Patch 6.1 – How Do I Protection?

Patch 6.1 arrives today, bringing a number of changes. I’ll try to keep this short (ha!) and just hit the points that matter to us.

Goodbye Glyphs

• Glyph of Alabaster Shield has been removed.
• Glyph of Harsh Words is now only available to Holy Paladins.

If you haven’t been following the prot DPS scene, you may not know that we were doing a lot of damage in early January. We got hit with a series of nerfs to help rein that in, including a nerf to Seraphim (down from 1000 of each stat to 750 for prot), collateral damage from a giant 80% nerf to Seal of Truth (see below), and a nerf to the Alabaster Shield glyph (from 10% per stack to 3%).

Harsh Words, while it wasn’t nerfed, was also a contributing factor. Harsh Word does a fair bit more damage than Shield of the Righteous, so this glyph was a pretty significant DPS gain. Of course, the downside is that you give up a lot of defense by using it. That’s a pretty fair trade-off, but it did lead to a lot of high-DPS parses in early Highmaul that ruffled the feathers of pure DPS classes.

In 6.1, both of these glyphs are off the table for Prot. That said, I don’t think either of these removals were actually targeted at reducing our DPS. High-DPS parses generally rely on stacking of Holy Avenger and Seraphim with potions, trinkets, and a lucky streak of Avenger’s Shield procs (ideally an AoE situation for maximum damage). Both glyphs obviously help, but they’re small effects in a relative sense.

What was the point of these removals, then? Choice. Your average paladin had quite a few glyph options for DPS: Alabaster Shield, Final Wrath, Harsh Word, and situational glyphs like Focused Shield and Double Jeopardy. Since many encounters favor the Glyph of Divine Protection, we really didn’t get a chance to use many other utility glyphs. There was no room for convenience or play style glyphs like Consecrator or Judgment.

In a sense, it’s the same problem that we had with the old talent trees (and arguably, some choices in the new ones) – it gave the illusion of choice rather than a real choice. I think the more important issue is that it made Glyphs feel like a checklist rather than something you could use to customize your play. Removing some of the DPS options helps provide room for more individual choices.

Recharging Empowered Seals

• Empowered Seals (Protection, Retribution) has been improved. Liadrin’s Righteousness now increases haste by 20% (up from 15%). Uther’s Insight now heals the Paladin for 2% of maximum health every 3 seconds (up from 1% every 2 seconds).
• Seal of Truth is no longer available to Protection Paladins. Seal of Command is now replaced by Seal of Righteousness instead of Seal of Truth.

Empowered Seals is getting a buff this patch, since it was lagging significantly behind the other two options. I’ve included the removal of Seal of Truth in this section because Empowered Seals is the real reason Truth is gone.

The initial speculation when Seal of Truth was nerfed by 80% via a hotfix was that it was a nerf targeted at reducing our DPS. But that’s really not the case – the DPS loss was at best collateral damage. The real reason was that it was a precursor to the removal of the spell. They can’t remove a spell from the game via a hotfix; that requires a patch. They can, however, nerf it to the ground so that it’s not attractive.

Empowered Seals suffered from a case of “too much complexity, too little fun” for Protection. Retribution only has to juggle two seals, which is pretty reasonable and led to a play style that many Rets actually enjoy. But juggling three seals was a bit too much, both in terms of buff management and GCD requirements.

The harsher GCD requirement meant you were casting fewer of your rotational spells, of which we already have a pretty full complement. So you were constantly skipping things like Consecration and Holy Wrath to swap seals. A lot of people found that to be less satisfying than hitting an actual damage spell. But regardless of whether you agree with the “fun” argument, there’s a more fundamental problem with the GCD cost of the pre-6.1 version of Empowered Seals.

Spending that many GCDs on seal swaps carried a fairly large opportunity cost, both in terms of damage (obvious) and survivability (harder to refresh SS at optimal times). That meant that the buffs needed to be very strong to compensate for that loss. At the existing tuning, the talent was about right (actually a little bit low) for Ret, but far too weak for Prot given the amount of effort.

There were several ways to fix that, of course. They could make the buffs stronger for Protection, much like they made Seraphim weaker for us. They could also take seals off of the GCD, but that introduces its own issues, because you could write a macro that made the talent essentially passive. They could re-design the talent entirely, but (a) that’s a lot of work, and (b) they like the theme, just not the numbers.

The solution they went with was to axe Seal of Truth for Prot and buff Liadrin’s Righteousness and Uther’s Insight. Since we like haste a lot more than Ret does, the buff to Liadrin’s Righteousness helps the talent much more for us than it does Ret. And obviously, the buff to Uther’s Insight only affects us, since Ret won’t be caught dead using Seal of Insight.

Only having to juggle two seals brings the complexity requirement down to a reasonable level. for both specs. Outside of its effect on Empowered Seals, Seal of Truth’s removal doesn’t affect us very much, since we can use Seal of Righteousness for DPS in pretty much any situation we might’ve wanted Seal of Truth before.

The one downside to the new implementation is that Liadrin’s Righteousness is giving a lot of haste now. In Mythic BRF gear we can easily hit the 50% haste cap with it active, which makes our stat choices a little wonky. For example, haste proc effects become significantly less valuable, and depending on exactly what gear we have, we may end up enchanting something other than haste or using a different food. This means we’ll be dancing around another cap when it comes to our gearing.

In the last blog post I provided a table of the haste-rating targets for different talents. Here’s an updated version of that table accounting for the new version of Empowered Seals (ES):

As you can see, it only takes 1319 haste rating from gear to hit 50% haste with Empowered Seals and the 5% haste raid buff. That’s not a difficult threshold to hit, and may even be attainable with an ideal set of heroic BRF gear.

In any event, we should at least briefly address how Empowered Seals performs after the buffs. First, let’s note that there are really three ways to play with Empowered Seals now:

1. Twist Insight and Righteousness, trying to keep both buffs up over 90% of the time.
2. Just use Seal of Insight
3. Just use Seal of Righteousness

I’ve run some quick simulations to show you how those different strategies affect your DPS output and survival with all of the different level 75 talent options:

Empowered Seals – DPS Rankings

Empowered Seals – TMI Rankings

Sitting in Seal of Righteousness is clearly the best choice for DPS, netting you around 2.5k-3k additional DPS. At the same time, it’s still a noticeable loss of survival, so it may not be a great idea for progression. At the other end of the spectrum, sitting in SoI is only a minor improvement over twisting, but comes with a significant DPS loss. Twisting is clearly the best strategy for trying to optimize both DPS and Survival, which is probably good since that’s what the talent is trying to encourage.

The next question, of course, is how this compares to the other talents. So let’s see how that comparison shakes out. I’ll include all of the regular setups, as well as a few setups where we sit in Seal of Righteousness to attain higher DPS.

DPS rankings for all L100+L75 talent combos, plus some SoR ES cases.

TMI rankings for all L100+L75 talent combos, plus some SoR ES cases.

These plots take a little more analysis than the previous ones. At first glance, you might conclude that Empowered Seals is still terrible. First of all, in the survival arena, it’s pretty competitive with HS and Seraphim. HS has a slight edge thanks to set bonuses, and Sera lags slightly, but it’s a small enough difference in each direction that any of the three talents is a reasonable choice. In practice, you’d probably choose the talent based on encounter constraints like tank swaps and the cooldown on dangerous abilities. Swapping to Seal of Righteousness hurts all of the setups, but appears to hurt Empowered Seals the least.

On the DPS side, Empowered Seals seems to be lagging the others by a fair bit. However, this sim over-values Holy Shield in two ways: the actors have 100% active time with the boss, and there’s a ticking DoT that can trigger it every two seconds.

If there’s an encounter where you’re tanking the boss continuously from beginning to end, then these results are fairly representative (they should, for example, be a fairly good indication of how Mythic Butcher would work in T17M BRF gear). They’ll still over-value Holy Shield slightly though thanks to the DoT.

If you have tank swaps, and assume you only have the boss’ attention 50% of the time, Holy Shield loses about half of its procs. That’s a loss of around 2k DPS, which would drop it back into the same range as Empowered Seals. Factor in the inflation due to pocs from DoT ticks and it will come out a little lower still.

In practice, that means Seraphim has a solid lead in DPS for the average encounter. However, if you’re going all-out DPS and using Seal of Righteousness, Empowered Seals is not far behind at all. And as we discussed earlier, that DPS comes with a smaller cost in survivability. The other advantage is that we have more control over when we go into DPS mode and when we go into turtle mode, unlike Seraphim’s fixed 30-second cycle.

Holy Shield brings up the rear for DPS, but has the best survival under most circumstances thanks to the fairly powerful T17 set bonuses (more on that below). It’s still a solid contender though, and will remain strong in fights where you’re tanking full time and/or you can abuse the spell and DoT blocking mechanics.

This change has really given us three solid options in the L100 talent tier. They all have strengths and weaknesses, but none are so far ahead or so far behind as to be irrelevant. I think Seraphim will still remain the preferred choice at the top tier of Mythic raiders, while Holy Shield will still be the preferred choice for the average player, but the gap has narrowed enough that I think all three talents will see use going forward.

Set Bonuses

I haven’t had time to post since Blackrock Foundry came out, so we should probably talk about set bonuses. First of all, the set bonuses are good, so you’ll want to pick up set items. Unlike previous tiers, set items can be warforged or socketed, so there’s no inherent disadvantage to using 4-piece on paper. In fact, the only remaining opportunity cost in theory is the itemization of the pieces themselves.

Of course, in practice you’ll still have to decide between the items you have on hand. So you may still find yourself deciding between a well-itemized warforged/socket off-set piece and a tier piece with neither of those features. Likewise, it’s hard to nail down which four pieces you should aim for because it depends very heavily on what items you have. Legs are the weakest tier piece on paper, but if you have a warforged/socketed off-set option in one of the other four slots then you may be better off using the tier legs to take advantage of that piece.

First, let’s look at how valuable the bonuses are. In this sim, I compare an actor with both set bonuses (DP_HS_4p) to one where the 4-piece is artificially disabled (DP_HS_2p) and one where both set bonuses are artificially disabled (DP_HS_0p). I repeat for an actor with the HA_Sera talent combination to see how it affects Seraphim.

DPS rankings for set bonuses

TMI rankings for set bonuses.

Neither set bonus really affects your DPS with Seraphim, but Holy Shield obviously benefits from the extra block granted by Faith Barricade. The four piece bonus (Defender of the Light) doesn’t directly affect DPS at all. The small discrepancy in the DP_HS actor appears to be a combination of statistical fluctuation and a ~1% reduction in Mark of Blackrock uptime caused by larger blocks.

On the other hand, both set bonuses are significant for survival regardless of talent choice. An improvement of 2k TMI will be better than swapping to two equal-ilvl items with better secondary stats, so there’s no reason not to grab the set bonuses.

The next question is whether it’s worth downgrading to pick up those set bonuses. To test that, I took a set of ilvl-685 Highmaul gear and compared it to sets where I swapped two or four ilvl-680 heroic set items in to activate the set bonuses. Here are the results:

DPS rankings for gear sets that trade ilvl for set bonuses.

DPS ranking for gear sets that trade ilvl for set bonuses.

The set bonuses are strong enough to make up for the loss of 5 ilvls in four slots if you’re using Holy Shield, so grabbing four pieces of heroic tier gear is a no-brainer. With Seraphim the story is a little different. Downgrading to pick up the two-piece improves survivability, although the improvement in DPS is minimal. The four-piece bonus doesn’t quite make up for the ilvl loss though, and comes with a heftier DPS cost. So it’s worth downgrading to get the 2-piece bonus if you run Seraphim, but I would wait to pick up the 4-piece until you get at least a sidegrade in base item power, like a heroic warforged or socketed tier piece.

Stat Weights

One question that’s come up a lot is how Mythic BRF gear affects stat weights. I set up a sim that uses the T17M gear set with a few different talent combinations to investigate that question.

The first actor uses Divine Purpose and Holy Shield, and produces a TMI scaling plot that looks like this:

TMI scaling plot for Divine Purpose + Holy Shield.

Remember that on one of these plots, the steeper (i.e. more negative) the slope, the better that stat is at increasing your survival. Bonus armor is clearly still king, with haste and mastery roughly tied in effectiveness. Mastery’s getting a bump from the set bonus, which may even make it slightly preferred over haste. Versatility, crit, and multistrike trail the other stats in effectiveness.

So with Holy Shield, you want

Bonus Armor > Haste / Mastery > Versatility > Multistrike > Crit

If we instead look at Seraphim, we get a very different story:

TMI scaling plot for Holy Avenger + Seraphim.

Haste’s all over the map here thanks to the myriad of haste breakpoints Seraphim creates. In practice, I think we should be averaging over those effects to account for latency and player error. That would put haste somewhere between bonus armor and the pack of other stats. The set bonus isn’t helping mastery as much in this talent configuration, so it sits at the back of the pack with, of all things, crit. I’m not sure why crit has lost its dominance with Seraphim here, but that may be due to the choice of Holy Avenger as a L90 talent. Crit may fare better when you pair Seraphim with Sanctified Wrath or Divine Purpose. Tough to say, but it’s something I’ll be investigating in the next few weeks as I have time.

For now, with Seraphim I think we want to stick with the stat priority

Bonus Armor > Haste > Versatility > Crit / Multistrike / Mastery

Empowered Seals provides a completely different scenario:

TMI scaling plot with Divine Purpose and Empowered Seals.

Haste performs incredibly poorly with Empowered Seals because we’re bumping up against the 50% haste cap. As you remove haste (going to the left on the plot), you can see haste’s value increase, but if you add haste above the cap (going to the right on the plot) it falls back in line with multistrike and crit. The standout winner here is mastery though, which seems to dominate when Empowered Seals is selected. Presumably the 91% uptime on Shield of the Righteous we get with this much haste is inflating the value of mastery’s contribution to SotR mitigation.

So with Empowered Seals, the stat priority is

Bonus Armor > Mastery > Haste (to 50%) > Versatility > Crit / Multistrike / Haste

Note that “to 50%” is after all raid buffs and Liadrin’s Righteousness. If you want a target to hit on the character sheet, you want to aim for 1719 rating without raid buffs, or around 19.1%.

Handier Purity

•  Hand of Purity now reduces damage taken by 15% (up from 10%).

A small buff to Hand of Purity makes it a little more effective against normal (non-DoT) damage. It also makes it a little more competitive with our default choice, Unbreakable Spirit. Thinking solely of Divine Protection and Hand of Purity for the moment, choosing Unbreakable Spirit gives you 16 seconds of 20% mitigation (or 40% against magic) every minute, while choosing Hand of Purity gives you 8 seconds of 20%(or 40%) mitigation and 12 seconds of 15% mitigation. Hand of Purity actually lets you cover more seconds of combat with a cooldown, albeit weaker ones. You could definitely imagine taking it to be able to chain cooldowns through a dangerous phase in an encounter, even if you can’t make use of Purity’s 80% mitigation against DoT damage.

That said, I think the fringe benefits of Unbreakable Spirit are hard to pass up. A 5-minute Lay on Hands and a 2.5-minute Divine Shield are really nice, and give you a lot of flexibility in dealing with encounter mechanics. That’s arguably stronger than 4 extra seconds of (weak) cooldown coverage every minute.

Still, I think having more options is better in this case, so a buff to make Purity a more viable competitor to Unbreakable Spirit is certainly welcome. And at any rate, it’s a flat-out buff any time  you receive it from a Ret or Holy paladin as an external cooldown.

Pardon Me, Governor

• Execution Sentence healing dealt by Stay of Execution has increased by 100%.

This buff may be targeted at Holy more than Prot. From our perspective, it looks like a buff that tries to make Execution Sentence a more interesting choice as a survival cooldown. The massive up-front healing from the first tick of Stay of Execution would quickly stabilize your health (if not top you off entirely), and the declining HoT gives you a few additional seconds of safety.

The real downside of using it this way is the sheer DPS cost. Regardless of which level 90 talent you use, they’re all a large portion of your DPS – 5% or more, in general. That’s a lot of DPS to give up to use Stay of Execution as a survival cooldown, especially when we already have serviceable alternatives in Lay on Hands and Word of Glory.

I’m sort of on the fence about this one. A 1-minute sort-of-LoH with a HoT sounds nice, but I’m skeptical that it’s worth the sheer DPS cost. I’ll have to play with it some to see how it feels, since this is the sort of thing that’s hard to quantify in a sim. Still, no sense complaining about a buff.

Multistrike – Not a Buff

You may have noticed that Shining Protector‘s tooltip changed slightly in patch 6.1. It now reads:

All heals you receive, including multistrikes, have two chances equal to your multistrike chance to trigger Shining Protector, healing you for an additional 30% of the triggering heal.

Before the patch, it said “a chance” rather than “two chances.” This may appear to be a 100% buff to the passive, but it isn’t. The spell has been making two rolls since Warlords was released, which is easily confirmed by checking logs:

Log showing Shining Protector giving two procs from a single heal.

In this example, which was taken some time during Highmaul progression, I received two Shining Protector procs from the initial Lifebloom tick and one proc from the Lifebloom tick’s multistrike. In fact, you could see up to nine healing events from a single heal if you got lucky: the base heal, two multistrikes, and two shining protector procs from each of those three events.

I was initially modeling this incorrectly in Simulationcraft based on the tooltip wording and testing I had performed early on in beta (at which point, it was only making one multistrike roll). When I noticed this, I reported the disconnect between tooltip and behavior and received confirmation that it was working as intended, which is what led to the tooltip update we see today. And of course, I’ve since updated SimC to handle the mechanic properly, so all of the sims in this post are modeling it correctly (hence why multistrike is performing a little better than before).

So, in short – the Shining Protector change is a tooltip update and not a buff.

WeakAura Strings

I haven’t had time to make and upload a video, but I’ve been updating my WeakAura strings throughout the last few months. The base package contains auras to help you manage Seraphim, and I’ve added a separate group to help manage Empowered Seals. I’ve also added armor potions and healing tonics to the cooldowns group.

Fire Away

As usual, if you have any questions feel free to drop them in the comments. I’ll try to follow up with and answer as many as I can throughout the day on Tuesday.