5.4 PTR: Now See What You Made Me Do? (Part 3)

A big update this time as someone pointed out in the comments of the last post.

The biggest change we were all expecting to see some form of was a change to Eternal Flame and here it is:

Mastery: Illuminated Healing no longer activates from periodic healing effects. (Source)

This is a pretty strong nerf because I believe people were thinking maybe they’ll take away part of it for the most part but they decided in fact to do away with all of it.  Ok, lets deal with this later.  Here are some comments about the recent changes to holy paladins which I think was a good move on Blizzard’s part to come out and have someone actually say something.

Intent behind the Eternal Flame change
We’ve made some changes to Eternal Flame because we feel the talent was being overrepresented within the Holy Paladin population.
Our intent for was for Holy Paladins to pick Selfless Healer if they like to Judge, Eternal Flame if they like heal-over-time spells, and Sacred Shield if they liked damage absorb shields. As it worked out, most every Holy paladin chose Eternal Flame, and we don’t want every Holy paladin to be a heal-over-time machine. (That’s a niche best served by Restoration Druids and Holy Priests.)
We don’t think Holy Paladin’s healing throughput is too high overall (though we’ll gather more information from testing on the 5.4 PTR), so our intent is not to nerf but to reduce reliance on Eternal Flame. One of the biggest problem with Eternal Flame is the way it interacts with the Illuminated Healing mastery, which turns the heal-over-time spell into a potent shield as well. If needed, we will compensate by buffing healing elsewhere.
Our plans for now, are to revert the nerf that was done to the initial healing component of Eternal Flame, but we are unlikely to change the Illuminated Healing back to the way it was.

Sacred Shield
Regarding Sacred Shield, we tried giving the talent as a baseline ability to Protection, but we’re unhappy with that experiment and are likely to revert the change. We agree that the current (old) version of Sacred Shield is more attractive to Holy than the (new) version. We also would likely have to nerf Protection to compensate for getting Holy Shield in addition to another talent. Overall, we think the current (old) version of Sacred Shield is a better design. Perhaps we can still make Sacred Shield more attractive for Holy and Retribution, and make the other two talents (Selfless Healer, Eternal Flame) more attractive to Protection.

Our goal
We understand that Selfless Healer requires a certain playstyle that not every Holy paladin will find attractive. So we want to make Eternal Flame and Sacred Shield feel like viable talent choices as well.
Mastery not benefiting from Execution Sentence or Light’s Hammer isn’t intended and probably just got caught in the change to make Eternal Flame’s heal-over-time not benefit with mastery. We will correct that.
Remember, at this early stage in the PTR, our designers’ focus are on implementation changes (mechanics) first, and tuning changes (numbers) later. Your feedback at this stage will be more helpful to be focusing on the mechanics (e.g. “I liked using hots” or “I like to Judge and wish I could use Selfless Healer more”) and less on “OMG 30% nerf.” (Source)


  • Nerf to EF initial heal likely to be rolled back
  • Nerf to Mastery affected HoTs likely to be permanent
  • Sacred Shield v2 likely to be rolled back with a new design in place for all talents of the tier
  • Mastery from Execution Sentence and Light’s Hammer is bug and will be fixed

Also good to know is the following tweets from Ghostcrawler.

Two examples: We will compensate hunters for Readiness and paladins for Eternal Flame, should those changes stick beyond a PTR build or so. (Source)

We appreciate that SH will just never work for some Holy players, but we think we can get SS to compete with EF. (Source)

Well I guess we all knew the change was coming but were more interested in what would make up for this change.  As you probably know by now, if your top heal on World of Logs is not Illuminated Healing then you’re doing something wrong.  To know how much of this illuminated healing is from Eternal Flame is not a straightforward thing since the World of Logs report lumps it all together with no relationship to what spell it came from.  For example in the picture below the percentage is 58.8% by we don’t know what part of it is from the direct heal and what part of it is from the HoT. However lets take a stab at it.

Dissecting Illuminated Healing
Let me outline the things we want to find out:

  • For Eternal Flame:
    • How much effective healing is due to the initial heal? How much from HoT?
    • How much overhealing is due to initial heal? How much from HoT?
  • For Illuminated Healing:
    • How much of the illuminated healing is from EF direct heal? From HoT?
    • How much will the nerf to IH for HoTs impact illuminated healing and overall healing?

Remember these values will be different fight to fight so I’m gonna focus on how to obtain these.  For most of this, there’s nothing more we can do than look at the expression browser and port the results into Excel and make sure to paste everything as plain text rather than simple Ctrl+V.  I have provided here two R scripts which will answer the questions presented.  Two file inputs are necessary for this to work.  Extract from World of Log’s Expression Editor with the following queries:

File 1: spell = “Eternal Flame” and fullType = SPELL_HEAL

File2: spell = “Eternal Flame” and fullType = SPELL_PERIODIC_HEAL

File 3 (make sure you fill in your screen name in the red portion surrounded by quotes): overheal > 0 and sourceName = “YOUR NAME HERE” 

The first two files is to dissect eternal flame heals for point 1 and the third is to dissect overheals for point 2.

The results are shown below with the original log displayed.  Any fight log can be used here and the main reason why you might get different answers when checking against the original log is that you haven’t recorded the data from World of Logs correctly.  If there’s any interest in using these programs for your own make some comment and I’ll send the file to you.  Remember I do everything in R 😛


Some ways we can check whether we pulled out the right data is  to add up the total amount of effective healing and see if it matches with the World of Logs report.  Here we have from just Eternal Flame a total of 2359879+2032451=4392330 which matches with what was reported on the page.  Lets go ahead then with the assumption that I haven’t done anything wrong and look at the results.

For decomposing the overhealing, we can see that a majority of the overhealing is from Eternal Flame.  This we kind of already suspected and could have easily been calculated from just looking at the report page.  From the expression editor however we can see the spell even more decomposed.  Around 50% of the original direct heal (the WoG component is overhealing).  Remember I said last post that a majority of the overhealing from this spell comes from the HoT?  Here you can see it.  A whopping 67% of the periodic heal of this spell goes to overhealing.    We do however see that in terms of effective healing, the HoT component of this spell is less powerful than the direct heal portion ~2M vs 2.3 M.

So with a mastery percentage of 34% (for me) before the nerf of this spell, we were getting a total of 34% * All Eternal Flame Healing (overheal,HoT) which equals to 10663796 (which we could’ve told from previously.  This gives us 3,625,691 contributing to the illuminated healing total in this log of 9,238,853 before 5.4.

After 5.4, since we won’t be getting IH from the HoT portion, we will only be getting 1,493,392 for illuminated healing.  This reduction of 2,132,299 is overall a 9.4% reduction in overall healing for the whole fight.

The fight in this log is from Heroic Ji-Kun which is not an extremely damage heavy fight.  We can take a look a few more heroic fights to see what the reduction is on healing for those.  The following is a summary table of both tables above for all of the heroic fights my guild’s completed.


The table above shows an example of each of the heroic raid bosses that my guild has done.  The first two columns shows the breakdown of the direct heal portion of EF into effective healing and overhealing.  As you can see, apart from Tortos the distribution between effective and over usually falls in two groups one where the values are split about 50/50 and the other specifically on Council of Elders and Primordius where the value are split 60/40.  Both of these fights are relatively healing intense throughout the fight so it’s therefore reasonable to say that on heavy progression fights, the direct healing portion of the fight comes more in handy.  This also just makes more logical sense.  Other fights which have distinct periods of heavy damage and farm fights don’t have as high effective healing on the direct heal. Good thing the nerf on this was rolled back.

What about the HoT part?  Once again apart from Tortos, there are specific fights where the ratio of effective healing for the spell is higher such as Primordius and Megaera both of which are heavy AOE damage fights.  Other fights especially those on farm, the amount of effective healing actually from the HoT goes down.  I think it’s fair to say fights with a lot of AOE raid damage will see higher use of effective healing from the HoT component of EF.  We can expect this trend to continue and in 5.4 when we may actually be switching between Tier 3 talents, it will be good to keep this in mind.

Lets take a look at the second table which shows how much healing we will lose with the nerf to IH benefiting HoTs.


There’s a lot of information here so bear with me.  The first column on the right “Total Healing” represents total effective healing for the fight.  All of the columns to the right of this look at the Illuminated Healing portion of the log.  The first 3 columns show the overall and effective healing from Illuminated Healing overall.  As expected, the effective healing of absorbs typically falls above 80%.  The second in red depicts information on Illuminated Healing from Eternal Flame.  The first two columns show IH gained from the direct healing part and the percentage refers to % of overall IH healing (rather than effective).  The next 2 show IH gained from the HoT portion of EF (the part we will be losing).  The last sum column is the percentage of IH from Eternal Flame.  We can see here that anywhere from 10-50%+ of Illuminated  Healing, our number 1 healing spell is from Eternal Flame overall while the HoT portion may represent upwards of 30%+.

The last column of the table is really what is of the most interest.  This represents the loss in effective healing from losing  the healing from the HoT of EF.  These numbers are a little bit generous because IH typically receives around 20% overheal in some situations.  However you can see that for some fights, the reduction in effective healing with the change may reach around 10.5%.  This is cut-throat 10% reduction in healing.

What is the meaning of all this crap?
The nerf to IH from HoTs will hurt.  There is no doubt.  Whatever compensation they offer us for this change needs to address this.  Paladins have always been strong healers but compared to how high Discipline Priests are versus other healing classes, we have never completely dominated.  There are some fights where we shine but the same is for any class.  The current changes in place are fine and acceptable as long as this strong nerf in healing is addressed.

Please keep in mind that depending on your gear and play style, the numbers above will change drastically.  If you have more mastery, you will most likely see higher nerfs with 5.4.  If you rely on Eternal Flame blanketing completely without weaving in any other types of spells you might see an increase in numbers as well.  The best way to see how the patch will affect you is to do the math yourself.  There are also additional considerations.  The refresh of the HoT was the reason why we don’t see high overhealing of the shield because it refreshes the duration. We may, without the benefit of free shield refreshes see a higher amount of overhealing.  On the same note, we may see higher benefits to effective healing of other spells because we no longer have as many absorbs or we may just get sniped more by other healers now.  A huge part of calculating these numbers also depends on your raid group and your other healers.  In 5.4 we will also be seeing different fight mechanics, getting differently itemized gear with tremendous difference in secondary stats depending on your luck.  It’s completely worth trying out some new specs  and finally moving away from what has always felt comfortable.

Cheers to everyone and to hoping that some idea of the compensation will appear soon!


6 thoughts on “5.4 PTR: Now See What You Made Me Do? (Part 3)

  1. This recent change to IH’s interaction with EF really gets me thinking now to the debate of Mastery vs Haste. Whereas before I never shot for the Haste soft cap since Mastery was so OP, if I continue to weave EF’s into my style of healing, adding more haste may make more sence. Guess it also depends on which heals they buff to compensate for the EF nerf. Thanks for another great post 🙂

  2. “The refresh of the HoT was the reason why we don’t see high overhealing of the shield because it refreshes the duration. We may, without the benefit of free shield refreshes see a higher amount of overhealing.”

    This is the key to the puzzle. Anyone still relying on broad EF coverage of their raid will see a very significant increase in IH overhealing for fights without near-constant AoE. Haste may see a come back for those who prefer EF, but Mastery will still be useful on every other spell including Radiance and Light of Dawn. Speaking of which: Light of Dawn may actually find its way back onto players’ bars.

  3. With the nerf to EF not proc’ing our Mastery Shields, I wonder if Haste will be > Mastery. Do you know if the Haste from T16 gear makes it possible to reach 45% Haste giving us that 15th off of EF?

    • Didn’t realize I posted here before in regards to Haste. I prefer EF over SS and will definitely give a Haste build a trial run. As for LoD, being in a 10 man team I use it to help with raid heals when grouped up which occurs in about 60% of the fights.

      • Sorry, been on vacation so no response for a while. Anyways, I’m not sure about a completely haste-based build but I think that hitting some of the haste caps for EF may be more widely used than it is now.

  4. […] There are so many examples of what kind of reduction in healing the nerf to EF will be.  I made a whole post on it with a farm log from real data a while ago.  No simulation bullshit.  No random fight mechanics skewing the data.  Primordius is a fairly […]

What do you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s