Spiritual Guidance: Shield spam and Divine Aegis, a theorycrafting story

A couple of months ago I found myself talking to a non-priest about the gems I had slotted on my character. He was of the understanding that disc priests wanted nothing but crit, and thought it was strange that I had gemmed straight spellpower on all my gear. Figuring he was behind on the times, I happily explained to him that I was using the standard gem set up for shield spamming disc priests, which works around the premise that if the majority of what we do is cast shields, then we should stack as much spellpower as possible in order to make our most used spell (Power Word: Shield) absorb more. This is the standard practice advised to shield spammers throughout the priest community, and I've advised it here on Spiritual Guidance before as well.
The non-priest still didn't understand though. He kept insisting "but crit ..." which inclined me to gently stroke back his hair and say "there there, poor little confused non-priest, it's all right." I allowed him his dignity though, and instead went on with my explanation. I told him that alternative stats like crit and haste didn't do much for shield spamming since Power Word: Shield can't crit, and Borrowed Time removes the necessity for haste since the talent carries us down to the 1 second GCD soft cap whenever we cast Power Word: Shield. The non-priest still didn't understand, so I explained to him that a disc priest's primary interest in crit was Divine Aegis, a talent which applies a second shield whenever one of your spells crits. "But shields don't crit," I reiterated. "The heal from the Glyph of Power Word: Shield can, but that would only add say ... 500 extra absorption from Divine Aegis. The spellpower is still better."
As I typed out those last words, they boomeranged back and hit me square in the face. Startled, I peeled the sans serif off my nose and and reexamined the limp letters in my hands. Suddenly I wondered, "is that really true?"
The question
After I pulled the last of the ink from my eyebrows, I quickly set to work at trying to figure out the math for my question. My stream of thinking went like this: The Glyph of Power Word: Shield causes each Power Word: Shield you cast to heal your target for 20 percent of the amount of the shield. If we say the the average absorb amount for Power Word: Shield is 10,000, then the glyph will heal our target for around 2,000 health. If the heal effect crits, then we'll heal our target for 150 percent of your heal, which is 3,000 health. But that critical heal from the glyph will also apply a Divine Aegis. The Divine Aegis will absorb damage equal to 30 percent of the critical heal, even if it overheals. So, if the crit heal from the glyph is 3,000, that means the Divine Aegis would absorb around. 900 points of damage.
I sat back and looked at my numbers. The numbers were inflated but still, I had never really considered just how much extra absorption Divine Aegis could do through Power Word: Shield. I never questioned the validity of spellpower stacking before. After all, I had come to the bubble spam scene late, pursuing a career in tank healing well into early Icecrown Citadel. When I did submit to the good of the raid, I just accepted the methods and shrugged off any benefit from crit as negligible. But now I started to wonder if crit and Divine Aegis had been accounted for in shield spammer throughput. How beneficial might it be to increase my Divine Aegis output as a shield spammer?
My thoughts wandered to the extra spellpower I had on my gear from gems and wondered what they did for me. I had 17 Runed Cardinal Rubies equipped, adding up to almost 400 spellpower. If I swapped those out for spellpower and crit gems, I could get an extra 3.7 percent chance to crit in exchange for about 200 spellpower. If I wanted to drop the spellpower completely and go with straight crit gems, I would be giving up that 400 spellpower for an extra 7.4 percent.
I immediately recoiled at the amount of spellpower I would have to drop to gain a sizable amount of crit. Plus, the gains seemed too small to make a big difference. What could 3 percent really do? Plus, I was still trading off spellpower for a chance to apply extra absorption, not a guarantee. The extra spellpower, on the other hand, would affect every shield I cast. Things seemed doubtful for crit, but I was determined to give it a fair shot.
I jotted down more scenarios and considered trading out my haste gear for crit gear, since it didn't require losing any spellpower. I wondered about diminishing returns, raid buffs, and crit from intellect. The more I thought, the more the variables started to pile up around me and it wasn't long before I realized I should get some more qualified help. So, I gathered up my notes and went off to see Zusterke.
The wizard
If you don't know who Zusterke is, no worries. I've mentioned him on Spiritual Guidance before only in passing. Zusterke is a theorycrafter in the priest community, most commonly found working his math magic at PlusHeal.com or even occasionally here at WoW.com, in the comments of Spiritual Guidance. He's also got his own little site, Zusterke's Corner, where you can find lots of useful calculators for priests. He has even been a guest blogger on the legendary World of Snarkcraft, a since retired priest blog run by Joveta and Serianna, the queens of priest blogging in early Wrath of the Lich King. (This is where I first learned of him.) Zusterke also happens to be one of my better friends in the WoW community, and I frequently bounce terrible article ideas off him before going further with them. So technically, you guys already owe him for sparing you a lot of rambling from me. He couldn't spare you this one, however.
Anyway, on a late July morning, I suggested the idea to Zusterke that crit could be better than spellpower, and asked him for his opinion. Immediately he seemed quite doubtful, and moments later provided me with some preliminary math.
Immediately the question flat-lined. My non-priest friend's idea wasn't going to hold up. I watch poor crit scurry off to a corner and whimper like an abused Shadowfiend; my heart swelled sympathetically for it.Zusterke: Well ... I made a raw throughput comparison. Much like the spirit and int technique I make the formula and derive for both sp and crit. Then compare both.
HPS = (Base + coeff * SP) * (1+Bonus) * (1 + 20% * (1+ crit/4591 * 95%))
d(SP) = coeff * (1+Bonus) * (1 + 20% * (1+ crit/4591 * 95%))
d(Crit) = (Base + coeff * SP) * (1+Bonus) * 20% * 95% * 1/4591
Then we check d(Crit) > d(SP) to know when crit rating is more valuable. (Not being entirely fair since we compare 1 crit to 1 SP and you can get more SP per point of crit) It works out to be: SP > 27146 + crit rating. Which means your SP should be over 27K before crit rating becomes stronger. But I might want to double check this.
For whatever reason, I got it in my head to try advocating for crit from another angle and wondered if there was some sort of compromise I could make on its behalf. I suggested to Zusterke that the rigidness of disc gearing and gemming was a cause for concern to a disc priest who also wanted to tank heal. What would happen if we started to apply the 15 second Weakened Soul debuff to the picture? Zusterke admitted his calculations were strictly for bubble-spamming, and that for tank healing he still recommended balance, something I also suggest here in Spiritual Guidance.
After that, I didn't think much of the failed theory until Zusterke messaged me a week later. I was surprised when he revealed he'd been working on it.
I stared stupidly at Zusterke's formulas. He had told me many times that people often misinterpreted his theorycrafting, then made it worse by trying to use their incorrect conclusions to support their own theories. This was the first time I saw for myself just how easy it was to misunderstand this stuff. Fortunately, to help me understand, Zusterke later gave me a tour through his daunting spreadsheet, and let me play with some of the crit values to see how the numbers worked with each other. (He even worked in some extra variables for me, like overhealing.) It certainly was impressive to look at.Zusterke: I've been thinking about your crit theory a bit. I thought I'd try something... Investigate whether crit or SP would deliver the biggest contribution to Divine Aegis. It seems this can depend on the spell, but things are looking a bit brighter for crit. Of course, this is restricted purely to Divine Aegis and not the whole throughput of the spell.
Dawn: What's your process?
Zusterke: Well, I worked out a big nasty formula. But it can be brought down to this:
SP = factor * critrating - Base/Coeff
Factor is the difference in stat cost between spellpower and crit rating. (For the stat cost I decided to base myself on WowWiki.)That would be 1.1667. Anyways, this equation shows the balance between both stats in terms of throughput. It is 6/7 in terms of cost. However I need to inverse that (7/6) because for one point of crit rating I can get 7/6 spellpower.
Dawn: What spells have you looked at?
Zusterke: I don't have to see many spells. Check this: imagine we about 40% crit on a target, based on talents, buffs, gear etc... This is the equivalent of 1836.4 crit rating. Our formula becomes:
SP = 2142.5 - Base/Coeff.
As such, for any reasonable amount of SP, crit is likely still to be the most valuable stat. The base heal and coefficient will combine to a negative number, in benefit for crit. So... Even if the base heal would be 0, any amount of SP above 2142 would imply crit is better than SP.

The new math intrigued both of us. Since I'd been rooting for it, I was pleased to see how influential crit was to the Aegis (something to keep in mind when us disco priests get back to tank healing.) Zusterke seemed to appreciate the working relationship between the two, commenting that crit took on a role very similar to hit, in that the Divine Aegis needed it to function.
Unfortunately for crit, the new formulas weren't enough to coax the little stat from its corner. Zusterke's formula determined crit was only stronger when Divine Aegis was isolated. Once the throughput of Power Word: Shield is added back in, crit no longer holds up against spellpower. I made a graph to illustrate this (read: not a spreadsheet.)
Unfortunately for crit, the new formulas weren't enough to coax the little stat from its corner. Zusterke's formula determined crit was only stronger when Divine Aegis was isolated. Once the throughput of Power Word: Shield is added back in, crit no longer holds up against spellpower. I made a graph to illustrate this (read: not a spreadsheet.)

I used 100 Power Word: Shields as my foundation, since it would allow an amount of Divine Aegis obviously proportional to the percentage of crit. 100 is also a good number to estimate for a normal fight. (To give you an idea how many shields us spammers use, in the heroic Lich King fight I might cast between 300 and 400 shields during the 15 minute fight. And that is casting shields for the first 2 phases at very specific points in the fight.) The absorption amount for Power Word: Shield is determined using the normal formula. The absorption amount for Divine Aegis is determined by multiplying the glyph crit amount by 30%. You get the total amount of throughput from absorption by adding up both totals.
The conclusion
So what did we learn out of all this? Well for one, that my non-priest friend was wrong. The theory I proposed on his behalf didn't hold up, and the spellpower stacking that the community and myself, however blindly, have advocated was the correct choice. Overall it's a nice feeling for me. I felt satiated to come out where I started, but now with the understanding that I've explored different routes as a spammer. Hopefully this exercise has done the same for any other disc priests who might have wondered why spellpower was the thing to do over everything else.
That said, between my talking with Zusterke and my own simplified math, I now have a more tangible understanding of the trade off between crit and spellpower. Min-maxers will recoil when I say this, but the difference honestly isn't that large. Over 100 Power Word: Shield casts, you lose a little from each shield by taking crit, but you can see in my graph it only adds up to about about 3 shields. It makes me think to myself that if you're interested in flexibility (e.g. the ability to swap to tank healing without regemming your gear) then you're not losing much to take a more balanced approach to gear. (Alternatively, you could also say you don't gain much by taking a little extra crit, but I have to draw the line somewhere.) Again, if you're a min-maxer then this probably sounds like appalling advice, but I think it's a fair trade.
I could be wrong, but I have in my mind an idea that most disc priests don't play the way I'm used to in my progression guild, assembly line raids. I rarely have to cross heal in a guild raid, but back in the day when I played with more pugs and less progressed guilds, I remember everyone cross healing more to cover each other. It's up to you really. Typically, I really like to advocate min-maxing as a way of getting the most out of what you do, but I understand it's not always applicable or needed. Plus, true min-maxing must be considered on a fight per fight basis. You can't necessarily pick one set up and be truly min-maxed for an entire dungeon. Spellpower stacking is no exception. If you do switch roles a lot, feel free to balance your stats a little. The crit will help out your normal casts and in turn your Divine Aegis output. So in that sense, my non-priest friend wasn't wrong about crit being what us disc priests want. We do want it, just so long as we do more than push the Power Word: Shield button on occasion.
Filed under: Priest, (Priest) Spiritual Guidance
Patch 5.4 patch notes
Virtual Realms feature revealed
The Proving Grounds are coming
The latest patch 5.4 news





Reader Comments (Page 3 of 3)
Prissa Sep 6th 2010 3:36AM
I'm in a slight predicament with my priest. My main spec is disc and I've been very sure to stack up the SP and a touch of int (coz I was going oom on lk). Now my guild is wiping on heroic sindy and my RL has asked me to swap my shadow OS to holy to use for this fight.
He's given me a spec which is based on renew spam, includes body and soul and does not have serendipity (which made me sad coz I love serendipity as a talent).
So obviously I don't have quite the right stats for holy, I know I need a lot more haste to be renewing on gcd without borrowed time. So I've jammed in a number of haste gems for now, but I have no idea how I should balance out my stats to cover both my specs without gimping myself in disc. Help plx :(
mrdragonfell Sep 6th 2010 4:11AM
Well I think you idea for the crit/haste trade off becomes viable in a pvp situation. many priests forgo crit for the haste off pieces neglecting to take into account borrowed time as haste and the benefits of aegis.
In a pvp situation haste does have it's benefits but a free absorption can do amazing things when your down to your last drop of mana and praying dots finish the other guy off.
Poltergeist Sep 6th 2010 4:25AM
From a 25 man perspective, I can't argue with this article.
However, if you're like me and prefer 10s, I can't stress enough that balance is by far more important.
In 10s, assuming only 1 other healer besides yourself, you'll need to do more than just bubble bot. I find myself relying more on my other heals, especially Pennance and PoH, in 10s, making haste a bit more valuable to me, even though I don't dare gem for haste.
I think the general rule of thumb for disc should be something along the lines of...
25s/Bubble bot: Gem SP
10s: Gem for balance.
Bvannas Sep 6th 2010 4:44AM
I saw two preists in a high end raiding guild chatting in /say and one said 'sp+haste gems generate 4 more hps per gem than sp'. I just lauged thinking 'What error bars are there on that?' Theorycrafting for specific changes has to work off a certain rotation and a certain gear setup, there are assumptions to make the calculations. Now these assumptions will differ player to player and raid to raid. The errors could be greater than the increase in healing you calculate, there is always a chance you are reducing your HPS by following the theorycraft.
But aside from that, is the change you are theorycrafting going to help save lives in a raid? More specifically, is it going to help YOU keep people alive? Your own healing quirks and those of whatever raid you are in matter.
My point is that once you have questioned the numbers from the theorycraft, question their relevance. For instance, is that DA doing anything? Is it hanging around not getting absorbed, (just for relevance, it probably is, but hypothetically, if it wasn't, then calculations based off it are not helping).
cupofbear Sep 6th 2010 5:01AM
The groups I'm running with on my Disc Priest, I find that the ideal situations don't happen all of the time, and I've needed to get direct heals out to a person on most fights that the groups haven't been doing for a while. I've been mostly going with spell power in red, spell/haste in yellow, and spell/spirit in blue if it has a decent socket bonus, with one spell/crit gem to just push me over the 30% self buffed crit line (it was sooo close, lol!).
This gives me 18.3% haste, 30% crit, and 3331 Spell Power just on my own, which going for a little more balance, like Dawn, and a few comments mentioned seems to have really worked well for me and my raid's needs. There are certainly times when all I need to do is bubble spam, and in regular content, since I'm only currently doing heroic modes on my Bear, I don't really feel too gimped for the loss of spell power, and I've never had an issue with mana either, which makes me feel safe loosing a little spirit as I progress with her.
cupofbear Sep 6th 2010 5:04AM
I should say, that on any fight, Glyph of PW:S is #1 for my healing, followed either by penance or POM, and PW:S is certainly top absorber, but the absorption from DA seems to be such a significant amount, that I'm happy having my crit where it is, if not higher.
Zinn Sep 6th 2010 8:05AM
If anything I go for int in yellow gems, for bigger Rapture procs. And it also adds crit of course.
Xaklo Sep 6th 2010 2:44PM
Doesn't Bi-weekly mean once every two weeks? I think semiweekly is twice a week.
Renee.Troughton001 Sep 6th 2010 8:55PM
A few things that have me unsure of the validity of the information here.
1) The googledocs screenshot (no link provided ftl) doesn't even list penance our primary heal after pw:s - where did this baby fit in the scheme of value on crit vs sp?
2) Pw:s being the only one where sp > crit seems a bit off when the googledocs show flash heal as benefiting from crit more (I am assuming we are only talking about DA procs here).
3) Was the maths based upon pure sp gemming vs appropriate spellpower + crit gemming (ie only gemming sp + crit if there is a yellow slot).
Dawn Moore Sep 6th 2010 9:15PM
E-mail me at dawn@wow.com and I'll send you the doc. I was not going to link a document attached to my private e-mail or Zusterke's. Penance can easily be plugged into the spreadsheet if you want to play with it. I'll have Zusterke answer your question tomorrow morning (Europe time) since the isolated DA was all his doing, and he can explain what he found out for the individual gems directly, rather than me copy pasting two conversations together.
Zusterke Sep 8th 2010 7:31AM
Some very good questions there. While I'm replying to your questions, I think this might answer other people's comments and questions as well.
1) Yup, I did not include Penance since, well.. I admitt that's my bad. I always had problems with pinpointing the exact measures of Penance. It just doesn't seem to follow any rules. In the end I cheated and ninja'd DrDamage's coefficient. I hope the author(s) forgive me and I hope this somewhat shameless plug to their addon compensates :)
http://wow.curse.com/downloads/wow-addons/details/dr-damage.aspx
I included Penance in the spreadsheet now and provided a link at the bottom of this comment. There are some other goodies added to answer your other questions.
2) I admitt I have somewhat trouble understanding your question. Let me rephrase some things more in detail. This is for the spells "entirely" (not just DA procs):
1 SP < 1 crit rating < 7/6 SP.
1 spellpower is weaker than 1 crit rating, but you get more spellpower per point of crit rating (7/6 in fact). This is a result of the lower stat cost of spellpower.
Exceptions to this are:
- PoM where 1 SP > 1 crit rating
- PWS where 1 SP > 1 crit rating
- PoH where 1 crit rating > 1 SP (a remnant of the coefficient nerf a while back)
For DA procs only, we see that all spells benefit more from crit rating than SP, including even PWS and PoM which are the big shots of SP.
Keep in mind this is valid for 45% crit and 3700 SP, with said overhealing. For other stats it might yield a different result, though it will be fairly reasonable to assume that for most circumstances:
- DA will still rely mostly on crit,
- PoH will mostly support crit,
- PWS and PoM wil mostly support SP.
The spreadsheet at the bottom can be downloaded and played with to heart's content. Feel free to experiment with some numbers and post any oddities you may find :)
3) The math was based foremost on 'just mathematical balance'. However, in order to obtain more concrete answers (read: understandable) we had to add concrete situations to it. For this we went as far as examining 1 SP, 1 crit rating, 7/6 SP (stat cost), 23 SP vs 20 crit rating. I take it your question is how hybrid gems weigh up against pure gemming?
Although the spreadsheet holds the basic information you need to answer that question, it does not answer this directly. A negligence I owe you my appologies for. I rephrased your question in 2 additional columns (K & L):
K - How much crit rating must a socket bonus yield before a 12SP+10crit gem outscores a 23 SP gem?
L - How much SP must a socket bonus yield before a 12SP+10crit gem outscores a 23 SP gem?
This yields low requirements for almost all spells except PoM and PWS. For some spells it even yielded a negative number. This would suggest that crit is more valuable but this isn't necessarily the case. What it means is that, while 23SP may beat 20 crit rating, the hybrid SP+Crit gem beats the pure SP gem for this spell. How can that happen? Because the 23 SP gem is slightly under budget (it should be 23.3 SP) and the hybrid gem is slightly over budget (it should be 11.7 SP). The difference is very small, but apparently enough for some spells to tip the balance.
Keep in mind however that these observations are only valid under the rigid situation we have examined:
- 3700 SP total
- 45% crit total (including raid buffs, target buffs, procs, talents etc)
- 40% overheal on average on normal heals.
- 50% overheal on average on crit heals.
For different situations, we can expect slightly different outcomes so it is well worth playing with the settings a bit.
The spreadsheet is available here:
Zusterke Sep 8th 2010 7:37AM
The form ate my links.. booh! :-)
spreadsheets.google.com/ccc?key=0AndYOGtXAJobdDNKMmJxRDhzdDZFOTZLTXVFdV8zaEE&hl=en
(add HTTPS yadi yadi)
img186.imageshack.us/img186/3333/calcs.png
(add HTTP yadi yadi)
It worked!
simon Sep 6th 2010 8:55PM
Same as others have said .. i have never played a preist and not really that arsed .. But this article was ACE , Great reading ... GZ
Renee.Troughton001 Sep 6th 2010 10:18PM
Thanks Dawn - what would be really interesting to know would be maths wise - at what point does penance/flash heal with crit + sp outway pure sp shield spamming.
The reality of the situation is unless it is LK defile prevention or known incoming mechanic prevention if everyone is topped up you probably aren't going to be spamming shields. The only thing you would be doing is tank healing or penance/flash healing weakened soul people that aren't quite full.
I would probably be doing very similar healing as most other disc priests where penance and flash heal only account for a total of 10% of my healing - but what if we shielded less and did heal more (like what is proposed in cata) - at what point of percentage total healing done by flash penance would crit + sp outway pure sp?
Ultimately this is probably all moot with the cataclysm changes but I would love to see the maths done when the expansion has finally been released.
Dawn Moore Sep 6th 2010 11:41PM
Without the math I can tell you right now that in the current game, Penance and Flash Heal won't overcome shield spamming for throughput, doesn't matter how you gem. You could in theory, but the numbers required are unattainable. Without Divine Aegis, a crit Flash Heal alone, even if it crits for it's highest amount every time, is still less than 1 Power Word: Shield with 1 non-crit heal from the Glyph of Power Word: Shield. A non crit Flash Heal doesn't come close. If you add in the amount of the crit that would convert to Divine Aegis you have something to work with, but Divine Aegis will still cap at 10k so your target would have to constantly be taking damage for you to fairly count each crit Flash Heal for the full value (the full value being with the Divine Aegis attached.) But in this scenario you still need to be critting with almost every heal you cast, and you'd probably need to be casting 1 second Flash Heals without Borrowed Time to boot. Even if you incorporate Penance at every CD, Renew, Grace, and a single Shield on every Weakened Soul CD, you'll still be behind shield spam most of the time.
Someone has probably already done the math on this, between PlusHeal.com and ElitistJerks.com. If not it wouldn't be too hard to draw up. Maybe Zusterke might want to do it for fun after he reads this post. =P
As for straight shield spamming, it really depends on where you are in the game. With the 30% buff in ICC it's true that you no longer need to shield spam, but it used to be that's what you did. H Putricide at 0-5% is the most notable fight that seemed to require a disc priest shielding from the beginning of p2 to the end, never stopping except for an occasional spot heal. This was just because there was so much danger of players randomly (or not so randomly) being hit by something that could 1-shot them. Blood Queen and Rotface were also good fights for spamming from the beginning to the end of the fight. Princes you can play preemptively and Festergut you're full on the raid for half the fight.
fr-schneider Sep 17th 2010 8:27AM
What about this : http://penancepriest.blogspot.com/2010/03/divine-aegis-revisited.html