Weighing Priest Healing Stats November 7, 2008

Last Updated 12/09/08, patch 3.0.2

Weighing Priest Healing Stats WotLK and beyond…

How we think about healing gear greatly changed with the introduction of patch 3.0.2 and the impending onset of WotLK.  Here is an analysis of what we are looking for.  If you don’t want the math or the assumptions it is based around, just skip to the end of the post for the conclusion.

Balancing

The first question that we need to answer is how we balance healing throughput and mana pool.  Throughput is how strong your healing spells are and your mana pool is how much mana you have to spend on them.  Haste, +heal, +crit, int, and spirit affect the strength of your heals.  Intellect, mp5, and spirit affect your amount of usable mana.  There are various talents that affect each.

Let me illustrate the throughput-vs-mana decision-making dilemma with an example:

If one item gives you 20 spellpower and another 20 mp5, which do you choose to use?

Well, hopefully you answered “it depends.”  If you have 2500 spellpower and zero mana regen, then the mp5 is a better option for you, and vice-versa. We cannot definitively say one would be better than the other in any given situation, or create a simple, linear model equating them. What you get out of your gear largely depends on your other gear.

Assumption #1: Your gear is fairly well balanced.

Gearing isn’t a linear task.  It is a balancing act.   Which stats you look to increase should depend on what you are lacking.

If we were to make the assumption that you weren’t clueless when it came to gear, we could set up a simple model to determine how you equate stats.  We wouldn’t have to worry about balancing bonus healing and mana pool in a cumbersome, cryptic equation.  We could just weigh them and sum them to get a score.

Sometimes, simpler is better.

What Do We Want?

Do we want to know how much healing we do until we go out of mana?

Hell no!  What use would that do us?

What we want to know is how much healing we can do in the allocated time of the fight. Time is the most important variable.

A number that gets thrown around a lot is five minutes … and it’s not a bad number!  Five minutes is a nice, round value and the average length of many fights.  Longer fights tend to have transitions built in to account for their extended length, giving you a chance to gain a lot of mana sitting out of the five second rule. However, that isn’t the case 100% of the time.

Assumption #2: You will know when to switch your gear.

Please understand that there will be some longer fights (and some shorter fights) that require you to wear more (or less) mana regen stats.  This is similar to how there are fights that require higher stamina, or fights that require spell resistance gear.  It never hurts to hold on to a few extra pieces of gear to tailor your stats to the needs of the fight. Be flexible.

Back to the issue of time, figuring five minutes for length of fight is perfect in some regards.  However, we want to have a little extra padding built in to our stat weights to allow for the tougher fights that “accidentally” go on a bit longer than they should.  We don’t want to go into a mana crisis if a DPSer dies too early and the fight goes on a little too long.

Assumption #3: Average fights may be five minutes, but we need to be geared for longer.

When looking at how we think about intellect especially, we’ll be computing it for fights of 6 minutes in length (20% longer). I have found that is was better to undervalue mana stats to allow for increased longevity in a pinch.  As always, your results may vary.

Spellpower vs. Mana

How much more healing can you do with +1 spellpower?  How much with +1 mana?

It’s the age old debate, being rehashed again.  Looking at the mechanics of one spell to model this, or how Blizzard has gems itemized, or the mechanics on one talent will NOT cut it.  I have seen so many people make these mistakes.

As I stated before, we aren’t looking to see how much healing we do before we run out of mana.  That isn’t our goal.  We want to maximize our healing potential in the time allotted.

It’s obvious how spellpower would play into that, but how would mana?

Overhealing.  If you have enough mana, you can toss that extra heal that’s 80% overheal.  You can burn that extra mana topping someone off that you wouldn’t be able to do otherwise.  You can put that extra stabilization spell up.  You can cut your IFSR time a little shorter.  The short answer: more mana lets you cast more spells.

Using my prior model and converting +heal to spellpower, we would weigh spellpower at 0.49.  Mana-related stats would be converted into their equivalent amount of mp5 for the fight, which would then be weighed at 1.0.  With patch 3.0.2 (the pre-WotLK patch), we see mana coming from different (and more frequent) sources.  Rather than adjust the mana stats accordingly, I adjusted the spellpower-related stat weights instead, weighing them 23% higher than before.  Yes, even despite the change to downranking.

Spellpower is now weighed at 0.60 per point.

This aspect of the weighings is not something I feel completely locked into, and it is the most likely to vary depending on playstyle and raid composition.

Assumption #4: You will adapt your gear to your playstyle and your raid conditions.

There are innumerable ways of playing the game.  The objective of the game is to have fun.  If you have more fun stacking mana regen and chain casting more, but healing for less, than so be it.  There are certainly less effective ways of healing, yes, but I hesitate to really call anything outright “wrong.”

My assumptions about playstyle and typical raid composition when making stat weights are my perceptions about how priests heal and how raids function.  Your results may vary.

Feel free to balance things differently, especially if you feel things aren’t working for you.  Experiment and  adapt.

Spirit/Intellect-Based Mana Regen

This is the big ticket item in terms of your mana.

Your mana will regenerate when you are out of the five second rule (meaning, after you have stopped casting for 5 seconds). This is your out-of-combat mana regen, otherwise referred to as out-of-five-second-rule (OOFSR) regen.  If you have 3 points in Meditation (which all PvE healers really, really should … you cannot spend three talent points in any better way), 30% of that OOFSR mana regen will continue while you are in the five second rule (IFSR).

Assumption #5: You have spent your talent points in logical manner.

Let me repeat: Meditation is a tremendous benefit to any PvE healer.  All gear weights listed here will assume you have this talent.  Certain other “core” talents are assumed in the weights at the bottom of the page.  See the test cases in the “Putting it All Together” section.

Our OOFSR mana regen (measured in MP5) is expressed by the following equation:

MP5 = 5 * (0.001 + sqrt(Int) * Spirit * LevelRegen)

LevelRegen is a coefficient based on your level.  At level 70, it is 0.009327.  At level 80, it is 0.005575.

To figure your mana regen gain from spirit or intellect, you need to know your spirit and intellect values. At the bottom of this post I outline a few gear assumptions.  For simplicity’s sake, I’ll mention up here that I’m assuming a level 80 priest with 1k int and spirit (early T7 content levels).

Given those assumptions, the OOFSR mana regen from a point in each stat becomes:

MP5 from Spirit = 0.8815
MP5 from Int = 0.4406

Those are the OOFSR values.  Say you spent about 80% of your time IFSR.  These weights then become:

MP5 from Spirit= 0.3879
MP5 from Int = 0.1939

That is per point of each stat.  If you have buffs or talents that modify your intellect multiplicatively, for example, you should take those into consideration as well (Blessing of Kings, Mental Strength).  Humans, with their  Human Spirit racial trait, get a 3% better return on spirit (which is not taken into account).

Intellect

For a holy priest, 1 point of intellect gives you 15 mana (16.5 if you have Blessing of Kings on you).  Thinking about mana gained, per point of intellect, we divide the amount mana from intellect by the length of the fight in seconds, then convert that value into mp5 by dividing by 5 (we’ll be factoring in the OOC/Meditation mana gain later). We convert into the unit mp5 as it remains the defacto unit of measure for mana over time.

Mana from Int = { [ Intellect x 15 ] / Time } / 5
Where: time is measured in seconds

Again, this is not figuring for Improved Blessing of Kings, or the fact that the priest is deep disc. Those both will increase the value of intellect in terms of mana gain.

With Improved Blessing of Kings (10% increase to all stats):

Mana from Int = { [ Intellect x 16.5 ] / Time } / 5

Speccing deep into discipline, you (must) pick up the talent Mental Strength, which increases your intellect by 15% (at max rank).

Mana from Int = { [ Intellect x 17.25 ] / Time } / 5

… and Imp. Kings on a deep disc priest, the benefits to intellect get multiplied, giving you 18.975 mana per point of intellect:

Mana from Int = { [ Intellect x 18.975 ] / Time } / 5

Now, if you look at stretching that mana over six minutes, one point of intellect gives you the equivalent of:

1 int = 0.208 mp5
1 int, w/Imp Kings = 0.2292 mp5
1 int, w/MS = 0.2396 mp5
1 int, w/Imp Kings & MS = 0.2635 mp5

And this isn’t taking replenishment, shadowfiend, rapture, or mana regen into account. Just its mana pool benefit.

Other Items Affecting Mana

There are spells and buffs that affect our spendable mana pool:

• Replenishment – When you have the Replenishment buff (from a shadow priest, ret paladin, or survival hunter), you gain 0.25% of your max mana back every second.  Increasing the intellect on your gear increases the effects of replenishment.  This will be a very common buff in raids.

• Shadowfiend – It restores 4% of your max mana per hit (it’s return is no longer based off of your spellpower or how much damage the little guy does).  Analyzing a number of combat log aggregators (such as WoW Web Stats) since patch 3.0.2 shows Shadowfiend to not be 100% effective.  That number is closer to about 90%. On most fights, it will be used only once (as the spell has a five minute cooldown).

• Innervate – Increases your regen by 400% while not casting and allows 100% of that total regen to occur while casting (for 20 seconds).  In most usual cases, the mana return from an Innervate would exceed your mana pool, making optimizing stats for its return somewhat irrelevant.  In any given fight, other casters or healers may have greater need of this buff than you, and it remains something not consistently cast on holy or disc priests.  The mana gain from this buff is not taken into account in any calculation, for those two reasons.

Spirit

From above, we know how much spirit is worth in terms of mana.  Holy priests with Spiritual Guidance get 25% of their spirit added to their spellpower (4 spirit = 1 extra spellpower).

Haste

How you use haste really depends on your playstyle and raid role, and similar to spellpower, will vary based on your personal preference.  While not high on your priority list of stats, once basic needs are met (such as having enough mana regen to comfortably make it through a raid encounter), you’ll want to focus more on collecting haste (potentially making the stat value of haste increase with gear level).

Having haste is like having finesse. It can allow you to pull off miracles of the speed-healing nature.  Fast healing can save a raid from a wipe.

For holy priests, haste is almost as significant as spellpower.  It lets you lands those heals quicker.  In burst situations especially, haste will allow you to keep up with heavy damage easier, and will let you top people off faster (allowing for more time out of the five second rule).  One point of haste is equal to about 75% of what 1 spellpower is worth.

For disc priests, having haste allows you toss out more healing spells between PW:S and Penance cooldowns, thus making it more core to your healing ability.  You may need it to be able to fire off a large heal, or extra stabilization spells, when necessitated, and still keep Penance and PW:S going frequently.  For disc priests, I value haste higher (80% of spellpower).

Crit

Crit rating will increase the chance that certain spells will deal 1.5 times their regular healing. It will also increase our mana savings through Improved Holy Concentration, Divine Aegis, Rapture, as well as chance happenings of larger heals that don’t overheal much, and procs of Serendipity that would not occur naturally.

At level 80, 167 intellect is equal to a 1% bonus chance to crit and 45.91 crit rating equals 1% bonus crit as well.

Base crit rating weight: 0.0480 (prior weight, adjusted for crit rating change at level 80)

Talents have modifiers that increase the value of crit:

• Rapture and Divine Aegis (disc): 0.0317
  Critical spells cause the Divine Aegis effect on the target, which grants a direct mana return from Rapture every time this happens.  As a disc priest, if you are casting about 25 spells that can crit a minute (Penance has two ticks, remember).  An increase in crit by 1% would give you about 0.2 more procs of DA per minute.  With max absorption from the shield, I was seeing about 80 mana returned per proc of Divine Aegis (in T7 gear). That works out to be a benefit of 1.3 mp5 per 1% crit, or 0.0317 weight per crit rating.

• IHC (holy): 0.0486
  Improved Holy Concentration gives you a 45% chance of entering clearcasting (mana free spell) upon having a critical hit.  Every 1% crit you gain will give you an extra 0.45% chance of clearcasting on healing spells that have a chance of proccing IHC (Flash Heal, Binding Heal, Greater Heal).  Assuming a low cast frequency of those spells (36 over the course of a encounter), we would see about 0.13 extra clearcasting procs over the fight.  Greater Heal is the highest mana cost spell to be usable by clearcasting.  Assuming that, we gain an equivalent of 2.23 mp5 for the fight per 1% crit gained.  Additional benefits can be gained from cheating the FSR with IHC, but that is difficult to model, and not always an option, hence not taken into consideration.

• Serendipity (holy): 0.0031
  Serendipity will return a 25% of a Flash Heal or Greater Heal’s mana cost if it overheals.  Damage is a little more predictable in WotLK, so we estimate that only about 25% of heals will not overheal.  Each 1% crit gained will increase the chance that the spell crits and mana is returned.  As before, assume a low cast frequency of Flash Heal and Greater Heal (33 over the encounter).  This would save the equivalent of 0.14 mp5 over the fight.

Total disc crit weight: 0.0797
Total holy crit weight: 0.0997

Intellect, since it grants crit, also increases in stat weight value due to these weights.  For its benefit to crit, Intellect gains 0.0219 for disc priests and 0.0274 for holy priests.  Modifiers to intellect (such as Blessing of Kings and Mental Strength) increase this weight.

Situational Stats

Stamina, spell resistances, hit, and other situational stats are not taken into account, as they are fight-dependent and something you do not often have to gear for.  See previous assumption about knowing when to switch your gear.

Gear Priorities

Gearing up, you have a number of priorities in order to reach your goals.

Priority #1: Enough spellpower to heal through burst damage and enough mana-related stats to not go out of mana.

The first priority is a bit of a tie, as neither spellpower or mana-regen is more important than the other. In terms of balancing spellpower and mana, I use the general rule of thumb: at lower levels of gear (level 70 raiding gear), I am looking for about a 4:1 spellpower to IFSR mana regen.  At higher levels of gear (level 80 raiding gear), I aim to stack spellpower a bit more, and aim for more of a 5:1 ratio.

Priority #2: Get crit!  Goal: 20% unbuffed.

This priority works to augment your priorities in #1.  It increases how much you heal by and your mana return from large heals (through talents like Serendipity).  The 20% seems to be a bit of a magic number.  At that point, raid buffed, you should go up to about 25% crit which is where certain talents (such as IHC) start to see competitive returns for their talent point investment.  Crit is something you continue to benefit from the more you stack it.

Priority #3: Grab some haste.

It doesn’t hurt to start picking up haste while reaching your other priorities, but I wouldn’t go too much out of your way for it.  Starting out, aim for about 5%-10%.  When you feel comfortable in your previous two goals, try to get somewhere in the 10%-20% range.  I am giving broad ranges for this as haste is a matter of taste to many people.  Players who continuously cast-cancel heals may see less benefit from it than players who spam heal certain spells (like CoH) through burst damage.  Try some haste on and see how it feels.

Setting Goals

Let’s quickly lay out what we are aiming for to start content (unbuffed):

• T4 Content
  • Spellpower: 635
  • IFSR MP5: 140
  • Spirit: 340
  • Int: 320
  • Crit: 11%
  • Haste: 0%

• T5 Content
  • Spellpower: 910
  • IFSR MP5: 185
  • Spirit: 360
  • Int: 350
  • Crit: 12%
  • Haste: 2%

• T6 Content
  • Spellpower: 1120
  • IFSR MP5: 225
  • Spirit: 400
  • Int: 400
  • Crit: 14%
  • Haste: 4%

• T7 Content
  • Spellpower: 1500
  • IFSR MP5: 300
  • Spirit: 850
  • Int: 850
  • Crit: 15%
  • Haste: 6%

• T8 Content
  • Spellpower: 2200
  • IFSR MP5: 450
  • Spirit: 1100
  • Int: 1100
  • Crit: 20%
  • Haste: 8%

Standard disclaimer: These are general goals.  If you are short in one area, but exceed in all others, you’re likely in fine shape to tackle the content.  Skilled healers (and raids) can get by with a lot less.

T4-T6 content is listed for purposes of aiding the transition from TBC to WotLK.  Understanding where you are coming from to where you are headed can be important for some people.

Putting it All Together

At this point, in order to get some stat weights, we need to create some test cases.

In both cases, the priest is assumed to not be human, and to have the talent Meditation.

Deep Holy Priest (Test Case)
(w/Spiritual Healing, Spiritual Guidance, IHC, Serendipity, Test of Faith, Empowered Healing, Circle of Healing)

• Use “T7 Content” goals for stats
• 80% in FSR
• 75% heals proc Serendipity
• 25% of heals proc Test of Faith
• 85% of time w/Replenishment

Deep Disc Priest (Test Case)
(w/Mental Agility, Mental Strength, Rapture, Borrowed Time, Divine Aegis, Penance)

• Use “T7 Content” goals for stats
• 78% in FSR
• Heals give 75% of Rapture returns
• 85% of time w/Replenishment

T7 content level goals are assumed for weighing stats as that is the goal (with slightly higher int and spirit, 1k each).  Ending T7 content, your IFSR mana regen from spirit and intellect will be slightly higher (a few hundredths of a stat weight each, if even).

Using the various stat weight models, test cases, and assumptions listed above, we have for following values for stat weights:

EDIT: These weights updated on 12/02/08

Holy Priest
Crit    0.1497
Haste    0.4000
Int (crit)    0.0361
Int (mana)    0.2295
Int (regen)    0.2133
Int (replen)    0.1753
Int (sf)    0.0825
Spellpower    0.6000
Spirit (regen)    0.3879
Spirit (spellpower)    0.1500

Disc Priest
Crit    0.1197
Haste    0.4200
Int (crit)    0.0301
Int (mana)    0.2635
Int (regen)    0.2530
Int (replen)    0.2016
Int (sf)    0.0949
Spellpower    0.6000
Spirit (regen)    0.4002

And in summary:

Deep Holy Priest Stat Weights:

• Spellpower: 0.60
• Spirit: 0.54
• Intellect: 0.74
• Crit: 0.15
• Haste: 0.40
• MP5: 1.00

Deep Disc Priest Stat Weights:

• Spellpower: 0.60
• Spirit: 0.40
• Intellect: 0.84
• Crit: 0.13
• Haste: 0.42
• MP5: 1.00

This gives you a starting point.  Tweak these stats to your own personal situation (example: human priests get 3% better benefit from spirit, not having Imp. Blessing of Kings decreases intellect’s weight by 10%, etc.).

Final Thoughts

This is my first real look at healing stats in WotLK. This model and its resulting stat weights will likely change as gear gets better, raid compositions evolve, encounters/strategies change, and more theorycrafting.  I’ll give this all a good second pass once Wrath is actually released.

I’ll have gear lists based on these stats up shortly (next day or two). They will include WotLK gear, and all gear released prior. (in the meantime, here is a LootRank list for holy priests, and here is one for disc priests).