Legion Theorycrafting [7.3, All specs]

[Mavick]

Effin, not many top hunters run haste/mastery anymore. I'd focus more on profiles coming out around 40%+ crit, 20-22% haste then whatever is left over on mastery. Both Stomp/OwtP/KC and Stomp/OwtP/AotB (particularly AotB) favor crit/haste above mastery. And both of those specs perform far better than any DF based spec, particularly when you have 2pc t19 and 4pc t20 eqiupped, which most good raiders do.

[Effinhunter]

Artifact sims

----- BM -----

Crit/Mastery Build, Qapla Boots

BM artifact ilvl and trait upgrades, Single Target, 63 trait base, 2pc/4pc T19/T20, 930/960 ilvl, (Simcraft Source)




BM artifact ilvl and trait upgrades, Three Target, 63 trait base, 2pc/4pc T19/T20, 930/960 ilvl, (Simcraft Source)



- - - Updated - - -

Talents

Crit/Mastery profile, Qapla boots

###### BM Huntmaster Legendary ######

BM top talent combinations, Single Target, 930/960 ilvl, 63 traits, Qapla Legendary, Simcraft Source


BM top talent combinations, Three Target, 930/960 ilvl, 63 traits, Qapla Legendary, Simcraft Source


- - - Updated - - -

Effin, not many top hunters run haste/mastery anymore. I'd focus more on profiles coming out around 40%+ crit, 20-22% haste then whatever is left over on mastery. Both Stomp/OwtP/KC and Stomp/OwtP/AotB (particularly AotB) favor crit/haste above mastery. And both of those specs perform far better than any DF based spec, particularly when you have 2pc t19 and 4pc t20 eqiupped, which most good raiders do.

Just to be clear, I see no indication that haste/mastery is not the top theoretical or not competitive in single target and not worth simming. However, like many others, I have been playing OwtP and gearing crit >= mastery > haste since the start of the tier. There are several factors that have resulted in most of us gearing crit >= mastery > haste.

1.) OwtP is more fun to play and results in having more naturally higher APM while experiencing less perceived downtime. We feel more active which feels more rewarding.
2.) OwtP/Stomp has natural AoE cleaving that allows for smoother rotation and less focus on rotation while dealing with mechanics
3.) OwtP/Stomp is very competitive in single target while also being better in natural AoE
4.) OwtP/Stomp is less focus starved and combined with AotB allows for dipping below the 90 focus threshold without impact on dps (allows you to cast more and feel like you are doing more). The inclusion of SS traits to reduce CobraS cast cost make AotB even more interesting in allowing you to cast CobraS even more. I think the APL for BM might need a bit more focus to see if there are some improvements on AotB dps, especially. The reduction of CobraS with the removal of the need to have enough focus for tandems of KC/CobraS for KC reset in KillerC builds makes AotB very attractive. I've been playing with AotB alot recently, and I've noticed you have too.
5.) Crit gear is much more prevalent in ToS and haste/mastery gear is in shorter supply, so it makes sense to gravitate to OwtP from a gear perspective also. I noticed I was naturally gaining crit versus haste without even focusing on it.

[3clipse]

I have been under a rock for a few months due to a lot of travel, basically just playing to raid (very casually compared to y'all) and dick around on weekends - did AotB get buffed, or is this just a scaling thing? It was turbogarbage before.

[bazzzyl]

Its a pitty that AotB dot does not stack lets say 3x while using boots, that would be sometning.

[Tehterokkar]

Just to be clear, I see no indication that haste/mastery is not the top theoretical or not competitive in single target and not worth simming. However, like many others, I have been playing OwtP and gearing crit >= mastery > haste since the start of the tier. There are several factors that have resulted in most of us gearing crit >= mastery > haste.

1.) OwtP is more fun to play and results in having more naturally higher APM while experiencing less perceived downtime. We feel more active which feels more rewarding.
2.) OwtP/Stomp has natural AoE cleaving that allows for smoother rotation and less focus on rotation while dealing with mechanics
3.) OwtP/Stomp is very competitive in single target while also being better in natural AoE
4.) OwtP/Stomp is less focus starved and combined with AotB allows for dipping below the 90 focus threshold without impact on dps (allows you to cast more and feel like you are doing more). The inclusion of SS traits to reduce CobraS cast cost make AotB even more interesting in allowing you to cast CobraS even more. I think the APL for BM might need a bit more focus to see if there are some improvements on AotB dps, especially. The reduction of CobraS with the removal of the need to have enough focus for tandems of KC/CobraS for KC reset in KillerC builds makes AotB very attractive. I've been playing with AotB alot recently, and I've noticed you have too.
5.) Crit gear is much more prevalent in ToS and haste/mastery gear is in shorter supply, so it makes sense to gravitate to OwtP from a gear perspective also. I noticed I was naturally gaining crit versus haste without even focusing on it.

You are right in saying that Killer Cobra variant values mastery more. But the AotB build values Crit/haste a lot more. My stat weights are still saying to go crit after over 13k rating(about 44%) with haste and mastery being roughly equal depending on gear.

I know it might be a pain in the ass to make 2 sets of gear, but AotB doesn't like mastery that much as the dot doesn't get mastery benefit.

[Mavick]

Just to be clear, I see no indication that haste/mastery is not the top theoretical or not competitive in single target and not worth simming. However, like many others, I have been playing OwtP and gearing crit >= mastery > haste since the start of the tier. There are several factors that have resulted in most of us gearing crit >= mastery > haste.

1.) OwtP is more fun to play and results in having more naturally higher APM while experiencing less perceived downtime. We feel more active which feels more rewarding.
2.) OwtP/Stomp has natural AoE cleaving that allows for smoother rotation and less focus on rotation while dealing with mechanics
3.) OwtP/Stomp is very competitive in single target while also being better in natural AoE
4.) OwtP/Stomp is less focus starved and combined with AotB allows for dipping below the 90 focus threshold without impact on dps (allows you to cast more and feel like you are doing more). The inclusion of SS traits to reduce CobraS cast cost make AotB even more interesting in allowing you to cast CobraS even more. I think the APL for BM might need a bit more focus to see if there are some improvements on AotB dps, especially. The reduction of CobraS with the removal of the need to have enough focus for tandems of KC/CobraS for KC reset in KillerC builds makes AotB very attractive. I've been playing with AotB alot recently, and I've noticed you have too.
5.) Crit gear is much more prevalent in ToS and haste/mastery gear is in shorter supply, so it makes sense to gravitate to OwtP from a gear perspective also. I noticed I was naturally gaining crit versus haste without even focusing on it.

https://www.raidbots.com/simbot/repo...EwVmhH9Su8wHDp

^ My statweights, you can view the gear and stats in the html link

And that hasn't really changed much, outside of ordering between crit and haste, in over 10 ilevels of ToS gearing. Even with KC, mastery was below both crit and haste for me at the time. I have no idea what you're looking at that values haste/mastery so much, outside of DF builds, which by and large are so rare to see these days, and typically not going to scale near as well as KC/AotB Stomp builds with 2pc t19.

[Effinhunter]

You are right in saying that Killer Cobra variant values mastery more. But the AotB build values Crit/haste a lot more. My stat weights are still saying to go crit after over 13k rating(about 44%) with haste and mastery being roughly equal depending on gear.

I know it might be a pain in the ass to make 2 sets of gear, but AotB doesn't like mastery that much as the dot doesn't get mastery benefit.

It's not a pain at all, and it's important to learn something from the gear, talent and ability interactions. Stat weights, as I've mentioned before, can be very misleading. They are instantaneous for a gear point and are not very effective for gauging stat trends, such as you might see in reforge plots. Haste, since Vanilla, has been especially misleading in instantaneous stat weights, and people take the darnedest conclusions from seeing their stat weights. Even if they have 13k crit and 11k mastery, they see stat weights that show haste ahead for the next single attribute gain, and extrapolate that the trend was solid for all gear points up to this point and all after.

Anyway, as I've said before, I'm looking around for profiles that are worth simming. Crit/haste shows some promise. Crit/mastery. Mastery/haste. I'll be simming all three until I'm satisfied that one or more of them are useless.

- - - Updated - - -

MM Artifacts with Deadly Aim readded.

----- MM -----

MM artifact ilvl and trait upgrades, Single Target, 63 trait base, 2pc/4pc T19/T20, 930/960 ilvl, (Simcraft Source)



MM artifact ilvl and trait upgrades, Three Target, 63 trait base, 2pc/4pc T19/T20, 930/960 ilvl, (Simcraft Source)



MM artifact ilvl and trait upgrades, Single Target, True Aim selected, 63 trait base, 2pc/4pc T19/T20, 930/960 ilvl, (Simcraft Source)

[Mavick]

It's not a pain at all, and it's important to learn something from the gear, talent and ability interactions. Stat weights, as I've mentioned before, can be very misleading. They are instantaneous for a gear point and are not very effective for gauging stat trends, such as you might see in reforge plots. Haste, since Vanilla, has been especially misleading in instantaneous stat weights, and people take the darnedest conclusions from seeing their stat weights. Even if they have 13k crit and 11k mastery, they see stat weights that show haste ahead for the next single attribute gain, and extrapolate that the trend was solid for all gear points up to this point and all after.

Anyway, as I've said before, I'm looking around for profiles that are worth simming. Crit/haste shows some promise. Crit/mastery. Mastery/haste. I'll be simming all three until I'm satisfied that one or more of them are useless.

Statweights aren't nearly as meaningless as you're making them sound. No, they're not always gospel, and yes, there has been some bugs with haste. But even ignoring the statweights you can simply add haste to your gear by switching enchants/gems, which is not an insignificant amount these days, and see both your sim and ingame results go up, as I have done.

I mean, I get it, it sounds sacrilegious to not be aiming for high mastery, and when the hunter community on discord was figuring out that mastery just wasn't all that important anymore unless you somehow manage to get below 50%, which is nigh impossible with TOS gear, there was a lot of ????? And in the end it's the result of several factors, probably the biggest two being the neutering of Kill Command damage through trait nerfs plus the repeated buffs to Cobra Shot in relation.

To put it in perspective, here's my leggo combination sims for AotB spec:

https://www.raidbots.com/simbot/repo...7Rue8Qqs9ZH6hP

Note how my 2nd highest simming leggo setup is Sephuz+Prydaz (no boots). That's a drastic reduction in the amount of Kill Commands I'd use in a fight, and by simple logic, a great increase in the number of Cobra Shots. Just from that fact combined with the raw stats of Prydaz that nearly equals my best setup for that spec, and that spec sims higher than Killer Cobra or Dire Frenzy.

[Effinhunter]

Statweights aren't nearly as meaningless as you're making them sound. No, they're not always gospel, and yes, there has been some bugs with haste. But even ignoring the statweights you can simply add haste to your gear by switching enchants/gems, which is not an insignificant amount these days, and see both your sim and ingame results go up, as I have done.

I mean, I get it, it sounds sacrilegious to not be aiming for high mastery, and when the hunter community on discord was figuring out that mastery just wasn't all that important anymore unless you somehow manage to get below 50%, which is nigh impossible with TOS gear, there was a lot of ????? And in the end it's the result of several factors, probably the biggest two being the neutering of Kill Command damage through trait nerfs plus the repeated buffs to Cobra Shot in relation.

To put it in perspective, here's my leggo combination sims for AotB spec:

https://www.raidbots.com/simbot/repo...7Rue8Qqs9ZH6hP

Note how my 2nd highest simming leggo setup is Sephuz+Prydaz (no boots). That's a drastic reduction in the amount of Kill Commands I'd use in a fight, and by simple logic, a great increase in the number of Cobra Shots. Just from that fact combined with the raw stats of Prydaz that nearly equals my best setup for that spec, and that spec sims higher than Killer Cobra or Dire Frenzy.

They're only useful if clear trends have been established, and I'm not necessarily saying that yours are useless, only that they have traditionally been extremely misleading. You are geared for heavy crit and mastery but are saying crit and haste are top because of what adding a point to your current high crit/mastery build does with AotB and the lack of mastery impact on AotB. I'm in no way arguing that crit/haste is bad to sim or any other combo. What I'm saying is that I'm interested in seeing clear stat trends for as many competitive talent/legendaries as possible and providing clear, reproducible results for others.

As someone very familiar with the simcraft code and the hunter module, I can tell you that haste is problematic, especially if any rppm items/abilities, or talents are involved. I always try to take the stat weight results for haste, especially, with a grain of salt.

[Mavick]

They're only useful if clear trends have been established, and I'm not necessarily saying that yours are useless, only that they have traditionally been extremely misleading. You are geared for heavy crit and mastery but are saying crit and haste are top because of what adding a point to your current high crit/mastery build does with AotB and the lack of mastery impact on AotB. I'm in no way arguing that crit/haste is bad to sim or any other combo. What I'm saying is that I'm interested in seeing clear stat trends for as many competitive talent/legendaries as possible and providing clear, reproducible results for others.

As someone very familiar with the simcraft code and the hunter module, I can tell you that haste is problematic, especially if any rppm items/abilities, or talents are involved. I always try to take the stat weight results for haste, especially, with a grain of salt.

It's not actually necessary to be super-familiar with simcraft coding these days to get a handle on how to sim yourself and weigh those results, particularly since the advent of Raidbots. Nor is it hard to figure out what is going on with the module when you can talk to the people who maintain them on discord. In fact, I'm not even sure if you're aware of the recent tweaks to how it handles AotB with boots equipped, at which point people started realizing just how better the spec scaled.

But anyway, we got kind of sidetracked on the haste issue, and your'e right, I'm not actually running a "crit/haste" build, I'm more "crit/mastery" and that's the way it will look for the forseeable future, because haste does run into breakpoints, and tends to fall off past 22% from what I've seen. But maybe that changes once I finish out my t2, I don't know, I'll have to re-sim and see.

But my biggest question was why you were originally favoring haste/mastery profiles for OwtP specs like KC/AotB when it should be obvious that crit/mastery or crit/haste should do better, and is actually what hunters are gearing these days. I mean, hell we already saw it on the thread: people look at your sims and assume DF then KC are the way to go when AotB actually does better the higher your ilevel goes. That's just misinformation tbh.

[Effinhunter]

It's not actually necessary to be super-familiar with simcraft coding these days to get a handle on how to sim yourself and weigh those results, particularly since the advent of Raidbots. Nor is it hard to figure out what is going on with the module when you can talk to the people who maintain them on discord. In fact, I'm not even sure if you're aware of the recent tweaks to how it handles AotB with boots equipped, at which point people started realizing just how better the spec scaled.

But anyway, we got kind of sidetracked on the haste issue, and your'e right, I'm not actually running a "crit/haste" build, I'm more "crit/mastery" and that's the way it will look for the forseeable future, because haste does run into breakpoints, and tends to fall off past 22% from what I've seen. But maybe that changes once I finish out my t2, I don't know, I'll have to re-sim and see.

But my biggest question was why you were originally favoring haste/mastery profiles for OwtP specs like KC/AotB when it should be obvious that crit/mastery or crit/haste should do better, and is actually what hunters are gearing these days. I mean, hell we already saw it on the thread: people look at your sims and assume DF then KC are the way to go when AotB actually does better the higher your ilevel goes. That's just misinformation tbh.

The Qapla changes to the APL were suggested in this thread. I was testing Qapla APL changes and sharing results long before it was reflected in the default APL.

If you don't understand the coding and model of Simcraft, then you can't really understand why relying on raidbots for stat weights can be problematic, since it relies on Simcraft.

The original profiles for haste/mastery were reflective of the DF build's superiority before this tier, but I have no intention of continuing to support such sims in the long term if they are a waste of time. There are examples I have shown here of OwtP that showed crit/mastery not beating haste/mastery at the time and I had requested profiles during this tier of anyone who could find (this was before nerfs to JoT and PL). OwtP started to gain ground with the changes to Call of the Wild that were a result of the changes to Mantle's effect and the overall 2 charge implementation to DF/DB and the resulting change to Call to compensate. This brought about interest in simming Huntmaster with OwtP and the trend of crit increasing significantly, which was reported here. If you've been following the thread, then you should be able to see why the profiles are in their current state and how willing I am to change things to help others.

As usual, you are extrapolating things from the thread that were not intended and claiming to speak for the hunter community. And then there's the misinformation meme that has been so common since the Hunter Discord started, especially. All of which are annoying. Please post useful things. Stop posting accusations (e.g., "misinformation" in the quote above) and misleading statements (e.g. crit > haste is optimal when you are gearing for high crit and mastery). I can look into these things as I get time, or you can post your own sims that help the community get more knowledgeable using profiles and ideas that will help them, if what you're trying to do is help.

- - - Updated - - -

P.S. Another potential useful thing that needs to be investigated is the interaction of the various crucible traits with stats. I'm definitely thinking about how to present this correctly. The armory still has not incorporated the crucible changes, so using Raidbots or even Simcraft (with armory=realm,server,character) will be a bit rough without direct importation of a character profile via the Simulationcraft addon (the one you use /simc to copy simc profiles with, which does include the crucible=* string).

P.P.S. Again, for the thousandth time, this thread was never intended to be just me posting useful sims. Everyone is encouraged to contribute in any way that you can to help us all learn. The first post will be updated with anything we consider especially useful, so please, feel free to help.

[Khallid]

The original profiles for haste/mastery were reflective of the DF build's superiority before this tier, but I have no intention of continuing to support such sims in the long term if they are a waste of time.

I don't think they're a waste of time per se, because it's impossible to know if one build continues to win over another without testing for it with new changes. My impression is that OWTP will stay ahead at least until tier 21 comes out though.

[Mavick]

The Qapla changes to the APL were suggested in this thread. I was testing Qapla APL changes and sharing results long before it was reflected in the default APL.

If you don't understand the coding and model of Simcraft, then you can't really understand why relying on raidbots for stat weights can be problematic, since it relies on Simcraft.

The original profiles for haste/mastery were reflective of the DF build's superiority before this tier, but I have no intention of continuing to support such sims in the long term if they are a waste of time. There are examples I have shown here of OwtP that showed crit/mastery not beating haste/mastery at the time and I had requested profiles during this tier of anyone who could find (this was before nerfs to JoT and PL). OwtP started to gain ground with the changes to Call of the Wild that were a result of the changes to Mantle's effect and the overall 2 charge implementation to DF/DB and the resulting change to Call to compensate. This brought about interest in simming Huntmaster with OwtP and the trend of crit increasing significantly, which was reported here. If you've been following the thread, then you should be able to see why the profiles are in their current state and how willing I am to change things to help others.

As usual, you are extrapolating things from the thread that were not intended and claiming to speak for the hunter community. And then there's the misinformation meme that has been so common since the Hunter Discord started, especially. All of which are annoying. Please post useful things. Stop posting accusations (e.g., "misinformation" in the quote above) and misleading statements (e.g. crit > haste is optimal when you are gearing for high crit and mastery). I can look into these things as I get time, or you can post your own sims that help the community get more knowledgeable using profiles and ideas that will help them, if what you're trying to do is help.

That's funny, as I remember witnessing the conversations taking place with Azor and nuoHep on Discord when the changes were implemented. Pretty sure whatever comments were made on the thread, it was probably the result of previous conversations on Discord, as it's vastly quicker at identifying things than these boards are, and has been all expansion.

I don't fully understand how the coding works on Simcraft, yet I know why to not fully trust implicitly the state weights presented by it. But I also know when to not disregard them when I can back the results up ingame. Funny how that works.

The misinformation theme persists because it's true. Nobody really gains much from the sims you post, to be downright honest, and they haven't for quite awhile. That's why more and more people regard these boards as a memepark these days. I don't have to extrapolate much when I can see firsthand the level of intelligence and useful information that comes from the hunter discord as opposed to these boards, and who contributes to both. You, on the other hand, have participated only in these board discussions while pretending that that is where good hunters go for their information these days, while completely ignorant of what is said and happens in the other.

And finally, I was trying to help you. I simply suggested you focus on profiles that favored crit and haste more for your AotB sims. You could have simply said "yeah maybe I'll look at that" but instead it sounded more like even tho you yourself was gearing that way you're going to keep presenting sims based on stat profiles that nobody (or at least very few competitive hunters) uses that show a bunch of irrelevant results.

Edit: Inb4 the next however many posts of people rushing to soothe Effin's ego, telling him how great and useful his work is for their normal/heroic progression and we perpetuate the cycle over again.

[Effinhunter]

I don't think they're a waste of time per se, because it's impossible to know if one build continues to win over another without testing for it with new changes. My impression is that OWTP will stay ahead at least until tier 21 comes out though.

From my tests, OWTP has been ahead basically since the changes to Call of the Wild after the Mantle nerf/change. I don't see any reason for that to change.


In regards to my earlier thoughts on stat interactions with crucible traits, I have found the information below. I am not sure which of these traits, if any, interact with crit and crit modifiers on gear/talents/cds. Has anyone monitored AotW's crit boost on any of the crucible traits or any interactions with Trueshot for MM?

Source: https://raw.githubusercontent.com/si.../allspells.txt


Crucible traits affected by stats (haste related so far):


Shadowbind (id=252875)
haste: RPPM of 2, affected by haste

Chaotic Darkness (id=252888)
haste: RPPM of 2, affected by haste

Dark Sorrows (id=252922)
haste: RPPM of 1, affected by haste

Secure in the Light (id=253070)
haste: RPPM of 3, affected by haste

Infusion of Light (id=253093)
haste: RPPM of 4, affected by haste

Refractive Shell (id=252207), non-dps trait
haste: RPPM of 2, affected by haste

Light's Embrace (id=253111), non-dps trait
haste: RPPM of 8, affected by haste


According to spell dump, the following proc-based trait is not affected by haste

Torment of the Weak (id=252906)
PPM of 4 (does not show a haste multiplier)


Anything that depends on Concordance of the Legionfall (id=239042), does not appear to be affected by haste either. This would appear to include:
Shocklight (id=252799)
Murderous Intent (id=252191)

Concordance says the proc rate is 2 and does not indicate a haste multiplier.

[Khallid]

In regards to my earlier thoughts on stat interactions with crucible traits, I have found the information below. I am not sure which of these traits, if any, interact with crit and crit modifiers on gear/talents/cds. Has anyone monitored AotW's crit boost on any of the crucible traits or any interactions with Trueshot for MM?

The crucible traits seem to crit normally with whatever crit chance I have at the time it procs. I only have logs with Dark Sorrow (my other traits are Shocklight), and it crits essentially in line with my active crit rating (I averaged 54% crit chance over 50 procs), which is higher than my passive crit rating (47%), and its crit chance looks totally normal in line with the rest of my abilities.

[Effinhunter]

The crucible traits seem to crit normally with whatever crit chance I have at the time it procs. I only have logs with Dark Sorrow (my other traits are Shocklight), and it crits essentially in line with my active crit rating (I averaged 54% crit chance over 50 procs), which is higher than my passive crit rating (47%), and its crit chance looks totally normal in line with the rest of my abilities.

Thanks! I'll put some log analysis on my task list, and see if there are any traits that are not behaving as expected. If true, crit and haste should elevate naturally then for most of the Crucible traits, and we'll want to post some sims that show how all of this will have effect. I see no reason why mastery might be stealthily useful in BM's case as none of these traits appear to produce a pet that might be taken into account. MM's mastery does not appear to apply here either, but it would be good to know if major CDs with crit or crit dmg modifiers are being applied to crucible traits and we can find that out in logs.

[DotJun]

Are we back to using the boots when t21 comes out?

[Effinhunter]

Are we back to using the boots when t21 comes out?

Haven't done T21 breakdowns and will wait for those for a bit closer to the content. I can say the boots are very good for AotB, but I need to rerun legendaries with more configurations (e.g., crit/mastery, crit/haste) before I can give definitive results. I'm working all this weekend to get robotics demos ready. I may not be able to do much this week. We have some time before next tier though.

[mishal]

How does torment the weak trait work? Is there any point in getting more than 1 rank since the stack doesn't go above 3? I have 2 ranks so far and i feel like i messed up big.

[ozusteapot]

What are your thoughts on the viability of the AotB build, with Qa'pla+Sephuz/belt when t21 gets released?
I've started to really enjoy the build and am now pouring some AP into my BM artifact so as to be ~3 traits behind my main, MM. From a basic look at BM's t21 set bonuses, to me it looks like the AotB+Qapla build will still be viable, but I'm looking into other people's opinions on it.
Currently this spec is a great alternative for going MM on heavy movement fights, and I really want to be using it while progressing on M Avatar.

Also, could someone explain how can Sephuz be among the best Bm legos even on fights where you can't interrupt, like Sisters or Avatar? Is the passive haste buff just that good?