I've put together a little excel sheet to experiment with the dmg from BA and ET.
Assuming the RAP coefficients Nerec postet are correct and inserting 17000AP and 0.42crit I get the following values (disregarding mastery and debuffs altogether because those affect both in the same way):
Black Arrow: 20098.68
Explosive Trap: 13224.08
The values differ because Nerec didn't subtract the crit chance from the normal damage (100% normal hits + 42% * 2 or 1.5 crit coeff results in too much damage). I'm at work right now so I can't upload the xls file.

Still, pichuca was of course right, BA does outdps ET on a single target and with the changes coming in 4.3 they both have 10 ticks with the same chance to trigger LnL.

Also I used the calculated numbers for another test. Assuming 1 target takes the full effect of the ET I created a second table calculating how much damage the ET does if x additional targets take the initial hit of the trap + y ticks of it. Not very relevant in the current tier, as said earlier mostly in the Yor'sahj encounter but some others like Blackhorn, Spine or Madness may be affected too.
According to my table ET will only do less damage than BA with 1 additional target and 4 or less ticks or with 2 additional targets and 1 tick. In every other case ET should outdps BA.
With the values stated above on a Yor'sahj yellow/black/red scenario (I'm not sure if ET is aoe capped for 10+ targets) ET will do up to 145k damage if all 10 adds take the full trap.

Its pretty simple math here.
Multishot does 120% weapondamage to any target it hits + damage from ISrS(via Serpent Spread). The alternatives, Arcane shot or Cobrashot (SS isn't applicable, as no SV hunter would be using it anyway) do respectively 100% weapondamage + 4.83%RAP + 289 (AS) and 100% weapondamage +1.7%RAP + 276 (CoS) - both AS and CoS is increased by SV mastery, as is ISrS.
An ilvl 397 ranged has a dps of 888.2, an ilvl 397 geared hunter will have ~28k RAP (from browsing a few FD profiles) the math then becomes;
Multishot cleave:
(1.2*(888.2+28,000/14)*1.8)+(0.3(28,000*0.4+(460*15/3))) = 6,239+4,050 = 10,289 damage per target per second at the cost of 40 focus per second. The 4,050 damage is multiplied by SV mastery.
Arcane shot "spam":
(888.2+28,000/14)*1.8+28,000*0.0483+289 = 5,199+1,352+289 = 6,840 damage per second at the cost of 22 focus per second - unglyphed, glyphed it increases to 7,661 damage per second at the cost of 22 focus per second. All of this damage is multiplied by SV mastery.
Cobra Shot spam:
(888.2+28,000/14)*1.8+28,000*0.017+276 = 5,199+476+276 = 5,951 damage per casttime at the gain of 9(18) focus per casttime. This damage is multiplied by SV mastery.
TO make a complete realitsic comparison, I need to factor in the SV mastery, browsing World of Logs and cheking the SV hunters, I will state that ~15 mastery (including the gratis 8 points) is a reasonable amount to expect in ilvl 397, leaving the following damageamounts:
MS: 10,897 (per target)
AS: 8,810
CoS: 6,844

Now factoring in the focus cost, this means gives us the figures of:
MS: 272.4 DPF per target
AS: 400.5 DPF

While these figues make it intuitive to go MS for 2+ targets over AS, we have to remember that you burn through your focus at twice the rate, meaning more CoS, faster, to keep your ExS going - while I haven't done any ExS comparisons above (because it has a CD so it isn't spamable), ExS, with the stated amounts of RAP and mastery, will deal ~28k damage with all 3 ticks for 44 focus, and is more DPET than any of the shots and more DPF than MS on 2 targets. This means that wuith 2 targets you want to use ExS whenever possible, and if there are no damage mitigations tresulting in a situation where you have to focus one target down before engaging the next, MS is more DPS even a single target when directly comparing MS and AS, but it is far from sustainable.
A rough estimate of sustained fire says you will need 1 CoS per AS to sustain (with 2t13) and 2 CoS per MS to sustain making the "real damage" as follows (on 2 targets):
MS: 35,482 damage over 1sec+2xCoS
AS: 15,654 damage over 1sec+1xCoS
With a CoS casttime of 1.6, this comes out to:
MS: 8,448 dps (over 4.2 seconds)
AS: 6,021 dps (over 2.6 seconds)
Keep in mind that ExS is still wanted on CD as long as there are only 2 targets (which is what we are looking at here), so while you can do 2 complete "AS spams" between ExS you can only do 1 "MS spam" and 1 extra ability.
Also keep in mind that this assumes that there is no kill priority, and doe snot take into account debuffs on the targets which will vary depending on which target the other members of your raid is on and what debuffs they apply and how.

A note about crit; as crit affects each of the shots looked at the same, I have ignored it to simplify the math as much as possible so you can easily use it for your own specific gear if you would rather math it out manually.

I was also curious what exactly SV hunters should do when there is only 2 targets
I know it's kind of a crude way to test it but this is what I did:

-Used my default SV profile and removed Rapid Fire from shot priority and disabled Bloodlust.
-Set fight length to 30 seconds and 1 minute for each scenario (to more accurately simulate when this will useful).

Scenario 1
-Used my default shot priority and recorded new DPS at shorter fight length (to simulate ignoring secondary target).
=> 55418.28 DPS @ 30 second length / 48494.00 DPS @ 1 minute length

Scenario 2
-Recorded normal DPS, then doubled Serpent Sting damage contribution (to simulate a scenario of just using normal rotation and just refreshing Serpent Sting on an secondary target).
=> 57858.05 DPS @ 30 second length / 50861.74 DPS @ 1 minute length

Scenario 3
-Substituted Black Arrow for Explosive Trap in shot priority.
-Recorded new DPS then doubled Serpent Sting and Explosive Trap damage contribution (to simulate both mobs staying in Explosive Trap the whole time while primarily single targeting a mob and just refreshing Serpent Sting on a second).
=> 58634.58 DPS @ 30 second length / 51572.33 DPS @ 1 minute length

Scenario 4
-Substituted Black Arrow for Explosive Trap , Arcane Shot for Mulit-Shot, and removed initial Serpent Sting in comparison to default profile.
-Recorded new DPS after factoring in doubling DPS gain from Explosive Trap, Serpent Sting, and Improved Steady Shot.
=> 55829.06 DPS @ 30 second length / 51011.96 DPS @ 1 minute length


- From this, I've concluded that it's clearly a DPS increase to use the standard SV shot priority while also maintaining rolling Serpent Sting on a secondary target and refreshing it with Cobra Shots for a shorter duration, but as the combat length extends to 1 minute, the advantage greatly diminishes to barely beating out the Multi-Shot focus dump. While it is a slight DPS increase to use Explosive Trap instead of Black Arrow, it should only be used if both mobs will definitely be within the area of effect for the entire duration.

It definitely isn't exact math, as the switch to from Black Arrow to Explosive Trap would increase the probably of Lock and Load procs, and I didn't accommodate for the shot replacement of a Cobra Shot and replacement of one Serpent Sting with Improved Serpent Sting (or alternatively a Multi-Shot) for the cases that started with rolling Serpent Sting on the secondary target, nor the focus impact of the initial rotation from doing such, as well as I'm sure various other impacts which couldn't be simulated at FemaleDwarf.

This made me curious about optimal play with 3 mobs where I assumed Scenario 4 would be best, but I figured I'd repeat the procedure just to see the approximate gains in each case.


Scenario 1B
-Again, used my default shot priority and recorded new DPS at shorter fight length (to simulate ignoring second and third targets).
=> 55418.28 DPS @ 30 second length / 48494.00 DPS @ 1 minute length

Scenario 2B
-Recorded normal DPS, then tripled Serpent Sting damage contribution (to simulate a scenario of just using normal rotation and just refreshing Serpent Sting on an second and third target).
=> 60297.82 DPS @ 30 second length / 53229.48 DPS @ 1 minute length

Scenario 3B
-Substituted Black Arrow for Explosive Trap in shot priority.
-Recorded new DPS then tripled Serpent Sting and Explosive Trap damage contribution (to simulate all mobs staying in Explosive Trap the whole time while primarily single targeting a mob and just refreshing Serpent Sting on a second and third).
=> 63546.20 DPS @ 30 second length / 55746.59 DPS @ 1 minute length

Scenario 4B
-Substituted Black Arrow for Explosive Trap , Arcane Shot for Mulit-Shot, and removed initial Serpent Sting in comparison to default profile.
-Recorded new DPS after factoring in tripling DPS gain from Explosive Trap, Serpent Sting, and Improved Steady Shot.
=> 62780.14 DPS @ 30 second length / 57790.99 DPS @ 1 minute length


- From this, I have to reiterate the crude math here, as Scenario 3B at 30 seconds comes out the highest, but also suffers the most from artificially adjusting shots without considering the casting of the second and third Serpent Stings (or starting with a Multi-Shot) and the awkwardness, and in some cases, impossibility during Lock and Load procs and during Kill Shot range without overflowing focus, of actually keeping three Serpent Stings rolling on separate targets.

With that said, it is clearly most practical as well as the highest DPS to use Multi-Shot and Explosive Trap on three or more targets, which is as expected. This holds true at both the shorter 30 second and longer 1 minute combat lengths.

Whats the reasoning for promoting the 757 haste plateau? With the 2cp T13, haste effects, LnL and other things, I dont see how gearing for this specific plateau will give any benefit since the times we cast 3x CS is quite rare.

At the time t11 was highest tier, mastery does scale slightly better as haste after the plateau. After several updates (BA Buff) in t12, mastery does scaled better than haste. How should haste be better with t13 2pc, which does reduce the amount of cobra shots needed even further.

The impact of haste on SV abilites are:

• shorter CobS cast time
• more auto-shots
• focus-reg
• does pets even scale?

Mastery have impact onto following abilities:

• BA / Trap
• Serpent Sting
• Arcane Shot
• Cobra Shot
• Nature Dmg procs (Vishanka)

In my last WOL Profile on Ultraxion Auto-Shot does 15%, Pet does 10% and just for the number: Kill Shot does 3%. Therefor haste is nice for 25% of our dmg, ~70 does favor mastery.

The goal of a haste plateau was to have a near focus neutral rotation which supports our Explosive Shot the most. Thus the best use of the given timeframe between 2 explosive shots. This leads to 757 haste + x modified through everyones latency to get as tight uptime between each Explosive Shots possible..

I recently reforged out of haste and I'm pretty pleased with the results.

One of Survival's challenges is it's erratic. LnL and TotH procs make it impossible to accurately predict what's going to happen in the next 5-second window you have between normal ExS. Still, I find myself using CSx3->ExS often enough and, more significantly, it's the *only* reliable shot sequence I ever encounter. If I'm going to optimize haste for something, then it might as well be the only thing I can be assured of encountering. Prior to reforging (CS was around 1.52 or so), I found myself with really awkward pauses between shots that caused me to either push ExS or delay shots more than I'd like to. It still happens, but less often than before.

If Mastery were really better than haste above 757, then when I reforge all the haste I have that's above 757 into mastery, I should see a dps increase.

I do not.


Reforging out of mastery as much as possible: 41109.69
Female Dwarf: Reforging out of mastery

Stats:
Hit Rating: 961
Crit Rating: 1881
Haste Rating: 1726
Mastery Rating: 515

----

Reforging out of haste as much as possible (try to get to the 757 softcap): 40662.98
Female Dwarf: Reforging out of haste

Stats:
Hit Rating: 961
Crit Rating: 1933
Haste Rating: 1169
Mastery Rating: 1020

I essentially gained ~500 dps by going from the recommended "softcap" value for haste to just completely dropping as much mastery as possible.

To check the mastery vs haste issue simulationcraft is a lot more accurate thanks to the variable fight lenght option. Since FD has sims every instance with the same fight lenght, a variaton on 1 haste rating can push one shot out of the simulation or can add a new one, biasing the value of haste.

This is my profile simulated on SC. default profile, default setup, etc

the mastery>haste values are consistent even if I manually drop haste far below 757



SC summons always my ravager instead of my cat, anyway changing it manually doesn´t change the stat values

I have some questions I'm curious about and think that the introduction page should probably explain more clearly.

Wild hunt (2/2), Go For The Throat (2/2), Sic 'Em (2/2) VS Bestial Discipline (2/3), Frenzy( 2/3), and Shark attack (2/2.)

While I understand that Wild hunt is a fair talent, it doesn't strike me as outstanding with Sic 'Em, here's why.

Wild Hunt requires your pet to be at 50 or more focus in order to receive it's bonus effect. Go for the Throat requires your auto shot crits to regenerate 10 focus, and Sic 'Em requires your Arcane, Explosive, and Black arrow to crit in order to reduce the focus cost of your pets next basic attack.

Where as Bestial Discipline grants an additional 20% focus regeneration (<insert haste equivalent to receive such bonus>) while also having your pet stack 4% attack speed that is up almost 90% of the time it's attacking. This is where Shark attack gets involved, while your pet has a 20% attack speed increase they're also dealing 6% more damage with all their attacks (white damage included.)

I think that Wild Hunt is more burst damage, rather than streamline damage; I would question whether it functions with Sic 'Em and which set of talents provides more damage. By no means am I saying that Sic 'Em isn't worth spec'ing into but, spec'ing into it along with Wild Hunt as a higher priority than Shark Attack, I would be concerned about.

I think there's a lot of talk about it theoretically and not actual simulations about the different talent choices - this is just my opinion.

Correct me where I'm wrong. If I'm onto something here, I think it is a valid option to present to other hunters when it involves spec'ing.

EDIT: Additional tests I've done spec'ing between both and trying some mix-and-matching.

It would seem that *strictly* one or the other are viable. Wild Hunt starts my Cat out with 3200 Damage (without Call of the Wild)

While Frenzy + Shark attack starts my Cat out with 3000 Damage (without Call of the Wild)

Both specs normalize around 2400-2500 depending on trinket procs and Profession abilities. Mixing and matching such as Frenzy + Wild hunt, doesn't seem too good (2600 start for a cat), or Shark attack + Sic 'Em (2400 start for a cat)

Character used Xaxas @ Lightninghoof - Game - World of Warcraft

I decided to give it a try on simulationcraft. In FD, switching to 7/3/31 and 2/2 SA 0/2 WH was a 500 dps loss. But when I simmed it on simulation craft it became an almost 200 dps gain



The first breakdown is 7/3/31 and 2/2 SA, while the second is the standard profile with 2/2 WH

edit: I don´t really know if SC is modeling correctly pet damage. Triying both specs in the dummy for several 5 min trys the pet with standard spec is consistently doing 300-500 dps more, while missing key buffs as 5% crit, kings, and flask. the dummy is not the best place to test, but having many crit buffs missing should bias the result in favour of the 7/3/31 spec

The value of mastery is way above haste on the tendon phase. As SV you will be proccing a L&L in advance of the Tendon and saving it. You will also probably save Chronohunter procs. This results in you casting mostly instants during the 18s burst.