Yes, MM is a priority shot selection spec. The term rotation is only used as a loose term to illustrate the ideal shot selection under certain conditions, with those conditions changing dynamically. Here is some suggestions to try to help you:
- Try to follow each CS and AI hardcast special with an SS pair. This helps to maintain the ISS buff and balance focus usage.
- If there is enough additional room in your rotation before CS comes off CD, cast another SS.
- MM AI procs shouldn't affect your "rotation" much. Plus, you do not have to use it right away. Never use it in the middle of an SS pair. Its more important to maintain the ISS buff than it is to have one more SS to possibly gain 1 stack of MMM. If your rotation only has 4 SSs, then replace the next hardcast AI with it if you have a tight rotation, or if your rotation has some deadtime before CS, do it before the CS and delay it a bit. If your rotation does more than 4 SSs, replace one of the extra SSs with it.

The reason for RK is not the xp since we are already xp cap. The reasons are clearly spelled out in the guide. The main is its synergy with RR, where you get 50 focus everytime you get a killing blow. This is great on trash or AoE fights. It can even help on boss fights with adds. I know my AoE during the Magmaw and Maloriak fights greatly benefit from it. The other reason is the damage bonus on the next applicable shot you take after a killing blow.

Testing my DPS last night I can at least say that only using CS to refresh SrS and using AiS as your main focus dump - and dropping Arcane shot completely - is very effective

Like Whitefyst said earlier, the optimal "rotation" for me wasn't a rotation at all; There was just a list of things that I had ticking in my head throughout the whole fight:

1. Have enough focus every ~13.5 seconds to fire a Chimera Shot.
2. Shoot steady shots in pairs to keep up ISS
3. Aimed shot every 2 or 3 steady shots
4. Exploit that bug with "Ready, set, aim..." for free aimed shots
5. Have preference for being focus capped vs focus starved, else you run the risk of losing SrS on target

That's basically it. For example, an ideal "rotation" for me looked something like this: SS > SS > AiS > SS > SS > AiS > SS > SS > SS > CS, repeat. Of course it was different when I got "Fire!" procs, during movement, and when I was trying to use that AiS bug to my advantage.

A thing to note for SV hunters is you can switch back to Ferocity pets now and put 2/2 Wild Hunt (it's actually better than a cunning now). Since there is no actual way of tracking uptime for Wild Hunt in game, I've resorted to FD for the answer on it's uptime: ~90% in my current gear (358 avg ilevel).

edit:

Info on my character:

• RF, AiS, SS glyphs.
• Reforged all mastery/haste into hit/crit, and then slowly reforged the excess hit into haste/crit. Basically, every single piece of gear I had was reforged so that I could minimize the amount of lesser secondary stats I had (haste/mastery) and transfer them into better secondary stats (crit/hit).
• Link to FD settings: Zeherah's Hunter DPS Analyzer. If there is any way to make FD only use CS to refresh SrS, please tell.
• Haste: 911
• Crit: 1902
• Mastery: 928
• Hit: 961
• I have 4pc.
• 3/3 Pathing

So, like pretty much everyone here I have been doing some simming in FD to play around with the new MM rotation. I noticed that FD estimates hard casting AI outside the CA phase to be a whopping +52 DPS increase over only hard casting during CA, and using AS in pairs for the rest of the fight. This seems inconsistent with what I am reading here, however it DOES make the rotation much more forgiving and movement friendly, which situationally speaking (assuming the theoretical DPS is as close as FD makes it out to be) would probably make the ASx2 outside higher practical DPS.

Thoughts? Like I said I know those sim results are not the apparent norm here, hence my question. Here is my armory link, in case it's relevant.

Vysin @ Sargeras - Game - World of Warcraft

Inconsistent with what I'm doing in FD. Save settings and make them publicly viewable. As for heavy movement fights, using arcane shot will definitely yield higher dps.

Silens I think the issue may be the fact that you don't have buffs enabled. The buffs are going to raise the gap between aimed and arcane. If you're trying to compare the difference between hardcast and just arcane shot during the portion of the fight that isn't affected by careful aim it might be easiest to just change the portion of the fight spent in careful aim to 0%. If I enable raid buffs for you and just look at a fight where careful aim and kill shot are never active, I'm seeing a 1.5k dps difference between hardcast aimed shot and arcane shot. This is also including your arcane shot glyph, which might not be the best choice if you're going to be hardcasting aimed.

Try the following for a noticeable increase in both FD sim and (I've no doubt) target dummy dps:

1. dump KS & AS glyphs for Aimed Shot (to see true benefit of the AiS focus dump rotation) & Rapid Fire
2. you're more than 80 hit rating over cap. Reforge 71-89 hit off of one of your pieces to crit
3. remove Arcane Shot from rotation (just drag it out to the 'inactive' list).
4. switch up pet talents a skosh: move 2 points from shark attack to wild hunt
5. i started by unreforging all your gear then reforging hit to crit (when possible) or haste (when no crit) to get as little over cap as possible. then moved remaining unreforged mastery to crit (when possible) & haste (when no crit), then haste to crit.
6. changed your agi/haste gem to agi/crit, since per the differentials, 10 agi > 20 haste dps wise

These changes took 14961 to about 15911. 6.3% or so isn't whopping, but nothing to sneeze at either considering it's with the same gear.

I see you wrote that you manually removed AS from rotation, but when I hit your setting and do the same, I see an increase & not a decrease in DPS, even without doing any of the other things.

Oh yeah...get the +15 stats for the tunic. Very cheap enchant!

Even on Atramedes & Cho'gall (a fight where I had the pleasure of misdirecting adds to the tanks, moving with the tank as he moved the add to the steps, and dropping traps & using AOE on the adds), which is what I would consider two of the most movement-heavy fights in this tier, it wasn't at all a challenge for me to get off hardcasted Aimed Shots. When you have to move, switch into Fox and shoot Steadys, or, if you're high on focus, replace an Aimed with a Chimera. There has yet to be a situation where I need to Arcane.

In the rotation you described (only using Chimera to refresh SrS) you can squeeze in a couple more steadys and, every other rotation or so, an extra Aimed since you only need to use Chimera every 12-14s depending on how far away from the boss you are.

Anyways, I don't think you should get in the mindset of 'this is my rotation, this is how I'm going to do it.' If you have to move, lay traps, cast Misdirect, Feign Death, or any of the number of things you need to do in a fight your rotation falls apart and you won't be able to keep up. I look at it as a priority system that look something like this:

Hunter's Mark
Serpent Sting
Aimed Shot
Steady Shot Pair
Chimera Shot (if SrS is about to fall off)
Single Steady Shot

In the CA phase I spam Aimed with SrS ticking, throwing in a Cobra Shot every 9s to keep it up. It gets a little messy with all the focus regen, but it's the highest DPS rotation I've been able to come up with on a dummy in the CA phase.

This entire argument is based off the assumption that waiting for less than a gcd for Chimera Shot to come off CD rather than fitting in another Steady Shot and slightly delaying Chimera is optimal. In an ideal theoretical situation I agree with you that that would be the case for sufficiently short wait times and the breakpoint for whether or not it's worth waiting for Chimera to come off CD rather than starting another SS could be found using:

t_wait/t_cast*D_steady = (t_cast-t_wait)/10*D_chimera

Working that out for a reasonable gear level gives ~0.5-0.6 seconds as the amount of time worth waiting for CS to come off CD. My point is that in an actual raiding situation the combination of time varying latency, reaction time and small amounts of movement will prevent you from encountering the situation where you are regularly waiting <0.5s for CS to come off CD which would significantly devalue hastes effect on specials. I think to maximize actual raid dps you will virtually always be better off queuing up an additional steady shot and delaying CS to minimize time not using an ability.

As for the plot, the priority queue used to generate it will always start another Steady Shot rather than waiting for CS to come off CD. Simming using a priority queue where the lowest priority Steady is only used if the CS CD remaining is >0.5s is universally a dps loss. As to your note about the maximum haste shown in the plot being insufficient for the 5/6s SS rotations that's not actually the case. The maximum haste is actually 2825, 0 on the plot corresponds to the default haste in the gear set (1425) so the range is really from 0 to 2825.

Of course you are correct that you can't add more haste without losing other stats. The point of the plot was to show that I believe in an actual raiding situation haste's value is roughly independent of your current amount of haste and is always greater than the value of mastery so it isn't worth reforging haste into mastery to match one of the ideal rotations, rather you should use additional SS's as possible to minimize time not casting even if it slightly delays CS.

Finally, contributions to dps that are independent of haste don't mean that dps doesn't scale linearly with haste, just that the coefficient is lower. As for the dps contributions that should scale as step functions in the ideal case I agree, but believe I've already made my point about why I dont think that will be seen in reality.

This statement is very false and misleading. While it is true that if you pick certain end haste points and you plot a line from the start point to the finish it may show an increase in DPS (but it may show a descrease) as you say. However, that is very misleading since there are actually peaks and valleys where adding haste sometimes helps and sometimes hurts.

To illustrate, below is a table produced using FD where I took my current character as is where I have reforged to what I think is as ideal a level of haste I can get with my current gear from empirical analysis and reforge limitations. Then I have added haste in steps of 50 with substracting first an equal amount of mastery until I run out of mastery rating (at 650) and then crit until FD will no longer allow me to subtract more crit (at -999 for a total change in haste of 1649). This analysis was done with waiting at most 0.3s for CS to come off CD. If there is longer than a 0.3s wait, then likely SS is cast instead, although in some situations, another AI could have been cast.

Haste Added	DPS Change Per Step	Total Change
0	27274	0
50	17	17
100	13	30
150	-22	8
200	-28	-20
250	34	14
300	12	26
350	-17	9
400	-30	-21
450	58	37
500	-23	14
550	-9	5
600	224	229
650	42	271
700	-5	266
750	10	276
800	-80	196
850	-83	113
900	-54	59
950	187	246
1000	8	254
1050	-37	217
1100	26	243
1150	-141	102
1200	-13	89
1250	41	130
1300	12	142
1350	-38	104
1400	5	109
1450	-51	58
1500	-118	-60
1550	-48	-108
1600	75	-33
1649	-50	-83

This table clearly illustrates that DPS does not increase linearly with gear for all haste trade off with other stats. If I could reforge more haste, I really do not get much of a benefit for the first 550 or so haste I reforge, where the maximum increase with the steps is 37 total DPS. Which I am fine without, especially considering that on movement fights where I can do less AIs that the value of that additional haste is worth a heck of a lot less.

Around 600 haste rating, there is a nice DPS increase since I was able to cast 1 extra AI over the full 5 min duration of the fight, but shortly after that point, DPS decreases with more additional haste.

At 1649 haste rating added, I actually lose 83 DPS.

Thus, in summary, always blindly adding haste is not always a DPS increase. While it is true that adding the right amounts of haste can be a nice DPS boost, just adding haste indiscriminately may or may not be.

While reading too much into one encounter by one of the best geared hunters in the world is probably a little foolish, almost everything he's doing does seem to fall in line with what's been stated in this thread. Hardcasting Aimed is the way to go, Chimera is now simply for extending SrS, and Arcane shot is best saved for heavy movement only. Nothing too astonishing as far as that goes. He is also prioritizing for Crit wherever possible in gear which, again, is the general consensus so far.

That said, again, reading too much into specific glyphs or enchants that he's done might be a bit premature. He's extremely well-geared, and the patch has only been around for a couple days, so taking everything he does (Spec, enchants, etc.) as gospel might not be the best idea.

One thing to remember is that the highest parses will often have much higher than average Aimed shot crit rates. That Chimaeron kill was about 5:45, and my kill was the same time length, but I had only 43% Aimed crit to his 67%.