Protection legion Discussion

[Deleted]

thanks marok, all i really hope is thaty they take revenge off the gcd if they want to push this trough i dont want the vengeance buff rotation conflicting with my GCD lock as it would just make it feel really clunky and frustrating.

[Deleted]

Why are we getting even more nerfs? I mean - now we have no interaction between Ultimatum and Vengeance (and Battle Cry), we need to either select living a bit more on multi-mob M+ packs or hitting them for 30 rage and applying Deep wounds, and let's not forget about the "cherry" on the top - a 50% increase in rage cost of Shield block... is this all REALLY offset by the increase to Thunderclap damage and it generating a whooping 3 rage?

[ReD-EyeD]

The changes are so damn random that I can't really say if it is clearly a buff or a nerf

[vsb]

It certainly is interesting that with the overwhelmingly negative response that demon blades recieved, they took this very clearly disliked and universally hated talent and gave it to another class.

It just baffles the mind. They said they wanted to make all talents viable in some way, and here is devestator, which will be a wasted talent slot that no one takes, unless it ends up the best in that row, in which case it just leaves two talents people want to use but cant.

In no way is this a good idea.

Well, if this talent is competitive, I would use it. I hate the concept of smashing every GCD with something and if I won't be gimped too much for taking this talent, I'll certainly use it.

[ReD-EyeD]

Well, if this talent is competitive, I would use it. I hate the concept of smashing every GCD with something and if I won't be gimped too much for taking this talent, I'll certainly use it.

I don't want to choose this talent for the exact opposite reasons. I really want to press something every GCD and as fast as possible. This is fun for me. Opinions, opinions.

[Enitzu]

Yes, Devastator is "bad", and it's supposed to be. It's not there for those who want to maximize their DPS/survivability. It's simply there so people who want to make Protection easier to play have an option. There are people (more than you'd think) who are going to prefer this more passive/easy playstyle. Now, I'll admit that it does suck for the more serious players because it essentially removes a potential talent choice, however, it still is a choice for a lot of other people. If for some reason Blizzard decides to buff Devastator to the point where it's better than the other harder to play options, then that's a big problem. Also, if you're wondering if Devastator will be worth taking for DPS or survivability, the answer is no. At least not in Prot's current state on PTR.

Here's something I wrote in the Warrior discord about the strength of Devastator:

You seem to have missed a lot of factors in this. Oh and before I begin, since DH came into play I have hated the passive playstyle aka Devestator type yet Blizz still made it the best talent for DH's. Saying it won't happen here is like saying Blizz listens to the playerbase, which we all know is BS.

First off, your assumption with dev is that it will generate 613 rage. What is that based on? Fight length? Time on target? Your personal haste values? There are a ton of things that can make this number go drastically up or down. So trying to base napkin math off of your personal stats will only lead to false conclusions.

Now if we go by actual fight numbers, let's use the rank 1 mythic parse on Nyth for this: https://www.warcraftlogs.com/reports...&ability=23922 then there were 105 melee attacks done. From those 105 alone, you would be getting 525 rage. Now with a 30% chance to reset shield slam you can safely assume that we would have gotten an extra 31 shield slams in for another 620 rage. For a total of 1145 rage. Now considering that without Vengeance, IP costs 60 rage we would have 19 IPs just from that rage gain alone. That is NOT taking into account vengeance procs for a reduced cost. Now lets say that we used all of these IPs after hitting revenge, like we do now with FR. That cuts the cost of IP to 39 per use and means that we could have gotten 29 IPs out of that same rage gen.

Looking at the same log, let's check out the current system. The fight was 3:56 long. That means without a cd reset on SS he could only have used SS 26 times. Yet he had 45 in the fight. So we can safely assume that the other 19 were resets from Devastate. That gives us an extra 380 rage which means even using the reduced costs, we wouldn't get 10 extra IPs out of it. Seeing an issue yet?

Since this guy was running Indom, his avg IP was 1.2 mill. Using that number and taking out the 25% for losing Indom gives us 900k. So only using the examples of "free" rage here, total IP healing with Indom in the current build would be 10.8 mill. Total IP healing with dev using upcoming build (what we know) would be 26.1 mill. Granted this is not taking into account any rage gained from TC.

Feel free to correct my math on this if I am wrong. But do so with actual facts and numbers, not made up information.

[kiklion]

I don't want to choose this talent for the exact opposite reasons. I really want to press something every GCD and as fast as possible. This is fun for me. Opinions, opinions.

Which is fine as long as the talent is balanced. I don't believe they will come close to balancing it with indomitable. However I do believe they can balance its damage output which is why I think it should be on talent line with endless fury and avatar.

- - - Updated - - -

...

Warcraft logs isn't loading on my phone but your math is off. 30% of autoattacks doesn't equal the number of bonus Shield slams. A proc while on global the global before SS naturally reset is wasted. A proc with 50% of SS cool down remaining is only half a SS gain.

Looking at live numbers you don't take # of SS - (fight length / SS cd) because that ignores the partial resets.

Do the same math for both. Take # of autos and multiply by .3 then take number of devastates and multiply by .3. Then add the extra 5 rage per auto. Compare them then.

Another way of wording your mistake is that you are essentially assuming every auto SS proc happens the same global that you SS and can immediately SS again with no loss vs as many devastate procs as possible happening the global before shield slam and getting ignored.

It loaded partially for me, he got 102 devastates which would be 31 procs for SS. So it's basically a comparison of 505 rage vs 25% extra rage on everything... which includes damage taken which you completely forgot about.

Not even trying to assign a value to 25% hp and that you get the 25% extra IP when you need it... immediately after a big hit which just generated a lot of rage from damage taken.

[ReD-EyeD]

Which is fine as long as the talent is balanced. I don't believe they will come close to balancing it with indomitable. However I do believe they can balance its damage output which is why I think it should be on talent line with endless fury and avatar.

I'm much more interested in rage gain and shield slam reset than actual damage tho. 5 rage every hit and 30% slam reset (don't forget it hits off gcd unlike devastate) seems quite A LOT to me.

[Enitzu]

Which is fine as long as the talent is balanced. I don't believe they will come close to balancing it with indomitable. However I do believe they can balance its damage output which is why I think it should be on talent line with endless fury and avatar.

- - - Updated - - -



Warcraft logs isn't loading on my phone but your math is off. 30% of autoattacks doesn't equal the number of bonus Shield slams. A proc while on global the global before SS naturally reset is wasted. A proc with 50% of SS cool down remaining is only half a SS gain.

Looking at live numbers you don't take # of SS - (fight length / SS cd) because that ignores the partial resets.

Do the same math for both. Take # of autos and multiply by .3 then take number of devastates and multiply by .3. Then add the extra 5 rage per auto. Compare them then.

Another way of wording your mistake is that you are essentially assuming every auto SS proc happens the same global that you SS and can immediately SS again with no loss vs as many devastate procs as possible happening the global before shield slam and getting ignored.

It loaded partially for me, he got 102 devastates which would be 31 procs for SS. So it's basically a comparison of 505 rage vs 25% extra rage on everything... which includes damage taken which you completely forgot about.

Not even trying to assign a value to 25% hp and that you get the 25% extra IP when you need it... immediately after a big hit which just generated a lot of rage from damage taken.

You bring up some valid points. My train of thought for comparing them however was that you will take the same dmg regardless. The only thing we are comparing here is the 2 talents so I tried to stay away from other sources. Where is this 25% extra rage on everything coming from though? The HP I completely agree with. It's a valuable talent no doubt but if the rage gen becomes that much higher with this new one then you would have to weigh mitigation/dmg to that HP and decide for yourself. Since IP stacks on itself the extra 25% to IP could easily be argued that it's pointless when you can get the rage for 2 IPs and have a bigger one. Really just all depends on how things end up going.

[Leih]

Yes, Devastator is "bad", and it's supposed to be. It's not there for those who want to maximize their DPS/survivability. It's simply there so people who want to make Protection easier to play have an option. There are people (more than you'd think) who are going to prefer this more passive/easy playstyle. Now, I'll admit that it does suck for the more serious players because it essentially removes a potential talent choice, however, it still is a choice for a lot of other people. If for some reason Blizzard decides to buff Devastator to the point where it's better than the other harder to play options, then that's a big problem

This is of course true and it makes sense that way, it's just that Demon Blades shows it's not impossible that this might end up the best talent in the tree at some point. If this is the best talent at any point for any kind of content, then it really sucks. If it's outright inferior in all cases, then I'm fine with it, I guess, it's just scary having it there after my experience with Demon Hunters.

[kiklion]

You bring up some valid points. My train of thought for comparing them however was that you will take the same dmg regardless. The only thing we are comparing here is the 2 talents so I tried to stay away from other sources. Where is this 25% extra rage on everything coming from though? The HP I completely agree with. It's a valuable talent no doubt but if the rage gen becomes that much higher with this new one then you would have to weigh mitigation/dmg to that HP and decide for yourself. Since IP stacks on itself the extra 25% to IP could easily be argued that it's pointless when you can get the rage for 2 IPs and have a bigger one. Really just all depends on how things end up going.

You will take the same damage, but the rage you get from the damage gets you 25% more IP than the same rage gets you without indomitable. Assuming you only spend rage on IP, you can model Indomitable as 25% extra rage from all sources. This causes Indomitable to be preferential when tank damage is higher as the % of your rage that you get from damage taken goes up while the rage per second from your abilities stays the same.

So rather than...

(Rage/IPCost) * IPHeal vs (Rage/IPCost) * (IPHeal*1.25)

It's just

Rage vs Rage*.25

This simplification falls apart if you use the rage for other means, but we are talking about only using revenge when it is free.

I don't understand what you are trying to say about IP stacking. This comparison boils down exclusively to how much rage you generate since we have no other rage spenders that matter except when they are free (so they aren't technically a 'rage spender' at that point).

Truthfully it gets more complicate than that because Devastator has more free globals to use on free revenges which lower the cost of IP while normal does not, however normal does still want to use revenge because, assuming a 60 rage IP, revenge essentially generates 26 rage (21 rage reduction + 25% extra from indom), however this is an extra global not in use in the current log we were looking at... assuming TClap replaces all of the revenges that did go out so this will lower the number of shield slams that occurred through lowering the number of devastates that go out.

Truthfully, I don't believe it can be gauged with math short of thousands of iterations of simulations since it depends largely on when procs occur.

[Davidsped]

amazing thread dude

[vorpalbladebb]

Can we get a short summary of what is affected by haste?

Is the following still true?


http://www.icy-veins.com/wow/protect...-stat-priority

Default GCD of 1.5 seconds. Seems to be unaffected for prot warriors.

Base weapon aps of the artifact is 2.6.

With 30% haste base aps is 1.82 and the cooldowns are as follows:

Devastator hits every 1.82 seconds. Grants 5 rage. 30% chance to proc SS.
Shield Block cd is 9.1s. Costs 15 rage.
Thunderclap cd is 4.2s. Grants 3 rage. 30% chance to proc SS.
Revenge cd is 2.1s. Costs 30 rage. Free on proc?!
SS cd is 6.3s. Grants 20 rage.
Devastate cd is 1.5s. 30% chance to proc SS.

9.1/1.82 = 5 Devastator procs per Shield block cooldown = 25 rage from auto attacks.

Any rage from damage taken, Thunderclap, SS and the 10 extra rage from Devastator earned in the 9.1s time frame can go into IP if you only use free Revenge procs.

Chance to proc SS from 5 Devastator hits = 1-0.7^5 = 83.193%

Would be interesting to know if autoattacks would allow back-to-back Shield Slams. Could be possible if the auto attack hits within the 1.5gcd and procs SS.

Couple things here. First, starting in legion, haste affects the GCD for Prot, so your 1.5 sec GCD is indeed reduced by your haste.

Second, 2.6 is not the weapons "aps" (attacks per sec) but rather Seconds Per Attack, which is very different.

Third, and this one is important, 30% haste does NOT mean that your cooldown is reduced by 30%, it means that you will fit 30% more attacks within the same timeframe... This is perhaps not intuitive, but the formula for the amount of cooldown reduction is: 1 - (1 / (1+haste)) .. so your 30% haste gives you: 1 - (1 / (1 + .3)) = 23% cooldown reduction. Put another way, your cooldown will be 77% of what it was before.

Think of it this way.. if you had 100% haste, you don't all of a sudden have no cooldown on any of your abilities, rather you get 100% more attacks (so, twice as many attacks) in the same amount of time, meaning each attack's duration (or cooldown) would need to be halved by your haste. So 100% haste gives you 50% cooldown reduction, or: 1 - (1 / (1 + 1))

So, your 2.6sec auto attack with 30% haste becomes 2sec, and your 1.5sec GCD becomes 1.15sec. I believe that there is still a cap on GCD reduction at 50% haste, which would give you a 1sec GCD. Bloodlust can still reduce the GCD beyond that 1sec floor, but your haste from gear can not.

[Deleted]

Thanks a ton! Great info.

[Llarold]

Couple things here. First, starting in legion, haste affects the GCD for Prot, so your 1.5 sec GCD is indeed reduced by your haste.

Second, 2.6 is not the weapons "aps" (attacks per sec) but rather Seconds Per Attack, which is very different.

Third, and this one is important, 30% haste does NOT mean that your cooldown is reduced by 30%, it means that you will fit 30% more attacks within the same timeframe... This is perhaps not intuitive, but the formula for the amount of cooldown reduction is: 1 - (1 / (1+haste)) .. so your 30% haste gives you: 1 - (1 / (1 + .3)) = 23% cooldown reduction. Put another way, your cooldown will be 77% of what it was before.

Think of it this way.. if you had 100% haste, you don't all of a sudden have no cooldown on any of your abilities, rather you get 100% more attacks (so, twice as many attacks) in the same amount of time, meaning each attack's duration (or cooldown) would need to be halved by your haste. So 100% haste gives you 50% cooldown reduction, or: 1 - (1 / (1 + 1))

So, your 2.6sec auto attack with 30% haste becomes 2sec, and your 1.5sec GCD becomes 1.15sec. I believe that there is still a cap on GCD reduction at 50% haste, which would give you a 1sec GCD. Bloodlust can still reduce the GCD beyond that 1sec floor, but your haste from gear can not.

All well said. Haste is not as good as the community assumes at first glance, but it's still very good. Nothing wrong with that.

[mugthrall]

<Thunder Clap> replacing Revenge as an aoe opener, and helps maintain cleave threat on adds which is essential to maintaining a tanking role. However with the 50% movement reduction, especially in mythic+ dungeons where I am required to move adds quickly, and maintain aoe threat will be extremely frustrating.
<Revenge>'s 30 rage cost, only reasonable if off the gcd, otherwise clunky and will not be used in favour of TC
<Shield Block> rage up to 15 rage??? Was this even a problem? Like really guys, rage is hard enough to come by AS IS... whyyyy?
<Best Served Cold> I like the addition of more damage focused talents for those of us bored on farm, BUT the extra rage generation was CRUCIAL in many mythic+ dungeons where I needed a higher up time on IP as we were moving so much i get hit in the back making SB pointless, feels like a nerf to a resource that was already nerfed
<Devastator> in the same row as Indomitable? lol Indomitable is the strongest talent in the entire tree, that's a shame it'll never see play.
<Vengeance> This talent has separated the good from the amazing. It's one of the defining talents of the spec, and should be treated as such. At the present time the lowly 35% rage reduction to IP and Revenge made this more of a dps talent than survivability. It felt meaningful to get right when it was 50%, I believe this # should be reverted to 50% as per the original build, and instead remove the rage cost from Revenge and return Heroic Strike
-On Devastator as a talent. A complete parody of 'Demon Blades' from the DH, it simplifies the spec to a mind-numbing pace of waiting on cool downs that don't proc ever and the feeling dread when your precious rage for IP goes into pitiful 155k Revenge, WITH Best Served Cold against 5 enemies, not worth over Avatar
-On the removal of Focused Rage, I was never a fan of focused rage as a dump as it equated to such minor dps increases, although the fun of chain cycling between FR + IP was engaging, a more rewarding dump was HEROIC STRIKE (rip) which tied into the Vengeance talent to combo with IP


Eliminating the fun play style of hitting core abilities when they proc a reset (used to be Revenge + SS) was great, especially with an OFF THE GCD component. When that button lit up it meant I crit or dodged, it felt amazing. That sensation of quick-time event play style that was enjoyable and tested my ability to maintain Vengeance, was ultimately rewarding. This new system is not.

IN SHORT... Please return heroic strike instead of revenge as rage dump, remove the rage cost for revenge and return it's rage generation, revert TC to the way it was (IT WAS AWESOME) and the extra 5 rage on SB.... really?

This is a step back in game play it feels worse than when shield slam generated no rage, please, reconsider these changes.

[Leih]

<Vengeance> This talent has separated the good from the amazing. It's one of the defining talents of the spec, and should be treated as such. At the present time the lowly 35% rage reduction to IP and Revenge made this more of a dps talent than survivability. It felt meaningful to get right when it was 50%, I believe this # should be reverted to 50% as per the original build, and instead remove the rage cost from Revenge and return Heroic Strike

I mean probably one of the strangest things about all this is that now since Vengeance isn't dependant on choosing Ultimatum, and Revenge resets fairly frequently on dodges and parries, there doesn't seem like a lot of point in ever taking Into the Fray or Booming Voice except, I guess, maybe the ultra niche situation where some raid boss might have no melee attacks?

Vengeance/Ultimatum even after the nerf was already a really powerful combo that kind of defined the playstyle but with Ultimatum gone and Revenge resetting by itself whenever you're actually tanking, Vengeance just seems like a no brainer.

At the very least, without it, Revenge seems like it barely has a purpose and not worth wasting a GCD on from a mitigation standpoint, even when you're capping rage and even when you get the free proc... At least Focused Rage was off the GCD.

Presumably there are yet more changes still to come, because I thought the purpose of all this was to bake in mandatory talents and make talents which are not picked more appealing. This does the opposite.

[Toybox]

UPDATE: Hmm, just put some more time in... It's growing on me.

[Sting]

Anybody else that hates skittish? Doing mythics with a frost dk / fire mage and my threat just ain't cutting it.

[Lysozyme]

You seem to have missed a lot of factors in this. Oh and before I begin, since DH came into play I have hated the passive playstyle aka Devestator type yet Blizz still made it the best talent for DH's. Saying it won't happen here is like saying Blizz listens to the playerbase, which we all know is BS.

First off, your assumption with dev is that it will generate 613 rage. What is that based on? Fight length? Time on target? Your personal haste values? There are a ton of things that can make this number go drastically up or down. So trying to base napkin math off of your personal stats will only lead to false conclusions.

Now if we go by actual fight numbers, let's use the rank 1 mythic parse on Nyth for this: https://www.warcraftlogs.com/reports...&ability=23922 then there were 105 melee attacks done. From those 105 alone, you would be getting 525 rage. Now with a 30% chance to reset shield slam you can safely assume that we would have gotten an extra 31 shield slams in for another 620 rage. For a total of 1145 rage. Now considering that without Vengeance, IP costs 60 rage we would have 19 IPs just from that rage gain alone. That is NOT taking into account vengeance procs for a reduced cost. Now lets say that we used all of these IPs after hitting revenge, like we do now with FR. That cuts the cost of IP to 39 per use and means that we could have gotten 29 IPs out of that same rage gen.

Looking at the same log, let's check out the current system. The fight was 3:56 long. That means without a cd reset on SS he could only have used SS 26 times. Yet he had 45 in the fight. So we can safely assume that the other 19 were resets from Devastate. That gives us an extra 380 rage which means even using the reduced costs, we wouldn't get 10 extra IPs out of it. Seeing an issue yet?

Since this guy was running Indom, his avg IP was 1.2 mill. Using that number and taking out the 25% for losing Indom gives us 900k. So only using the examples of "free" rage here, total IP healing with Indom in the current build would be 10.8 mill. Total IP healing with dev using upcoming build (what we know) would be 26.1 mill. Granted this is not taking into account any rage gained from TC.

Feel free to correct my math on this if I am wrong. But do so with actual facts and numbers, not made up information.

Your maths is wrong because when you reset shield slam you don't just get an additional shield slam, you reset the cd. This means you can use it again immediately. If you reset the cd when the cd was half used up already, you're only really getting half a shield slam. If you reset the cd when you are less than 1 gcd from being able to use shield slam, you get effectively nothing.

Marok didn't "just make stuff up", just because you don't know where the numbers came from doesn't mean the maths is incorrect - he just used an example over a particular fight length.

Your maths is very wrong though. You can take a log and say "this is how many extra shield slams he got from devastate resets, assuming he was using it on cd", but you can't take a log and say "if this had a 30% chance to reset shield slam, he'd have 30% of that amount more shield slams".

Further, you are comparing "only using the examples of 'free' rage here" - why? Even if do get more rage from Devastator than you do without Devastator, and you get more out of Ignore Pain from that extra rage, you get nowhere near as much from Ignore Pain with devastator as you do from Indomitable throughout the entire fight. Which is what actually matters - the totals.

Plus in any given cast of Ignore Pain Indomitable is better because any particular Ignore Pain will be larger. Plus Devastator is slightly more susceptible to rng because you are more liable to a shield slam reset being wasted than if you didn't have it. Plus Devastator is worse for DPS. Plus you don't have an extra 25% HP.