that I can test. What's the difference in dps if I separate the pet into a stand alone entity with his own crit rate and attack speed as opposed to lumping it with the hunter. I will let you know. Personally I think the effect is not going to be profound since in my current implementation I already waive the 1 sec cooldown for hunters because I understand that 0.7 attacks per second that I obtained from the parses includes the pet attacks.

I also have a new version of the spreadsheet that has faster run speed as I am trying to take Vulajin's suggestions into account. I am testing it right now and checking for bugs, and will have it available soon.

i use 6/21/44, and what is weapons is better?
Currently use Omen of Ruin MH and Webbed Dead OH, maybe Webbed Death MH and Hand of Nerub is better?

I think what you're missing here as far as why this build leans so heavily on eviscerate is that you eventually pass a point of inflection where there is no button you can possibly push that will do more damage than eviscerate. Rupture costs the same (0 energy with 5/5 relentless strikes), but will never outpace your eviscerate damage-- even if there is target switching, you still have to push a button to get the damage onto the mob. With this logic in mind, and having the gear you should have after farming raid content for months now, there is no reason to use any finisher except eviscerate (except SnD, when it needs to be refreshed).