A guy called Twistedmind on the WoW Forums did a write-up with some math-stuff on it (goes onto the next page too), not sure what you guys make of that.

Could you elaborate on the 40% mastery number he pulls out of his sleeve there? Since not all disc priests might be able to put in so many points in haste due to mana issues it might be an option to hang around this turning point where mastery apparently starts to lose a lot more of it's value? Or am I just understanding him incorrectly?

What he claims he did is run a simulation adding fixed amount of each stat and found how much each of them increased his overall healing. That 40% is the point where the value of mastery equals that of other stats. I really can't evaluate it with what is there on his post. I think he made more of an "FYI this is what I found" post, so it all comes down to whether you trust him or not.

I personally don't trust it at all. My own guess on the subject is trade as much mastery as you can for haste, but still reforge away all crit and stack as much spirit as possible. For those interested in looking at how I arrive to that conclusion my thought process is below:

===================================================================================

Basically the scaling for all stats has "diminishing returns".

If you think how much your healing increases as you get more int mastery and crit then a fixed amount of these stats always increases their respective stat (healing done, increased absorbs, chance to crit), by a fixed amount, but as these gets larger and larger the % increase represented by a fixed amount decreases.

For INT the % increase is

%increase_Int = 100*(Int*B*k)/(1+spellpower*k), where spellpower is your current total spellpower, B is a factor depending on SP and int boosting buffs (1.05 unbuffed to 1.33 fully buffed with inner fire) and k =0.000096564716 ~ 0.0001

As you can see the % increase gets smaller and smaller the more spellpower you have.

With mastery

%increase_Mas = 100*Mas/(1+mastery), where mastery is your current mastery expressed as a fraction (i.e. 20% = 0.2) and Mas is the amount of mastery added again expressed as a fraction.

or in terms of mastery rating:

%increase_mast = 100*(mastRat_added)/(16656+mastRat_total)

Crit works the same way

%increase_Cr = 100*Cr/(1+crit)*CF, where crit is your current crit rate, Cr is the amount of crit chance added and CF is the crit bonus factor (i.e. how much extra healing is added by a crit).

Intellect is a better stat for both absorbs and heals as we have shown earlier. For gems in 5.1 whether 2 mastery is better than 1 int depends on how much of your healing is absorbs (including spirit shell) and what % of your heals overheals by any amount (note that this is not overheal). In 5.2 mastery scales with spirit shell in the same way that the basic heals do. i.e. the scaling is negligible for single target heals and poor with PoH. With spirit shell out of the equation it is highly likely that we will be looking at 3:1 equivalence or more between mastery and int.

For PoH crit is a bloody terrible stat compared to mastery and here is why:

Mastery scaling for PoH is (1+0.5*(1+mastery+Mas)/(1+0.5*(1+mastery)) = 1+0.5*mas/(1.5+0.5*mastery)), where mastery is your current mastery expressed as a fraction (i.e. 20% = 0.2) and mas is the extra amount of mastery added expressed as a fraction i.e. (Mastery rating added)*0.025/600

CF for PoH equals 1 (i.e. a crit always adds 100% extra healing), since crits no longer proc extra aegis. Thus crit scaling for PoH ignoring overheal is simply 1+Cr/(1+crit).

Adding some values of 40% mastery and 12% crit which are pretty much the lowest values you can have in 496 gear. Then 600 rating adds 2.5% more absorbs for mastery and 1% extra crit chance, so PoH increases in value by 1+0.025*0.5/(1.5+0.5*0.4) = 1.007352941176 for mastery and 1+0.01/1.12 = 1.00892857142857 for crit. Inner focus is reasonably 1 out every 15 PoHs so that reduces the value of crit to 14/15 of normal. Which means the total benefit is 1.008333. Notice that the values are really close even without taking overhealing into account.

To look at what happens when we trade crit for mastery a real life senario of a disc priest stacking mastery (62% mastery fully buffed), but having zero crit rating (12% fully buffed). Let us now take out 10% mastery and add 4% crit and see what happens to PoH.

PoH value is Base*(1+crit)*(1.5+5*mastery)), so the change in total PoH is 1.16*(1.5+0.5*0.52)/(1.12*(1.5+0.5*0.62)) = 1.007103393844. That means changing 10% mastery to 4% crit increases PoH healing by only 0.7%, before even factoring in overheal. On the flip side the % of the PoH base heal that is an absorb has gone from 90.72% to 88.16% (this includes crit). Now let us think of a senario where you have very frequent, very strong damage spikes. You cast x PoHs at full health followed by y PoHs which won't overheal at all. With the mastery set PoH heals for 3.8016 times the base healing while with the crit set it heals for 3.8372 times the base healing. Those 6 PoHs therefore net you 0.0356x of base healing extra with the crit set but you lose 0.0256 base healing for every PoH you have cast at full health so 0.0256y. The total %gain is (0.0356x-0.0256y). The benefit of the crit set becomes zero when x = 0.72y. That means if you are casting 1.4 PoHs at full health for each PoH which won't overheal at all, the crit set has zero benefit. Any more than that and PoH now has zero value.

In a continuous senario what is the overhealing on crits required to completely remove the benefit of the crit set? The answer is simple. You now have 16% crits instead of 12%. So if a crit overheals more than a normal crit by a factor of O then changing 4% of heals to crit heals increases overheal by (0.84+O*0.16)/(0.88+O*0.12). To eliminate a gain of k overheal needs to increase by a factor of (k-1+0H)/(OH*k). At 20% overheal then an increase by 2.78% in overheal is required to eliminate a 0.7% gain in overall healing. (0.84+O*0.16)/(0.88+O*0.12) = 1.0278 ---> O ~ 1.76

Basically at 20% overheal if crits overheal 1.76 times as much as normal heals do then the crit set has zero benefit. This is almost always the case for anyone whose playstyle involves spamming PoH as much as possible even when healthbars are high.

Thus the benefit of trading mastery for crit is negligible for PoH and vanishes in any senario where you are spaming PoH for the whole encounter. If your play style is to only spam PoH during spikes and do maintenance healing otherwise, you may get some benefit from trading mastery for crit, but even then its not certain. Since no matter what you do crits always have more overheal then normal heals the overall benefit of trading mastery for crit is always going to extremely tiny. Even the extra raw throughput that you get is too small to actually be noticeable. But the biggest problem with crit is uneven healing.

I think if your main spell is PoH trading mastery for crit is never going to work well.

The only time crit is of any consequence is for single target healing in senarios such as the howling cackle groups on the sha of fear or when you actually have to provide off-dps. For tank healing crit is just not a great stat, because most of your heals when you are tank healing overheal to a lesser or greater extent so crits have very high overheal compared to normal heals.

In contrast haste not only has no diminishing returns, but it actually multiplies the value of proc haste buffs like PI and borrowed time and it increases shadowfiend return and weapon enchant procs.

Being able to chain cast is more important to disc that having big heals, so having enough spirit to chain cast the best spells for the job is the number 1 priority. You want to be able to trade all the mastery you can for haste, so having enough spirit to do that is essential.

Consider that the healing formula and the damage formula are different for penance. At 25k spellpower penance used defensively heals for 22-23k ticks.
I just did a quick log and AA stacks multiplicatively with TOF, as Evangelism also stacks multiplicitively with TOF. The log was done with 22.5 spellpower after Inner fire. Tooltip damage for penance shows 24357 average damage tick, 20908 average healing ticks. Smite at 21641, HF 26137 initial and 6080 over time. I am using a +3% crit effect meta gem.

--Compare--
Penance crit:
-- expected damage: 2*1.03*1.2*1.15*24357=69242 (critical*meta*evangelism*twist of fate*average damage)
-- actual log damage: 69404 (+.2%)
-- expected atonement: 69242*1.03*1.25*1.15*.5=51260 (damage*meta double dip*archangel*twist of fate*self heal penalty)
-- actual log atonement: 51381 (+.2%, Notice how these numbers match. The formula is perfect.)

Holy Fire crit:
-- expected damage: 2*1.03*1.2*1.15*26137=74302
-- actual log damage: 73919 (-.5%)
-- expected atonement: 74302*1.03*1.25*1.15*.5=55006
-- actual log atonement: 54723 (-.5%, again, an exact match)

I hope this shows definitively that the damage and healing bonuses, at least archangel, evangelism, and twist of fate are multiplicative bonuses. I see no reason not to expect that touch of the titans is the same. Please understand that twist of fate alone is a 32% increase to atonement healing while active, and as such you will not see any heals above 750% unless that talent is chosen. My 185k penance tick that you pointed out is around 740% (assuming roughly 25k spellpower). This matches perfectly with my predictions.

D hymn including buffs:
5 targets * 7987 (+ 154.2% sp) 4 ticks over 8 sec, +10% healing, +50% touch of the titans, +25% chakra
5*1.1*1.5*1.25*(7987+1.542*25000)*4/8=239956 hps (assuming all targets have the +10% buff)

Offensive penance no buffs:
940 to 1061 (+ 103.7% sp) 3 ticks over 2 seconds
(1000.5+1.037*25000)*3/2=40388 hps

Buffed dhymn to unbuffed penance ratio: 5.94

So in order for penance to have higher hps than dhymn, you would need a combination of buff that increase healing to at least 594%. (shown in red)
Even factoring in echo of light, adding maybe 10%, or 15% more healing. That only raises the bar to 6.5 or 6.8. Or maybe I missed something?

I used to run a haste spec in the final months of DS - will be interesting if 5.2 brings us the same conclusion.


For me - the numbers are interesting but the reality of game play once the patch drops will dictate choice, at least it will mean less pain in swapping specs during raids.

Thanks for the feedback guys

ToF also buffs divine hymn so if you include ToF anywhere in the calculation you need to buff divine hymn by 15% first.

These are my numbers with 25% chakra and inner fire without any other buffs and with whatever gear I happened to have on when I tested.

Divine hymn 54914 per target with 22% mastery, 7.8% crit.

Total is 54914*1.22 *(1+0.078) = 72 221

Apply ToT and you get 108 331.5

DH = 108 331.5*5/2 = 270828.75 HPS for 8s and applies a 10% healing buff to all targets, which increases HPS done by the priest and other healers (a utility not shared by penance used offensively). For extra benefit it can be chained with power infusion, an intellect trinket proc (WoE trinket for example) and/or and intellect potion.

With ToF DH DPS is 311 453

With PI 324 994.5

With an intellect trinket proc (~3.5k intellect) and power infusion DH HPS is ~350k.

This is before any raid buffs and consumables which increase overall healing by at least 10%.

DH generally produces close to 400k HPS. In that phase.

Penance does 25172 damage without evangelism and before any buffs and I have 44% mastery with 7.8% crit

25172*(1+0.078*(1+0.5*1.44))3/2 = 42824 HPS through atonement. It is impossible to reliably chain an int trinket proc with all the buffs you need to maximal penance output.

Unbuffed penance to DH ratio is 6.3 without factoring in talents, procs and stuff.

Unbuffed penance to DH ratio with ToF is 7.3

Unbuffed penance to DH ratio with power infusion is 8.

In 10 man the healing buff is almost certain to add 5% overall to DH so all these are 5% higher.

More important you cant really compare the 3 heals from penance over 2s with the 25 heals from divine hymn over 8s. To get the big penance you need to spam low HPS spells before and after so you produce a shallow spike. Holy spams high HPS spells gets a massive heal from DH and then gets a hefty 10% buff both for its own heals and the heals of other healers, so it produces a massive spike.

A combination of PoH/PoM/Divine hymn from holy ends up being considerably higher HPS than pure smite spam from disc and more importantly it ends up being more powerful for saving people, because you are providing a blanket heal to the whole raid instead of a massive heal to a couple of people.

Its pretty hard to find anyone who is using ToF and smites through the last phase in the top 10 disc parses in WoL. The reason is this:

Healing done - 19-12 19:35 - Ktjn Logs - World of Logs

Look at what happens when the priest spams PoH at the very end instead of atonement.

The rank 3 parse for disc looks like this.

Healing done - 03-12 21:10 - Shin - World of Logs

Notice the holy priest hitting a nice 400k spike due to divine hymn and the 10% buff.

Atonement spam is not as good as geting 5 stacks for AA --> PoH spam.

I tested SS on PTR and formula for PoH is exactly as it should be:
SS_PoH=Avg*(1+Crit%)*(1.5+0.5*Mastery)

If you regem and reforge from pure mastery to Int/Crit your SS part of HPS will drop by ~3-5% and other part of HPS will rise significantly.

Solace is imbalanced right now.
Pros:
1) It makes HF instant
2) It makes HF free
3) It makes HF restore mana
4) It will give you 2 GCD in 3min (you will summon SF instead of MB)
Cons:
4) If you never cast HF before, it makes you do so =)

In terms of mana regen:
Mindbender (11 attacks, 1.43% mana each, 3 summons per 3 minute) = 11 * 1.43% * 3 = 47,19% mana per 3 minute

SF+Solace (9attacks, 3% mana each, 1 summon per 3 minute, Solace on CD = 18 times per 3 minute) = 9 * 3% + 18 * 1% (mana regen part) + 18 * 1.8% (free mana cost part) = 77,4% mana per 3 minute

So, MB talent gives you estimate 47% mana per 3 minute and Solace gives you 77% mana per 3 minute.
Solace is nobrain choice for Disc right now.

I think Blizzard balanced solace w/o free mana cost part. Mana regen for Solace w/o it per 3 minute:
9%*3+18*1% = 45% mana per 3 minute which is rather close to MB. If you never ever cast HF (but you really shoud) you will take MB. If you cast HF on CD your nobrain choice is Solace.

I think solace is balanced around the logged use of holy fire for priests.

The real question here is whether holy fire is now 1.5 cast or if its still 2 sec. Since there is no longer a glyph to make it instant taking solace reduces HF cast time by 0.5s. If that is the case the either you have to spend more mana casting smite or you will also lose 0.5s everytime you cast holy fire.

If it is 2s without solace then its an absolute no brainer to take solace. You only need to cast

47.19 - 18 = 39.19 --> 39.19/3.8 10.38 i.e. 11 solace per 3 minutes to beat the mana return from mindbender. That equates to 1 solace every 18s, which you less than you need to for keeping evangelism stacked. Since you will have to cast 11 holy fires with mindbender then overall you are also gaining 0.5*11 + 3 = 8.5 seconds of casting time per 3 minutes. If you choose to use smite to save time instead you will use an additional 0.7% mana each time, which makes the mana gain from solace even more significant.

==================================================================================

Comparing reforging to haste vs reforging to crit

My values: ~45% mastery, 5.25% haste, 7.56% crit unbuffed

Buffed: 57.5% mastery, 10.25% haste, 12.91% crit

From the character sheet: 2418 haste --> 5.49% so 425 rating for 1% haste
600 crit --> 1%
600 mastery --> 2.5%

At ~6000 mastery reforging allows to convert 2400 rating to crit = 4% or to haste = 5.65%

Final buffed value after reforging

crit: 47.5% mastery, 10.25% haste, 16.91% crit

Haste: 47.5% mastery, 15.9% haste, 12.91% crit

First of all lets see what happens to shadowfiend:

9*1.1025 = 9.9225, so still 9 attacks

9*1.159 = 10.431 so 10 attacks instead of 9, meaning an additional 9000 mana every 3 minutes.

Estimated mana return is ~16000 per 5 seconds for my spirit & Spirit procs + 3750 per 5 from solace/fiend + 700 mp5 from hymns + ~600k over 6 minutes from rapture is ~ 1.8million mana. The additional mana from shadowfiend haste is 18000, which is ~1% more mana.

In contrast 4% crit increases the maximum HPM of atonement and all single target heals by (1+0.1619*1.475)/(1+0.1219*1.475), which is 5% before overheal. It increases the HPM of PoH by 3.57%

Factoring in inner focus for greater heal and flash heal at 1 use every 35 seconds or on average 1 in 17.5 heals with 2 PWS casted means an overall increase in crit of 5.7%. That reduces the HPM contribution for single target heals to 4.7% and for PoH to 3.4%

Reforging to crit instead of haste results in having 1% less mana to spend but it increases all single target heal HPM by 4.7% and 3.4% PoH HPM before overheal. Since crits have inherently high overheal the actual benefit is likely to be much smaller, but for now lets use the maximum.

Haste stacks multiplicatively with inner focus and borrowed time:

5.69% from haste rating. Adding 15% from borrowed time --> 21.54%, 1.0569*1.15 = 1.215435
Adding 20% from power infusion --> 26.83, 1.0569*1.2 = 1.26828
Adding both borrowed time and power infusion 45.85, 1.0569*1.15*1.2 = 1.458522

With additive stacking it would be 20.69, 25.69 and 40.69.

In other words the value of haste is unaffected by temporary haste buffs.

The HPS benefit from reforging to haste is 15.9/10.25 = 1.051247165533 for all spells and it completely unaffected by overheal or temporary haste buffs. However it also means I will need to cast 5.1% more spells meaning I need 5.1% more mana. Shadowfiend return increase by 1% potentially means I only need 4% more mana.

For spirit shell specifically with the extra haste PoH cast time is reduced from 2.27s to 2.16s. 15/2.27 = 6.6 2.5s casts but 15/2.16 = 6.94 casts.

Theoretically there is no increase. Adding PI reduces the cast time to 1.89 or 1.8, which means 7.94 casts with the crit set or 8.34 casts with the haste set. I.e. 8 casts instead of 7 or an increase of 14%, which is fairly dramatic considering that crit only adds 3.4%.

More importantly even when PI is not present casting PWS just before activating spirit shell will reduce the cast time of the 1st PoH to 1.972 or 1.88. That means you can now cast 1+(15-1.972)/2.27 = 6.74 2.5sec casts or 1+(15-1.88)/2.16 = 7.1 PoHs.

7 PoHs instead of 6 (and 0.1s allowance for lag) or an even more dramatic increase of 16.7%

Using both PI and PWS however allows 8 casts in both cases.

The more haste you have the easier it is to guarantee extra casts during spirit shell, or you can decrease the length of time you use spirit shell for before a spike and add some more aegis before it instead.

Since spirit shell is no longer an actual HPS increase, this is rather unimportant under sustained damage conditions.

Thus haste is dramatically higher value than crit for PoH spam and it is the best way to mitigate the loss of mastery scaling from spirit shell. Using your haste buffs intelligently will actually allow you to get more out of your spirit shell.

==================================================================================
Intellect versus spirit. With the increase of rapture to 250% of spirit and the need to stack haste the value of spirit for disc has skyrocketed.

Lets have a look.

Assuming you are at the point where PWS is mana neutral you now gain 2.625 mana for every additional point of spirit just from rapture. Assuming one rapture proc every 18s that means ~0.73mp5. Reducing my spirit by 1447 reduces combat mana regen by 817mp5. thus 1 point of spirit adds ~0.565 mp5. Correcting for the approximations that sums up to 1.294 mp5 per point of spirit. For disc the approximate average cast time is 2/haste or 1.75 or so for me after reforging to haste and average cost of spells is 11k when aoe healing or 13k when single target healing. That means in 5 seconds I spend on average 31.5k mana aoe healing or 37.2k mana single target healing. 1.294 therefore represents a decrease in mana cost of 0.0041% for aoe heals or 0.0035% when single target healing. This equivalent to an increase in HPM of 0.0041% and 0.0035% (because the values are so small 1/(1-x) ~ x). This is not affected at all by overhealing.

For intellect

%increase_Int = 100*(Int*B*k)/(1+spellpower*k), where spellpower is your current total spellpower, B is a factor depending on SP and int boosting buffs (1.05 unbuffed to 1.33 fully buffed with inner fire) and k =0.000096564716 ~ 0.0001

Using B = 1.05*1.1*1.1*1.05 = 1.334025 (inner fire, 10% SP buff and 5% stats buff) the formula tells us that at 27000 spellpower

1 point of int increases healing by 100*( 1.334025*0.000096564716)/(1+27000*0.000096564716) = 0.00357% before overhealing.

That means 1 spirit is now better than 1 intellect for disc and the more your overheal the better spirit becomes.

That tells us that for gems and consumables spirit is best. For alchemists spirit flask is a no brainer.

It also tells us that stacking haste and spirit is now the best gearing option for disc. A value of at 15k spirit before any procs would probably allow to chain cast with ~15% haste and be at roughly the same HPS as we were before with mastery stacking.

The new stat weights for disc are now 1 spirit > 1 int > 2 haste > 3 crit > 3.5 mastery. So basically if the gem bonus is is anything other than 120 spirit/int then gem spirit, since 100 spirit is better than 200 haste, 300 crit or 350 mastery.

Havoc, I think you calculated your value of Crit in a funny way. At your numbers, it should give you a 7.6% increase in healing (though does not change things for PoH of course). I'm also not convinced that crits contribute massively to overheals. Intuitively, it's an appealing notion. Practically speaking, however, I've looked at my ratio of crit/normal effective heal amounts for the last several weeks, and they're typically very close to 2 (and, in fact, more often above than below, between the crit meta and int procs raising both the chance to crit and the amount healed). Crits are particularly unlikely to overheal during periods of heavy damage, when healing is most needed. Of course, an argument can be made for Haste's reliability, but given the number of smart heals flying around in the typical raid (and the fact that, due to defensive abilities, people will baseline often be taking significantly different amounts of damage) I'm not convinced.

You've also left out the increase in Crit due to Int.

Finally, in your Haste and Spirit calculations, you're assuming that all spellcasts are equal. Because some of our spells are better than others, and the better ones tend to have CDs, being able to cast more due to more mana or more haste often means casting crappier spells, and thus they don't necessarily scale as linearly as you've proposed. Having more mana or more Haste does nothing for Cascade, PoM, Penance (unless you're casting Smite), Spirit Shell (except when crossing a breakpoint in the number of spells you can cast) or the healing output of PW:S.

A crit is a double so the amount of aegis added is 0.5 of a double heal, which is equal to a normal heal.

A crit is basically a 2x heal and 0.5*2xheal*mastery heal as aegis this works out to a total crit modifier of 2+1*mastery. Compared to a normal heal this adds 1+1*mastery extra healing, so the calculation is base healing + (1+1*mastery)*base for every crit. The way I calculated it is correct.

I included crit from int. I have 0.56% from rating everything else is int and I factor in the 5% buff and then 7% buff

PWS does not have a CD and it is less HPS than PoH, PoM has a CD but its even less important than PoH. Haste can dramatically increase the amount of healing you can do because you can increase your PoH to PWS ratio and it dramatically improves PoM since you can now manage to squeeze it in more easily. That means your actual healing can see a bigger increase than you expect just from haste.

But even with cascade which does a CD haste can add dramatic improvents. It can mean that you lose less CD time or that its easier to use it optimally. Being late with cascade because you had to stack that extra 5th stack of evangelism can result in a big loss of healing. But that is difficult to digest the most important thing is windows. Let us say that you could cast 7 PoHs and 1 cascade without haste before everyone is full up and your 8th PoH overheals by 80% because it landed 0.5s too late. If with haste you can squeeze in an extra PoH a little earlier and it overheals by 40% then assuming cascade heals twice as much as PoH (It doesn't just illustrating a point here) then you have the equivalent of 9.2 PoHs without haste or 9.6 PoHs with haste. That is a 4.3% increase. Assuming 2.3s cast time for PoH and 1.4s cast time for cascade to gain an extra 0.5s you only need 2.9% haste. I.e you see a 4.3% improvement for just 2.9% haste.

This is how haste can result in non-linear increases, especially since you can proc borrowed time effectively at will.

As for crit not overhealing more, you are making an important error I think. First of all close to 2 is not 2, the difference can be quite dramatic, here is why:

Lets assume you have 60% overheal on normal heals and 60% overheal on crits. If that is the case then you will have a strict 2:1 relationship on the meters. First of all inner focus adds a bias on crits for low health targets. How important is that bias depends strongly on how frequent your crits are. Here is an example:

10 PoHs with 5 targets = 50 heals so 5 crits with 75% overheal and you got 5 crits with 0% overheal. the average is just 35% overheal. For normal heals you got 50% overheal, however your benefit from crit was zero. If you had 50% crit then you would have 25 crits with 75% overheal and 5 crits with zero overheal so 62.5% overheal.

So low crit rates makes inner focus add a significant bias for crits on low health which reduces overall overheal for crits, but that also receives zero benefit from crit rating. Adding 10% crit rating here will just increase overheal and give zero benefit.

Look at this for example: Details for Th�*llo - 11-12 15:33 - Ellui - World of Logs

Look at PoH crits having nearly the same overheal as heals, but look at cascade look at atonement. The reason is inner focus adding a bias for crits at low health. Otherwise it is mathematically impossible to have overheal rates of nearly 60% and equal overheal on crits and normal heals. Since crits are 2x bigger then in order to get the same overheal they would have to magically have a preference for low health targets. Its actually not magic its inner focus and its specific to PoH. That does not mean crit is good. It makes crits even worse.

Lets check if this is true for the same log by looking for inner focus PoHs

[21:45:00.553] Th�*llo casts Inner Focus
[21:45:01.269] Th�*llo begins to cast Prayer of Healing
[21:45:03.624] Th�*llo's Inner Focus fades from Th�*llo
[21:45:03.712] Th�*llo Prayer of Healing Th�*llo +*64981* (O: 7405)
[21:45:03.720] Th�*llo begins to cast Prayer of Healing
[21:45:04.001] Th�*llo Prayer of Healing Req +*73093*
[21:45:04.001] Th�*llo Prayer of Healing Primal Fire Elemental +*0* (O: 72722)
[21:45:04.001] Th�*llo Prayer of Healing Tiberria +*72976*
[21:45:04.001] Th�*llo Prayer of Healing Primal Fire Elemental +*0* (O: 72419)
[21:45:04.001] Th�*llo Prayer of Healing Enragedwife +*6987* (O: 65640)

[21:48:30.223] Th�*llo Prayer of Healing Saburro +*63347* (O: 17386)
[21:48:30.223] Th�*llo Prayer of Healing Wýcked +*81257*
[21:48:30.223] Th�*llo Prayer of Healing Xaruki +*81383*
[21:48:30.223] Th�*llo Prayer of Healing Shadjr +*80681*
[21:48:30.223] Th�*llo Prayer of Healing Ellui +*51560* (O: 46058)

[21:49:40.465] Th�*llo Prayer of Healing Studliest +*0* (O: 76811)
[21:49:40.465] Th�*llo Prayer of Healing Attittude +*0* (O: 76227)
[21:49:40.465] Th�*llo Prayer of Healing Knatt +*0* (O: 76481)
[21:49:40.465] Th�*llo Prayer of Healing Eni +*0* (O: 76418)
[21:49:40.465] Th�*llo Prayer of Healing Verotha +*0* (O: 83822)
[21:49:42.928] Th�*llo Prayer of Healing Saburro +0 (O: 35081)

[21:51:09.244] Th�*llo Prayer of Healing Th�*llo +*81355*
[21:51:09.320] Th�*llo begins to cast Prayer of Healing
[21:51:09.583] Th�*llo Prayer of Healing Req +*41682* (O: 39874)
[21:51:09.583] Th�*llo Prayer of Healing Tiberria +*81363*
[21:51:09.583] Th�*llo Prayer of Healing Primal Fire Elemental +*0* (O: 81338)
[21:51:09.583] Th�*llo Prayer of Healing Enragedwife +*81491*
[21:51:09.583] Th�*llo Prayer of Healing Zechz +*81713*

[21:51:59.324] Th�*llo Prayer of Healing Th�*llo +*68795* (O: 7980)
[21:51:59.568] Th�*llo Prayer of Healing Req +*49101* (O: 26975)
[21:51:59.568] Th�*llo Prayer of Healing Primal Earth Elemental +*0* (O: 76517)
[21:51:59.568] Th�*llo Prayer of Healing Tiberria +*76589*
[21:51:59.568] Th�*llo Prayer of Healing Primal Earth Elemental +*0* (O: 76053)
[21:51:59.568] Th�*llo Prayer of Healing Enragedwife +*76210*
[21:51:59.570] Th�*llo Prayer of Healing Zechz +*76395*

I think its obvious that inner focus crits have a lot lower overhealing and I just linked 30 out of the 84 crits here.

So the overheal of crits for PoH is artificially lowered by inner focus. already here 30 out of the 84 PoH crits would not benefit at all from an increase in crit rating and the rest would probably have massive overheal. If a heal overheals by 50% or more then the only benefit of crit is divine aegis. The rest of hte crit overheals by 100%