Seraphim as a defensive CD

Posted this on Blizzard's forums. Would love to get some feedback from the maintankadin community.

Ran some tests with seraphim on dummies and in dungeons

I believe seraphim itself can be used as a form of active mitigation, much like divine protection, ardent defender, and shield of the rigteous. Defensively, seraphim increases armor, versatility, multistrike, parry (crit), and mastery. The armor, versatility, and mastery alone reduces physical and magic damage taken, though less so magical than physical. Multistrikes improves the healing intake as mentioned in the above.

Here is the new rotation I am currently testing:

Divine Protection ---> Seraphim ---> 2-3 charges of shield of the righteous ---> Divine Protection ---> Seraphim

Here are the logged results on a dungeon dummy:
Damage taken value without any active mitigation; the dummy averages around 30k-38k a swing.
http://www.warcraftlogs.com/reports/txJ ... iew=events

Damage taken with blanketed active mitigation:
http://www.warcraftlogs.com/reports/6MQ ... iew=events

Results:
Damage taken during seraphim active was the equivalent to having Diving Protection up. However, the chance to parry, block, and multistrike was significantly higher, making seraphim stronger than divine protection alone as a defensive CD.
I tested this rotation in several dungeons and always ended up being the top DPS and healer.

Conclusions:
Even at 9% baseline haste, I was able to keep this rotation going smoothly with an additional shield of the righteous in between seraphim for added dps. Additionally, I consistently found multistrike to be higher in dps and defensive value over haste. Multistrike may be the strongest dps/defensive stat for prot paladins at lower gear levels with this rotation. I believe it may outperform haste.

I'm not as astute with simulators as some others, but I'd love to see simulations done on the benefits of multistrike over haste. I haven't tested anything out mathematically, however I can see how multistrike may have a stronger dropoff point in rate of increased benefit than haste by virtue of what it does. The current simc models for beta paladins doesn't seem to take into account the potential for outside incoming heals to proc Shining Protector, which skews the results of calculated benefit from multistrikes.

I know Theck is doing some testing left and right, but for the more thorough tests, it might be better to wait to close to 6.0, as they're tinkering with the numbers a lot, right now. Unless of course you want to tell Blizzard something in time...

And yeah, between Seraphim and Empowered Seals, we got quite a few rotations that could be tested.

The problem with Multistrike is that it increases the healing done by other healers, that aren't present in SimC, so it's gonna be harder to test.

the sims I did with Seraphim showed that Seraphim is a survivability loss because of the SoTR uptime you lose for spending 5 Holypower. but a huge DPS increase. so it is more of a trade off.

have to see tho after we know more about the tanksquish.

as stamina/armor/bonusarmor and mastery should lose some value here

But SotR should as well, so it might even out...

âs I said, my sims show it is a huge survivability loss. you can check them in the simcraft thread.

Yes, but I guess the question would be whether your sim is using seraphim as I've detailed in the above. From what I've tested, seraphim is basically the nearly the equivalent to divine protection. So the question would be, is your sim chaining CDs without overlapping them and treating seraphim as a seaparate CD? Does your sim account for the multistrike proc increased rate during seraphim while being healed externally? In order to do this, your sim must specify that there is steady source of healing externally, taking into account multistrike passive procs. These are critical variables to consider before disregarding seraphim as a strong defensive CD. Moreover, does the TMI shift when multistrike is being prioritized?

SotR may be useful against physical damage but does nothing for magic damage, which seraphim does account for in the form of shining protector, versatility, multistrike procs on sacred shield, and multistrike procs on seal of insight.

Moreover, your sim may show a huge survivability loss, but I can tell you now that every healer that has healed me against a raid tank dummy or mythic tank dummy told me that it felt way easier to heal me while I was using seraphim in the way that I've detailed in the original post. They couldn't heal through the damage while I was using empowered seals.

I have a feeling that your sim may not be fully taking into account what multistrike does for prot pallies on beta.

The Thing is. We are siming for TMI.

if we consider having a healer in the sims we are basically screwing the whole thing :) also it might eb dificult to implement, because what exactly will the healer do? use the complete sim model of, let's say the holy paladin? is there even such a thing? :) healers are pretty hard to model as healing is more an art than science. the only thing you can go for healers is theoretical effectiveness, but still you might have to adjust depending on your raid size, raid skill, taktiks, Boss, mission and so on....

it will get utterly complex and to a point where the results you get out of it won't represent anything because you just have to many variables.

also yes only physical Meleehits are really what its all about. Magical attacks are soaked with HP and hard Cooldwons that you time for. this is playstyle and has nothing to do with siming or TMI, but defines the skill of the player.

also not every boss has magical attacks, also there are very few magical bosses. in fact in MoP the only pure amgical Boss was Lei Shi.

and yes I agree for a lei shi style Boss seraphim might be a golden Defensive Cooldown. but this is an exception, not the rule. ^^

in example, if you have a magical Boss, even today in MoP, you screw the whole haste thing and go full stamina.

in my sim seraphim is used on cooldwon when it is ready and 5 Holy power are gained.

like you said it is a huge DPS gain, but overall a TMI loss. even compared to not using a lvl 100 talent.

being top in DPS an HPS has nothing to do with TMI either. you want to reduce spike damage, and even the loss of 2 SoTRs for Seraphim screw you in that regard.

although it might be may be interesting to test a steady fixed healflow (like heal for X every Y seconds) with a definable multistrike rate while fighting a pure melee boss. I'll have to go and see how I'd define something like this.

BUT before this get's to far. Let's wait for the announced tanksquish. our secondary stats should become much more valuable. and EH slightly less so I guess.

also the reworking of our hard CDs like DP and GoAk might have an influence. the rescaling of mastery for our Active mitigation will for sure.

so let's wait until then, and retest / resim again ;-)

It's still hugely misleading to say that using Seraphim over other talents is a gigantic TMI loss without the warning of specifying that it does not include externally healed multistrike procs. Moreover, I'd still like to know how you're utilizing your CDs in your sims. This may even call for the need of a new survivability index tailor-made for prot pallies.

like you said it is a huge DPS gain, but overall a TMI loss. even compared to not using a lvl 100 talent.

Maybe I wasn't clear in the way I worded it, but no I'm not saying that Seraphim is a major TMI loss. I'm arguing that it may actually be comparable to the other talents and your sims may not utilizing Seraphim correctly.

if we consider having a healer in the sims we are basically screwing the whole thing :) also it might eb dificult to implement, because what exactly will the healer do? use the complete sim model of, let's say the holy paladin? is there even such a thing? :) healers are pretty hard to model as healing is more an art than science. the only thing you can go for healers is theoretical effectiveness, but still you might have to adjust depending on your raid size, raid skill, taktiks, Boss, mission and so on....

Yes, TMI explicitly omits external healing. This is the reason why there may need to be a new survivability index for prot paladins due to shining protector, which is a significant influence on our survivability, ESPECIALLY since seraphim increases our multistrikes by a huge amount.

it will get utterly complex and to a point where the results you get out of it won't represent anything because you just have to many variables.

Yes, this is my point. There may be a need for a new measure for prot paladin survivability.

being top in DPS an HPS has nothing to do with TMI either. you want to reduce spike damage, and even the loss of 2 SoTRs for Seraphim screw you in that regard.

I think you're missing my point entirely. And self healing does influence TMI.

in my sim seraphim is used on cooldwon when it is ready and 5 Holy power are gained.

This is what I'm primarily interested in. Is ardent defender or divine protection overriding when seraphim is being used or are all of your cooldowns being used adjacently in a manner that would always blanket you with a cd? Is SotR being spammed whenever it's available or is your sim actually saving Holy Power for when seraphim falls off to followup with 2 charges of SotR?

Please revisit the following log site to see what I mean by this: http://www.warcraftlogs.com/reports/6MQ ... end=156304

I guess Theck already covered external heals being included: http://www.sacredduty.net/theck-meloree ... -document/

ETMI may potentially be the most accurate measure for our effectiveness as a tank. TMI may no longer be a useful measure unless we're planning on going into encounters without getting healed.

Due to the nature of our multistrikes passive http://wod.wowhead.com/spell=159374/shining-protector , TMI may be an outdated model of our tanking effectiveness. TMI omits external healing. A significant part of the WoD prot pally's tanking effectiveness is contributed to external healing proccing our multistrikes passive. ETMI may be the more accurate measure starting WoD.

UPDATE
http://us.battle.net/wow/en/forum/topic/14058407204

With the upcoming tank squish, Seraphim may be more powerful than ever. Total armor is being reduced on plate and shields, making each added bonus armor a gigantic increase in survivability. FURTHERMORE, resolve is being nerfed even further, making multistrikes HUUUUUUUUUUUUUGE for our survivability.

I will be going even far as to say that every prot paladin will probably need to prioritize multistrike then haste or mastery—most likely haste.

I think you read that incorrectly, Resolve is getting massively buffed.

In any event, it's far too early to say whether multistrike will be our best secondary stat. One of the reasons TMI excludes external healing is that, generally speaking, when someone is actually healing you you don't die.

TMI measures those 6-second windows when you're on your own because your healer is dead, disconnected, running out of fire, playing Bejeweled, or doing whatever else happens to be distracting them at the time. And in a period where nobody else is directly healing you, multistrike's value will be primarily due to its effect on your own self-healing.

I did misread the changes to resolve. Thanks for the correction. I wrongly assumed that resolve would also be squished when they listed resolve underneath the large list of incoming squishes. I guess that's even better news for anything that has the ability to absorb or heal. And as you said, we still do have to see what happens, though, with this tank squish.

I still do believe, however, that squishing every other CD in the game makes seraphim even more attractive.

These are all speculations, of course, with how things have been changing thus far.

I'm not disagreeing with your speculation. It obviously depends on the details, but a tank squish certainly has the potential to make Seraphim stronger relative to the other talents. That said, keep in mind that they could also nerf Seraphim...

I also thing multistrike may end being a strong stat if the damage and healing model is heavily triage/throughput-based rather than spike-based. But again, we'll have to wait and see whether that's the case or not.

The tank squish notes are out. Post tank squish, I strongly believe that prot paladins are now the best tank in the game.

http://www.reddit.com/r/wow/comments/2f ... _state_of/

I keep an open mind. And might test the Multistrike build tonight on Hans & Franz mythic.

Not sure if those nerfs make Paladins shine or not, judging from merely the tooltips. Sure, DKs, Druids and Warriors lost over 20% of the Armor they had, while Paladins lost less than 10%, but the other changes to AM are hard to compare to that.

Our base SotR went down from 8% to 6%. How does that affect scaling? Does that mean that it's just 2 lower than it was before regardless of total mitigation (say 46% total instead of 48% total)? If so, we don't really care. Alternatively, if it means that it'll be 25% lower than it was before (say 36% total instead of 48% total), then we care a lot.

I don't really know how much % my SOTR took away before on the PTR. but if I remember correctly it is the second option.

I had sth in the 40ies Percentage wise and if I remember correctly it was only sth it he 30s yesterday.

but I didn't play on that character to long, as the levelservers had trouble and I wanted to try out multistrike, so I used my premade PVE character for raidtesting yesterday.

The overall tank squish effects and reduced AM makes Seraphim a major defensive (and offensive) cooldown relative to all other defensive CDs. Seraphim becomes that much more of a necessary CD in the AM blanketing rotation (if this is your playstyle, that is). Also, our shining protector ability did not get nerfed though SotR did. In other words, the benefits of haste and mastery were significantly reduced (more so haste than mastery).

The benefits of multistrike remain in tact. Resolve has been buffed, which also means sacred shield and overall healing in a raid environment have been buffed. All healing abilities have the capacity to multistrike, including sacred shield, shining protector, seal of insight procs, WoG, and lay on hands. This also means that multistrike scales with resolve. I don't think haste can compete anymore, though there will be a baseline requirement of haste to keep this rotation going:

Divine Protection ---> Seraphim ---> 2-3 charges of shield of the righteous ---> Divine Protection ---> Seraphim

http://www.warcraftlogs.com/reports/6MQ ... iew=events

TLDR: Everyone got nerfed. Seraphim and Shining Protector did not get nerfed, and they're essential for our tanking functionality (in my opinion). Paladins barely got nerfed compared to the other classes in my opinion. I don't believe our tanking functionality will center around SotR anymore. I believe the prot pally paradigm has shifted towards multistrike healing and seraphim. SotR is shifting towards being more of a filler like divine protection.

All this is to say that multistrike has huge theoretical potential. I have a feeling that multistrike also outperform haste in the DPS department at lower levels, though I haven't been able to sim it. My simc keeps showing DTPS in the results even though I select scale over DPS. The results felt consistent with my speculation of the multistrike build while raiding with my guild on mythic encounters, but I was more focused on trying to figure out the encounters rather than analyzing the most optimal spec for prot pallying. My perception could have been biased.

If I were to put a stat list priority for the first tier gear to focus on a decent mix of DPS and tanking effectiveness, I'd probably go with:

1) Enough haste to achieve the Seraphim AM rotation
2) Bonus Armor
3) Multistrike
4) Versatility
5) Crit
6) Excess haste
7) Mastery

If someone is able to come up with sims of our DPS scale factors, that'd be greatly appreciated!

Thels wrote:Our base SotR went down from 8% to 6%. How does that affect scaling? Does that mean that it's just 2 lower than it was before regardless of total mitigation (say 46% total instead of 48% total)? If so, we don't really care. Alternatively, if it means that it'll be 25% lower than it was before (say 36% total instead of 48% total), then we care a lot.

SotR was 25% + mastery%, which is why it said "8%" before (that's the base amount of mastery you get from your mastery spec passive).

It is now 20% + 0.75*mastery%, hence why our mastery spell reads "6%" now.

Also, I doubt this is the end of the changes. Resolve was buffed, which was supposed to go hand in hand with healing nerfs. I suspect we'll see WoG, SS, SoI, etc. nerfed soon.

Danielspally wrote:I don't believe our tanking functionality will center around SotR anymore.

If SotR is no longer important, then something's screwed up. AM is supposed to be important, else our rotation is no longer important. It's certainly not for keeping threat, with our 10x threat modifier...

theckhd wrote:
It is now 20% + 0.75*mastery%, hence why our mastery spell reads "6%" now.

where dot he 75% come from? what do I miss.

what I understood form the dataming:

SoTR live: 25%+8%+mastery%(gear)

SoTR squish: 20%+6%+mastery%(gear)

what I understand you are saying, if a paladin has 20% mastery from gear:

Live: 25%+8%+20%=53% reduction from SoTR
Squish: 20%+(6%+20%)*0.75 = 39.5% reduction from SoTR

correct? if so where do the 0.75% come from? :P

Schroom wrote:where do the 0.75% come from? :P

The number displayed on the character sheet didn't change, it still reflects the 8% base; the 0.75 multiplier adjusts that number to reflect the Mastery contribution to SotR.

Using one of my level 90 beta copies as an example:
The character sheet displays 31.70% Mastery (with Might), 75% of that is 23.77%. Mouse over Mastery and the tooltip indicates that SotR is increased by 23% (which we know Blizzard has a habit of flooring tooltips instead of rounding them, so this is expected).

To give an idea how the much SotR was affected by the change: on the Live servers (again, with Might) my SotR reduces physical damage by 57%, post-squish it's 44%. Take that with a grain of salt, as reforging (and the impending removal) has a slight effect on the initial number and the resulting difference.

ah ok 8*0,75=6 now I get it ^^ I was a bit confused as to how the relation of 6 and 0.75 comes from :) thanks.


so old formula: 25%+8%+mastery%
new formula: 20%+(8%+mastery%)*0,75

The 0.75% is in the spell data:


Each of those effects is part of our mastery. For example, effect #1 adds 1.0000 * our mastery to our block%. Effect #2 isn't used anymore, I think (hence why it's named "dummy" now). Effect #3 adds 0.75 * our mastery to Bastion of Glory, effect #4 adds -0.75 * our mastery to SotR's mitigation (which is stored as a negative, i.e. the base is -0.2 because the total mitigation is (1 - 0.2 - 0.75 * mastery)... don't ask).