Usually I do add in earth shields, not so much healing stream and it counts lifebloom end ticks err nevermind those get accredited to the people who it proc'd on.

I'd like to know what posters here think of a personal stat weighting method I'm working on. This is designed to figure out my own weighting between mp5, crit, and +heal, in order to make gear choices (such as prioritizing badge gear purchases).

1. I picked a fight that in my opinion stresses my healing stats; that is, where I feel that more healing from me would be useful. I know most people who post here are much more progressed than my guild, but we're 5/6 SSC and 3/4 TK. I chose the Fathom Lord Karathress fight. I'm the Caribdis group healer; I'm always out of mana at the end of the fight, and I think I would make use of more healing throughput as well. This and Morogrim are the only fights that really stress my healing throughput and regen; the others in SSC and TK are mostly strategy and positioning (or perhaps dps, but certainly not lack of healing stats).

2. I loaded up a WWS report for that one fight and looked at my use of healing spells. There were a few guesses and assumptions I had to make. I no longer have this combat log, but when I get the procedure right and do this again I will grep the log to actually count casts in the fight rather than having to guess. In this case I assumed that a) Each of my renews ticked an average of twice (I think I overuse renew); b) All prayers of mending on the Caribdis tank are attributable to me and no others are (this is probably a bad assumption); and c) My circles of healing and prayers of healing hit an average of 4 people each. I also had to guess how many of my greater heals were GH3 and how many were GH7, since the average GH according to WWS was about halfway between my GH3 and GH7 averages, I assumed I cast an equal number of each. I combined it with some data from Jayde's spreadsheet to get the following table. I would like to point out, also, that the precision of these numbers is not commensurate with their accuracy.

The first two columns are from WWS. The number of casts are based on the assumptions above. The mana costs are from Jayde's spreadhseet. The total mana is the number of casts times the mana per cast. The heals per mana is the total heal divided by the total mana. Overheal is from WWS (I'm not using this information at all; I'm not sure how I should). The next two columns, the amount of healing that one more point of +heal or +crit would add to the cast, is from Jayde's spreadsheet (uses my stats and talents). The last two columns are the number of hits during the fight multiplied by the previous two columns. So the last two numbers in the bottom right of the table should be, if I've done this right, the total amount of healing over the course of the fight that I would get if I had one more point of +heal or one more point of crit. So in this particular fight, every point of +healing would have added 174 healing, and every point of crit would have added 75 healing. The fight lasted 402 seconds, or 80 5-second ticks. One point of mp5 would then have given me 80 mana. My overall casting style in this fight resulted in 8 heal per mana. So one point of mp5 would have given me a total of 640 extra healing.

This gives: 1mp5 worth 640/174 = 3.7 +heal
1 crit worth 75/174 = 0.43 +heal

According to the table I used 65300 total mana. I started with 12388 mana, so I regenerated at least 52912 mana over the course of the fight. My insightful earthstorm diamond returned 4359 mana, my shadowfiend gave me 4124, and spellsurge 300. I got clearcasting twice; I'm going to assume I used them for greater heal 7 for a total of 1650 mana. My Eye of Gruul procced 4 times for a total cost reduction of 1800 mana. I don't see any mana potions in the WWS report (should they be there?) but I'm pretty sure I used 2, so let's say 4800 mana from that. That leaves 35879 mana regenerated through standard regeneration means for the rest of the fight. Since I get 267 in the five second rule and 850 outside, that works out to 21% outside the five second rule. I think this is an overestimate, actually. I suspect my assumptions about renew and PoM overestimate my mana usage.

Anyway, at 21% oofsr, from Jayde's spreadsheet again (which incorporates spirit's contribution to both healing and mp5), I get spirit worth 2.48 +heal and intellect worth 1.8 +heal.

I don't have any way of weighting haste as yet. I also value stamina and for now am just using an arbitrary number for that.

If I decide I like this method, I may extend it to averaging multiple fights. Now that I have the infrastructure in place to do the calculations, I can do it for a number of fights and just take the mean stat values. Or even make minor switches to gear between fights, if different fights require different stats. Also, using a boss we have pretty much on farm (though it takes us a couple of attempts, usually) may not be the best idea, but again, the progress fights aren't failing due to lack of healing, but due to us still working on strategy issues etc.

Thanks for any feedback.

I do not like the method you have used on many things.

1) Assuming equal casts of GH7 and 3 is far from ideal. Consider this.

GHav = (N7*GH7H+N3*GH3H)/Ntotal, where GHxH = average heal from GH rank x, Nx = number of GHx casts, Ntotal = total number of GH casts and GHav = GH average from WWS reports. You must include crit in the calculation for GHxH and use total healing including overheal. Its not going to be 100% accurate, but it will be close enough.

Thus N7 = [(GHav*Ntotal) - N3*GH3H]/GH7H and N3 = [(GHav*Ntotal) - N7*GH7H]/GH3H.


2) The use of Jaydee's spreadsheet weighings here makes no sense to me. Why are you using a stat weighing model, that includes factors other than the actual amount healed to predict how much healing you got from your crit? You should look at the data directly. WWS gives you this info. Just increase overheal by 1.5 for crits. My own value (from having an extra +3 crit from buffs) is that 1 crit adds 0.3% healing on critable spells. This is very different from the value you have got.

3) Instead of using WWS reports alone, I suggest using a real time damage meter. I remember that SWstats calculates your total HPM (i.e. not effective HPM, it includes overheal). For each of your spells. This will allow you to accurately determine how much mana you spent on each spell. You can also calculate your effective HPM by simply adding in the overheal. Also recount and PoM tracker will give you a more accurate account of how much healing you got from PoM.

I personally do not like the method you have used at all. It just tries to tailor your gear closer to your healing style and arrives at values that are only true given not only your current healing style but also your current gear.

In my oppinion the best way to determine what you need the most. Is to calculate your average effective HPM for a fight, your total regeneration and overheal then determine how much healing you can push out. Set up a total healing target and see how much regen and +healing you need to reach that target. That should give you the most accurate prediction of what stat is most effective for you at each particular point in time, but more importantly of what you can do with your current gear. That is the only way to determine how much haste you can squeeze into your gear, without compromising your output.

Note: I have often healed caribdis and I have never had the benefit of having the tank and 2 DPS in the same group as me. In fact in most cases every member of the team is in a different group. Without the benefit of being able to use CoH, I have found that 2x binding heal, one renew, PoM and one gheal rank3 is the sequence I require to top up everyone after the water bolts, combined with rank 3 GH and the occasional rank 6 gheal on the tank. The group composition I have is warrior tank, rogue, elemental shaman. I have healed this senario with a fury war tanking, who was taking 1.2-1.5k DPS from caribdis, making the encounter quite hard. I generally keep renew on the rogue and ensure he has a PoM on (bounced from the tank) when a water bolt hits. So all I have to do when water bolt hits, is BH the shaman (to prevent insta gib from chaos bolt on him and me), gh3 or 6 on tank, bh the shaman and let renew heal the rogue (he has already healed from PoM). I respond to chaos bolt hits with FH/PWS, if a water bolt is imminent or gheal rank 7 if a water bolt has already hit. A possible alternative strategy is maintain 2 renews (shaman rogue), keeping a PoM on the rogue (after tank bounce) and going BH shaman, BH tank after each water bolt then rank 3 on tank. Yet another senario is keeping 3 renews on yourself, rogue and shaman, then going BH shaman, gheal rank 3 on tank after each water bolt.

When the raid arrives, all you need to do is keep PoM on yourself and PoH/CoH everyone after each water bolt. That leaves you a lot of ooFSR time.

The new alche trinket and the earring of soulful meditation (if you dont have them) are your best choices for finding extra mana in this fight.

Havoc, I am in need of your help.

I remember when I first brought my haste thoughts to this thread and the previous one you definitely were significant in me figuring ways to do the calculations (I assume you are just skilled with mathematics). So I have a couple questions for you.

1) How do I take crit into account for healing for spells. I am wanting to find out the average amount healed by a spell per cast, including crit.

2) I am working on developing a model that shows how to break down each of the gems in terms of healing throughput. So socketing a [Teardrop Crimson Spinel] vs. [Gleaming Lionseye] vs. [Quick Lionseye] and the effect that 22 healing/10 haste/10 spell crit has on your HPS/Throughput.

(I have not had time to check this over with my roomate who is a math major, however he is also a DPS warrior and probably could care less about healing.)

If you can help, that would be super. If not, I understand.

I'm no math genius but I think I can tackle this one.

To work the formula for this one out, imagine you heal for 1k noncrit and 1.5k crit, and you have a 10% crit rate.

That means that over the course of 10 heals, you can expect one of them to be crit. In other words, over 10 heals you'll do 10.5k healing. This means that crit has contributed 0.5k/10 to your average amount healed per cast.

More generally this arrives at the conclusion that your average amount healed per cast is going to be:
noncrit amount + (crit amount - noncrit amount) * crit %

I don't think that you can evaluate crit quite that cut and dried given it's not-dependable nature. That's a reasonable assumption for a paladin given how much crit is given importance, but for a priest all direct heals must be assumed to non-crit. As such, most crit is going to be overheal, far more than normal healing is. This is especially true on CoH, where you have the following cases:

1) if you needed to do just one CoH, any crit is going to be overheal. Reasoning being that one CoH would have covered the group enough, so more of that would have been extra.
2) if you needed to do multiple CoH, the crits will help heal faster but not heal any more. The reason being that you will need to do multiple CoHs for many people anyway, and the crit will just heal one person in that group more quickly. The others in that group will still need to be healed, making the CoH on the critted person wasted. Again, overheal.

In the best case, crit's value can be determined by (crit rate*crit amount*HPS), but that's the absolute best case.

Kalbear, I am not sure where you are in progression (this is by no means a mockery) but in my current raiding setup and with the healers we bring each night to the sunwell fights, so far as I can see I love spell crit and would take more of it.

1) Almost always I have to CoH a group more than once.
2) If by chance I crit on one or a couple of the people in my CoH cast and 2-3 are at full HP, I move to another group. If 3 or more still need healing I'll just CoH them agian.

average heal = (1-x)*(average non crit heal) + x*1.5*(average heal) = (1+0.5x)*(average non crit heal), where x = crit chance (i.e. if 10% crit x = 0.01)


A = total amount (including crit) healed by a particular heal at a particular value of +healing
B = base cast time or GCD
c = 1/1570
T = converted cast time or GCD
S = spell haste rating
H = +healing additional to what you already have
k = constant converting +healing to actual healing for the particular spell

T = B/(1+cS)

HPS = (A+kH)/T = (A+kH)*(1+cS)/B

Calculating k and A has to be done separately for each heal, but you need to include all modifiers, including spiritual healing and set bonuses and crit. Just use the (1+0.5x) modifier to add crit.

Plug the values in this formula in a spreadsheet and then you can see how the relative values fluctuate depending on your current value of haste and +healing.

My experience with crit is that at 0.3% more healing per point of crit and at 20odd rating per crit point, makes geming for crit counterproductive.

We're in 5/5 7/9 right now. And I'm not saying that crit is bad or that you shouldn't want more - only that a lot of it is wasted. In your second example for instance, the crit is useful because you healed someone faster; you end up casting the same amount of heals on the same group. In that case the actual crit for the heal isn't overheal but the effect of the crit is overheal.

Like you said, if you get it 'by chance' it's great. It can heal faster and more efficiently. But because it is a chance, it isn't dependable. You can't rely on healing 2000 damage done to each member of a group using one CoH, which means no matter where you get that crit, you're going to do more overheal.

That isn't necessarily bad; overheal isn't a bad thing a lot of the time. But it's not a good thing to look at spell crit's value and automatically assume that it's going to be more productive than healing or haste, particularly on something like CoH where the coefficient for healing is poor on a per-individual basis.

Go back about 10 pages and read from there if you want a discussion of crit. We did this already; there's two schools of thought, and they're both defended there.

Crit is really all about Inspiration up-time, not HpS increase. Ya, the HpS increase can save tanks occasionally, but depending on it as a priest is a bit too risky. We're not paladins running 40% HL crit ... we run 10-14% crit, typically, depending on your choice of talent points in Holy Spec.

Edit.

Tyaz, it is viable. Though much better used in the hands of DPS caster classes, not to mention it would probably upset them.

I'll happily cheer the day I see a holy priest get one of those trinkets. It's not going to happen anytime soon. Here's some points to consider:

1) Trinkets aren't guaranteed drops from M'uru, and are hence quite rare (low % drop rate).
2) There are typically 8 dps casters in a raid (sometimes up to 10) on M'uru, all of whom wish this trinket, as it is BiS for every single one of them (I believe; it is for shadow priests, mages, and ele shamans, probably warlocks too).
3) Holy priests come after all of the dps casters.

So yes, if you get 10 of them to drop, sometime around Christmas, you might get one. Might. Or someone's alt dps caster might get it. To this day, I still haven't managed to get a [Hex Shrunken Head] for my shadow kit; we still have 2 casters who need one before I can take it when/if it drops. And that's from ZA, a 3-day reset instance that we clear *twice* every reset for just that purpose.

So ya ... I'm not really concerned with it. It's an incredibly unlikely occurrence for a holy priest to end up with one.

Here are some interesting calculations for those who are interested. I have used this formula:

H>= m*HPStg*Ft/( k*(mana pool + (mana regen)*Ft)(1-O) ) - A/k

Which determines the minimum +healing (H) required to maintain target HPS (HPStg) over fight time Ft, with a spell that has mana cost (m) base healing (A) and conversion factor (K --> +healing to actual healing, including all modifiers), given a certain mana pool, mana regen (in mana per second) and overheal (O).

Note that the formula is casting time independent, so it assumes that you have enough haste to maintain HPStg with that particular spell at this minimum +healing.

The purpose of this formula is to provide a more accurate way of linking mana regen, +healing and mana pool based on real fight conditions.

I calculate mana regen (as mana per sec) as follows: Assume 1100/400 regen and 10% ooFSR (includes time from clearcasting - 0.9*400/5+1100*0.1/5) add potions with alche trinket (2400/120*1.4) add shadowfiend (55/5), add inner focus (705/3/60) and add earring (I equal it to 1500 mana every 3 mins so 1500/3/60).

Mana cost includes clearcasting and talents.

Base heal includes all modifiers including crit, talents and t6 set bonuses.

Gheal rank 7
Based on WWS reports no more than 900-1000 HPS is required from single target healing. Hence setting HPStg = 1000


1) Mana regeneration (mana/second). Hred is +healing requirement reduction. Hred/1mp5 is Hred per 1 mp5.
2) Mana pool (in increments of 10 int). Also shown is Hred per point of int.
3) Overheal (in 5% decrements). Also shown is Hred per % of OH

Lets also have a look at Gheal rank 6 and CoH.

Gheal rank 6
Circle of healing
There are many interesting conclusions to be drawn from this, but in my oppinion the most interesting one is just how viable rank 6 is as an alternative for rank7. I have actively begun replacing rank 7 with rank 6 as much as I can and I am definately noticing a very large benefit in mana saved.

Another important point iswith 2500+healing and 1100/400 regeneration its perfectly possible to exceed the healing output required by any fight.

Constant,

To your main page when you have the time what do you think of the following stuff:

Adding [Hammer of Sanctification] and [Book of Highborne Hymns] to the BiS for main hand/off hand combo, and comparing those two to a Golden Staff.

Adding [Battlemaster's Alacrity] to go along with the +88 healing one. Also noting how good 40 spell haste is for certain encounters (especially for CoH priests). I think it would be important to know that the clicky use is also very effective on Sunwell fights due to how mana intensive they are, the clicky part can be used as another means of gaining health, along with a healthstone. (This is assuming our potion is always on cooldown due to chain chugging mana)

Lastly, as for cookie cutter specs, what do you think about changing your 20/41 so that the 20 points in Holy are with Absolution, instead of without Absolution.

Also here is a quick guide/list of ways to use lightwell effectively (since getting raid members to use it is like teaching a dog new tricks), and on which fights it is effective for.

Lightwell - Since the change in 2.2 it gives 100% of your +healing. 2361 health over 6 sec., given that many BT level priests are at/over or around 2000 HSE we are looking at ~4400 health in 6 seconds (3 ticks of near 1500 hp). Not to mention Lightwell gets +10% healing from spiritual healing so would be closer to ~5000 healing in 6 seconds.

The shitty part about this skill is it breaks on ANY type of damage (cheer for WotLK fixing this). Now on the other hand, does your raid DPS bandage? If so, then they should be taught to use a lightwell. It provides 5 faster, more powerful bandages. Here is a quick list of fights Lightwell is viable for (5/5 is very useful, 3/5 moderately, 1/5 slightly, 0/5 not)

Naj'entus - When Naj bubbles is a great time to use lightwell (this should be pre-emptively dropped) - 4/5
Supremus - Drop it for the first chasing phase, should be up later in the fight (dependent on your raids DPS) - 5/5
Shade - Don't really need healing - 0/5
Teron - Not too viable, could be used but randomness of damage makes it not too viable - 1/5
Gurtogg - Again lots of raid damage, but if timed right can be used - 1/5
RoS - Can be great on phase 2 - 3/5
Mother - So much random raid damage, tough times to click it (maybe those running back from FA) - 1/5
Council - It can be either very good on this fight, or very bad due to the randomness of raid damage - 3/5
Illidan - I actually use lightwell on this fight more than any other. Very good, multiple usages - 5/5

Kalecgos - Too much raid damage being taken, only viable in the nether realm - 1/5
Brutallus - Great for healing up a 3rd meteor slash - 4/5
Felmyst - Too much AE damage to make it even practical - 0/5
Twins - Again too much AE damage to make it practical - 0/5

That's as far as I have been.

I agree it is not the best talent, but I think with the changes it is getting in WotLK it will be much better. 10 charges, faster cooldown, only breaks on direct damage.

Imagine learning Archimonde with 2 holy priests (probably 4x lightwell throughout the encounter) and being able to click a lightwell if you got the doomfire debuff. I think the changes are a step in the right direction, but the spell could still use more love (more charges imo, or no cooldown).

I don't think you can model the healing required by using the DTPS from WWS.

On m'uru, our sentinel tank takes ~ 1900 DTPS throughout the fight, which could possibly be healed by a single healer, but at times the tank will be subjected to ~5000-6000 burst DPS for short periods of time, which means we have to have a second healer to mitigate this damage (in our case we use a resto druid rolling hots).

I don't see how this could work, unless they stop dps'ing for 6 seconds?

I haven't stepped foot in SwP yet, but it looks much more viable there than in BT. We seldom leave Naj'entus's shield up long enough to get more than 1-2 ticks off a lightwell, and that's assuming you were already standing right next to it.

I first picked up lightwell when I was leveling in hinterlands and used it extensively while leveling and in pretty much every 50+ non raid/heroic instance in the game.

I actually love having this spell when I am instancing/grinding as holy. It is really very useful. Exactly because I know this spell like the back of my hand I respecd out of it the instant I entered karazhan and have never included it in any raid spec I have tried.

Although it may seem that you can put lightwell down before the fight. You must remember that lightwell only lasts 3 minutes.

Lightwell in its current form is much less efficient than any other talent it competes with and its not very easy to use.

Only in 5mans where fight durations are very short and you are the only healer does lightwell have a purpose.

Its also very useful when you are leveling as you can reduce your downtime a lot, but most ppl level as shadow

Lightwell either needs to have no CD or not break on damage. Increasing the charges is not going to make it really useful. The reason is that lightwell is best used as a passive healing solution, to do that you cant afford to have a CD.

Turgid, I'd rather them stop dpsing for 6 seconds than kill themselves. Also during that six seconds they could get double the effect and bandage too.

Or you could just heal them in other forms and they don't die nor stop DPSing. Mana is only an issue in that fight if you let it be - by lacking the DPS to kill him before time runs out. By stopping to click a charge, you're actually hurting the raid.

I had Lightwell up through Najentus, as I was asked to by my guild. It's handy on a few fights in SSC/TK - Morogrim, Kael, etc - and I think that the changes will help make it more viable, however I do NOT think that they will make it completely viable over the other possible spec choices you have. Plus, the biggest issue I noticed when I had it was simply that no one even clicked it. I would say on vent before I put it down if I was going to, then when I put it down, then sometimes even a reminder later on, and people hardly noticed it. Plus, even if it was ticking, the other healers would be so quick it could MAYBE get one tick off.

It was good in five mans and smaller groups like a Karazhan, and amazing for solo or duo farming, but in a raid with six other healers it is completely lackluster and noneffective. The idea behind the spell was great but the implementation is not.

I think a great change to it would make it an instant heal rather than a HoT. Maybe a little overpowered but honestly, lolwell deserves it.

Oh I agree, but if your DPS is lacking, and your mana pool runs to 0/0 at that point it becomes viable, but not before.

On another note I have a quick math question again.

If calculating stacking bonus' to certain spells, I'll use Greater Heal with bogus' figures here for a sec.

Assume GHeal hits for 2k health, untalented, with zero HSE. I want to figure out how much a talented, +2000 HSE will hit for. Assuming all of the following possible modifiers:

Spiritual Healing +10% healing, Empowered Healing +20% to GHeal, 4 piece Absolution bonus +5% healing.

Which of these does it go (because they all give me different answers) *note using 2000 healing as base:
2000*(1.1 SE)*(1.2 EH)*(1.05 Absolution) = 2772 healing
2000*(1.35 all added together) = 2700 healing

Any help on this would be great, thanks!

The correct form is [base + (3/3.5+0.04*(empowered healing ranks))*HSE]*(1+0.02*spi healing ranks)*1.05(=4T6 bonus)

with 0HSE, 5/5 spi healing --> base*1.1*1.05

This does not include crit