Well it is pretty accurate if not completely accurate. But really if you need to get "brownie points" for your PoM being boosted by warlock armor multipliers I'm not sure what to say about your healing perofmrnace.

PoM is a great spell for tank healing period. It's a great spell for splash damage fights. It's a great spell. It's a portion of your total healing, albeit not a huge one. Use it often. Judge your performance versus other holy priests with similar assignments. If you are way down on the meters figure out why and then decide if you need to climb them.

Recount may get confused, but this can happen without any amplifiers if two priests happen to have exactly the same amount of +Heal (after all effects). Just as you have to know you can't rely on Omen if two mobs happen to have to the same name, you have to know the limits of Recount, too.

If you mean the effect of the healing amplifiers as a contribution to total amount healed (not mixing PoMs issue), then it's just the same situation as with other healing spells. Either the tool takes the effect into account or it doesn't.

Same story as always: a fool with a tool is still a fool.

Is getting the haste items in BT even worthwhile for a healing priest? The low % casting time increase even with all 3 items seems so small that it's not worth the loss of regen. We've been wondering what to do with the cloth haste healing items from BT and we're thinking of not using them.

Does anyone have any other suggestions?

The numerical theorycrafting on spell haste for healers is pretty straightfoward: its a linear throughput increase for spells with cast-time > 1.5 seconds (until reaching the 1.5 second cap), more consistant than crit and at any gear level being discussed higher throughput increase than the +healing upgrades one can get. It is the same HpM. It currently replaces regen itemization on the vast majority of items for healers.

As for "suggestions", I think the big thing is: how is your mana? If you consistantly have little to no mana problems, then spell haste is probably worth picking up. The main benefit I see, while partially anecdotal, is it shortens the duration between cancel casts, as well as spammed low rank. Quantifying the benefit of delivering your Gheal 2/3 every 2.2 seconds versus every 2.5 seconds is difficult, same with quantifying changing a 3.5-4 second "cancel gap" (cancel at 1-1.5, land 2.5 later) to a 3.2-3.7 "cancel gap", but its hard to determine and quantify for comparative purposes exactly how much that .3 seconds is worth in MP5.

I would postulate that if you have spare pot cooldowns, you will derive greater benefit from spell haste than from your standard incrimental gear upgrades. Even flash/binding gains some benefit: oftne when you are casting these spells tenths of a second matter for landing them.

I'm currently attempting to quantify this, does anyone have a link for boss swing speeds? A search did not reveal any good data, my apologies if it is easily findable and I simply failed.

Well, I do use recount and I have noitced being a bit higher than usual on it, I just chalked that up to the new gear I got about the same time that I got the mod. In general, healing meters are a way I judge if one of my healers happens to be out performing other people significantly, or if someone is *really* slacking (some situations I just throw away the meters). It's more of the overheal values that I'm concerned about. I'm trying to get my entire guild's worth of healers to keep their overheal in the 15-20%, I don't know if this is what I should be aiming for. But as it stands we're usually really good, and of course specific fights I like different numbers, but rarely are we over that. The only place right now we're over is on prince, since I tell my healers that overhealing is better than our tank (druid) take 3 crushings in a row and get jibbed. The situation that I'm concerned about threat is a bit more complicated.

I use it as a way for our tank to build lots of threat, since I've noticed that a fair few of our guildies don't use threat reduction talents (our spriest didn't have them until I looked at her spec and mentioned that it might be why she's pulling agro a lot). So using it at the right time, so there's minimal overheal and the full healing threat gets applied. However, I know it's a moot point here, but the difference between 20% reduced threat and the extra however much our tanks are generating (like deffensive stance for warriors) and other talents for +threat. Now, if I took SR out of my spec, I know it wont effect the healing agro since it counts as the targets healing, but is there any "I cast a spell" threat similar to buffing or PW S threat with PoM. I haven't seen any, but I'm curious for more of a knowledge reason than any practical one.

The other thing about PoM, really the only thing that I'm trying to get straight, is the choice of which target it will jump to. I've been monitoring it, and it does seem to have a max range, and on the down side if there's no on in range it just pops the head and the remaining charges are gone (or at least that's what my DoTimer tells me). I've noticed it does tend to jump to people who have dmg taken over people who are at full, though it does *not* seem to discriminate between people currently taking damage or not (i.e. have aggro). So does it jump randomly to someone in range who isn't at full health or is there an actual target it jumps to based on % of max or deficit or current hp. I've been testing this in Kara lately, and will continue to do so, I'm just hoping some people have tried this and can give some more info on the jumps.

Also, it's a bit OT but Curse of Elements from a warlock, does that effect the Shadowfiend's damage? (We're currently testing raid dps between our locks using different curses). I would very much like to know if this will increase the mana back from my fiend's hits. Also if there's a shadowfiend thread out there because I swear I saw one and just can't find it again, can I get a link to it? Thanks.

What do you mean by Cancel gap?

Short answer: Yes, CoH +full points in Empowered healing move you much further up on healing meters than being the IDS buff-bot.


Long Answer: As everyone has pointed out above Healing Meters are subjective and can really hide the true value of a healer spot. I.e. you provided a buff to every caster in the raid (with a decent regen buff for the Priests/Druids)
But second you have to break down how different heals work and in different situations.


Long Analysis to follow of how meters representation occurs in typcal raids with probably a few errors:

Situation 1: Four healers MT Healing.
Tree: Stacking Lifebloom, and keeping Rejuv up while hitting the tank with regrowths if necessary. Now since Lifebloom ticks every second, it has first chance to also apply healing to a tank, and thus gets always counted on a meter. Lifeblooms is also the most efficient druid spell in the book. Plus, every 20seconds, the tree has a mini-NS by swiftmending a HOT for full heal whenever a spike hits.


Paladin: A Paladin will be chain casting FoL or in large damage situation (and well Geared) low rank Holy Light to keep Light's Grace up. Since these spells with either a 1.5s or 2.0s period, they have next highest chance of healing and showing up on a meter

Priest: Keeping renew going on MT. Occasional PoM. Then chain cast GH2-4 or Cancel-Cast GH7. Since this spell is 2.5s (like the Shaman's Healing Wave) it is tied for the slowest period main use spell. Thus for small amounts of damage, most of its value never gets used, since HOTs will tick on random pattern(3s), then Lifebloom every second, followed by Paladin Spell of choice. Once the damage is greater like 6-7k+, then part of the Greater Heal will count, since not all the spells will fill up the full deficit.

Shaman: While not the perfect ideal for MT healing (CH owns all but more later), there are still more than adequate for the job. Earth Shield is one of there best MT healing spells, but don't get counted, since it shows up as self heal for the MT. For direct heal spells, they are using Lesser Healing Wave (1.5) or Healing Wave(2.5). Yes, Lesser isn't that great and doesn't stack Healing Way, but it is slightly more efficient than a comparable Priest's Flash Heal. So either they are on the same Paladin frequency of healing or on the Priest's frequency of healing. Either way for healing meters, they don't really do well on a single MT healing fight.

From this timeline and spell usage, in general, on a single MT healing fight, the Tree and the Paladin should win a meter race, purely based on frequency there spells hit the most often.


Situation 2: Various Healers in Raid Healing (In order of descending meter exploitation)

Shaman: (people in relative close vicinity) Chain Heal. Chain Heal became the best damn spell in the game the day for multi target healing the day Blizzard fixed it so it the next lowest health person in range.

Tree: Random raid member takes damage. Every healers clicks target and places heal. The tree has two options: 1) place a lifebloom, that will guarantee to tick once before the avalanche of FoL,LHW or any other spells hits, and may overheal the target. 2) Rejuv, and then swiftmend it at 1.5s and come in tied with the FoL/LHW/FH category. Truly good stuff in meter exploitation.

CoH Priest: When multiple people in a group get hit by damage, this spell is better than anything listed above. This is truly how a priest can "compete" on a meter for a consistent basis. It loses some of its efficiency as less people need it, but it still can be used. (Other spells fall under the IDS Priest category below)

Paladin: Fast,Efficient 1.5s heal that can hit a single damage target for a good amount, and they can chain cast this from here till the cows come home.

IDS Priest: Hmm, what are the options. Shield, renew, FH or GH. Shield good in cases, but no meter exploitation. Renew: very good since it will eventually fill the target, but since its 3 seconds before first tick, other heals might come in first. FH: Now you're trying to compete with a Paladin with a spell that has 1/3 of the mana efficiency. And then the normal GH which is solid, but as mentioned before, on smaller damage amounts, you end up canceling or overhealing the complete amount many times.


Now in typical SSC/TK fights/trash, you end up with a mix of both situations. I.e. most of the time, you are healing tanks, and then every 20-60s and AOE/cleave with multiple raid members. In this situation, healing becomes slightly a free-for-all, because the expectation is to keep the tanks up and then bring raid members back to full on interleaving spells or such. So on the Tank portion, the priest is doing okay, but slightly behind the Druids and Paladin's. But when an AOE hits, its now time for Meter Exploitation/Catchup. The Shaman get in a few CHs that throw them up meters some. And in 2-3 instant casts, a CoH Priest can account for a quick 10k-15k healing. That 10k-15k might be more healing than you will generally do in 20-30second period while simply MT healing with Renew and GH.

This last part is where the COH Priest will just absolutely push past the IDS Priest on healing meters.

Icetro:

No, thats exactly correct. Its the theoretical time ending with landing a heal, beginning with either landing a heal or cancelling a heal and starting casting, with a single heal cancel in the middle. Effectively, it will be the longest amount of time between landing relevant heals, as if you see the tank needs heals or expect her to by completion of cast, you will not cancel. The concept is based the worst case play-related scenario of tank getting hit as you cancel your heal, so you "should've" landed that heal but did not due to latency.

A legitimate arguement could be made for looking at the reduction between when the heal would land after cancelling, instead of the full "heal to heal" gap, although it only matters for comparison of "percentage increases". The premise is you get to make more choices of canceling per minute, and have more chances to land heals. This is perhaps an overly "conservative" methodology, as in many cases the relevant cast time is after movement or an ability, and thus the baseline time would be the cast time of GH or FH, and result in greater percentage increase numbers (as it is static numerator adjustment from spell haste, but the denominator is greater for this model).

in case others are interested in working out math on quantifying the benefit of spell haste, i was looking at doing an analysis of a 'per minute' fight involving a boss swing speed, as well as various special abilities on set timers (pull a couple from various sites), to quantify how likely spell haste is to improve your chances of landing a signficant heal between attacks, as multiple abilities/attack without recieved healing is what is lethal.

If anyone has any other ideas or suggestions on how to quantify this, I'm interested in hearing them...I'm a sucker for spreadsheets but developing a concept of how to make objective emperical conclusions and correlations on this topic are currently eluding me.

I agree, this sounds like it could be a really interesting project. Some thoughts:

Bosses' swing timers could be difficult. I'm thinking specifically of Gurtogg Bloodboil and his DW shenanigans here, but any boss that hits hard and fast could be difficult almost to the point of making this discussion moot - unless you have serious, serious mana issues, or unless the tank is going to be at 95%+ HP when your heal lands, you let it land. I rarely cancel heals on Prince phase 2, to use a lower-tier example, because most of the tanks I run with can easily eat a thrash and die.

To revisit the mana issues point: The fact that spell haste is in lieu of - rather than accompanied by - regen on Hyjal/BT gear is a significant point. Unless you're comfortably finishing fights with a cushion of mana / a shadowfiend cooldown up, you won't want to sacrifice regen in favor of haste, which will, theoretically, run you out of mana faster.

Lastly, I'm still skeptical of spellhaste in general. I used the Scarab of the Infinite Cycle for awhile, and to be honest the haste buff was really underwhelming. Haste in general seems a bit like armor penetration (which makes sense, as they're both T6 stats) insofar as you need a lot of it to nice a difference. Call me old fashioned, but my MP/5 and + healing are still my stackable stats.

Thanks for the in-depth reply, Sordee, that helped me see things more realistically. I also plead guilty to the "used to be a DPSer" thing... I rerolled from hunter, so I am used to evaluating myself in a different way. In hindsight I think the reason I perceived a change as we went into SSC had more to do with some of the other healers getting geared up to the point I was already at, than an actual change in what anyone was doing.

The cancel point takes into account "proper play": ie, with what you know at the time, the tank will be at 95% when your heal would land.

Comparing it to Scarab isn't particularly valuable, primarily for consistancy. You *will* cast faster if you're wearing spell haste, you know exactly how long it will take to cast a spell, while with scarab (like the 1/2 cast meta) you generally only get full benefit in a chaincasting situation, where it will increase your throughput.

And the armor penetration comparison is misleading: your comment is relevant to every piece of gear upgrades ever gotten. Individual incrimental increases, be it attack power, crit, hit, +healing, mp5, all do nearly nothing by themselves, its the sum of the parts that matters. We'll pretty much never be able to tell if 30 spell haste or 8 mp5 would have won us that fight we lost, but we try anyways.

The postulate for spell haste over regen is that if you have "mana to spare", this itemization is better. Padding a stat that effectively does nothing for you besides reduce the number of pots you need to use only has the benefit of less gold spent, and more cushion for "looser" gameplay (this is not an insult, nobody can play perfect 100% of the time, and if you have the capacity for landing more heals you can get away with looser play, being less effecient). "Trimming" that padding for a benefit in throughput and "reaction time" feels worthwhile. Obviously if you're maxing your mana regen and still running out (and have tightened your play as much as humanly possible), you shouldn't be looking at spell haste. But from the information I have, T6 play tends to involve much less mana potting. So itemization that might save a life at the expense of increased mana potting is likely worth seriously considering.

I am the raid leader when we're doing Kara, or at least one of the for the guild. I take healing lead when we hit the 25mans. I do see the point in caring more about whether people stay alive or not, it's always been my opinion that it's when the heals hit as oppose to how big they hit, of course that's not 100% true in all cases (See: FoL spam pallies). The thing about overheal is the mana lost for those heals, and while there are no problems now or in Gruul's I can see problems coming up in the larger longer fights with healers going oom because their heals are hitting full targets.

But you're right, the meters for healers are generally not looked at. Threat meters I ignore because it's either a tank dead (90%+ chance to wipe the raid) vs a healer dead (chance to raid depends on how far into it you are). Not to say healers are above everyone else, but a slacking healer usually means a wipe. It's far more important for us to be on top of the game, and therefore ignore meters in favour of keeping the raid up.

You're also right with the fight dependance, against Gruul, I'll let my heals land if there is any damage on the tank or hateful tank at all. They get spiked and die, it's our (healers) fault.

But back to the PoM tangent for a little bit, I've been trying to pay attention to it's jump targets, and I have discovered that it will not jump to someone with full health if there is someone who has damage taken, so that's part of what it is. The part I've noticed is that a lot of the time it'll throw itself onto a warlock or other player who is in range of the jump that has next to no dmg but again isn't at full. I'm going to be looking at it a bit more with max vs %age being more important. But as of right now I'm absoutely positive that the only part of distance that matters is inside range or outside range (can jump vs can't).

My opinion about overheal is that if you are not asking for innervates or complaining about mana potion consumption, then your overheal is no one else's business. In fact, I'm confused by people who disagree with that opinion. You can't always watch swing timers and even when you can watch those, you can't always also watch splash damage (i.e. cave ins, random cast effects, etc). If you can spam the target and not cast-cancel and you are not moaning about mana, overheal all you want. I suppose you can choose to optimize overheal anytime you want when damage is not spiky. But then, it's not exactly hard to do that, is it?

Last time we did Leotheras, we had three healers on the druid tanking the demon (maybe I've already related this story above? let me tell it again....) The paladin and the druid were landing all the easy heals long before I landed a single Gheal when the damage wasn't spiking. Cast cancel at say 2 seconds in the guessing game and then restart and I'm maybe landing something useful some of the time. Just land a Gheal3 or 4 every 2.5 seconds period and the druid and the paladin can know this and work around it accordingly. And then, if they are falling behind, switch to a Gheal 7. Massive, massive overheal but a much better chance of success. Now, we can discuss the merit of healing the guy this way or having a druid tank the demon but I could spam that heal for the entire fight with ease. Should I have done something differently? Hard to say.

After I read your first post, I had a look at this during this night's (horrible, as usual) SSC. I almost always had PoM on cooldown, and I think I can confirm your observation. Perhaps a test scenario outside of a real raid would be best.

Reemphasizing Mideci's example:
In the end this is what your roll was expected of, and you did it. You might have the most healing off all, and be relatively low on the effective heal meter, but your "tank" target stayed alive in large contribution because of your steady stream of heals were always coming in. Which works really well when you have a caster mob sending a large shot every 2-3 seconds to a single target.


And yes, overheal used to be a big statistical no-no before TBC, but now with elixir and mana regen changes, its all about getting the job done with whatever mana you can throw at it.

I would like to see an addon that shows accurately boss swing timers. It would help cast-cancelling A LOT while healing MT. I've tried to search but are there even such?

I've spent some time reading through the Compendium posted here and seen some really interesting things.

I did some investigating on the armory at the 2 main Hyjal/BT guilds we have on Dragonblight, and the specs of their holy priests. They don't seem to have even one IDS spec priest! I had a chat with one of them ingame, and he encouraged me to try a new spec.

Our raid setup is normally with a minimum of 6 healers, 3 IDS priests, holy pally, tree druid, and a resto shaman. The pally and/or the shammy normally take the MT healing. Having been looking on here, I decided having 3 priests IDS spec is a waste, so I recently went CoH, and I love it. I can't wait to try out healing the melee on Void Reaver (the shammy normally does this with a pally for backup).

My dilemma is, that I'm the only priest with a better than 95% raid attendance (being an officer and Raid leader I'm pretty much on all our raids). When it comes to 25mans, its rare I'm the only priest. But if I was, I wouldn't want the raid to miss out on DS. In terms of ZA, I'm usually the only priest (with a pally and shaman).

Our guild are mainly doing Void Reaver in TK, Gruul, and ZA/Kara. Would it be more beneficial to the guild for me to go back to IDS spec? Just seems a waste 3 of us having the same spec when all priests do attend, but then I hate the guild missing out on a buff if I was the only one on that particular raid.

Oh and talking about healing meters, we never have a priest on top, due to having paladins in the raid. I'm always 3rd after the holy pally and shaman. I hardly ever look at them any more, as long as my targets stayed alive, then I know I did my job.

In ZA and Kara, IDS isn't mission-critical, since you just don't have that many players in the group with enough willpower for IDS to matter. In 25-mans, IDS matters, but as you say, you are typically not the only holy priest there. As for ZA, I can only say that CoH is brilliant there - for some of the more dangerous trash packs and especially on Malacrass and Zul'Jin phase 4. For ZA, however, I do carry a stack of WP scrolls.

Consider also that a CoH spec gives you two more points in Empowered healing. This is a very nice boost that even scales with your equipment.

Well, we too have paladins in our raids, and since I have started using Grid+Clique in order to be able to toss some flash heals around without compromising my tank's safety, the pallies no longer post heal statistics.

Questionable healmeter fun aside: try this for Zul'Aman, it will boost your reaction time there. Damage on the group is so spiky and unpredictable there, you can't really talk about "my targets" vs. "the other healer's targets" any more. Any healer will regularly need help on "his" group, and if you can quickly throw emergency heals around, so much the better.

I'm unfortunately not up enough on programming to be able to write a program that digs out relevant data out of WWS, and short time enough to not be able to dig out enough samples at this time (and it would be much faster with a program): could anyone, if interested, either direct me to someone who could and is interested, or direct them to me, or just work off of this?

The generic formula for Gross Boss Swing Speed is below:
Sg = Gross Boss Swing Speed (number without any parry acceleration), Swings / s
Sn = Net Boss Swing Speed (taking into account parry acceleration), Swings / s
P = Number of Boss Parries during a fight, #
A = Number of total (melee) attacks made by the boss during a fight, #

Sg = Sn / (1 - (.26 * P / A))

This is slightly less useful for a DW boss, due to the difference in MH & OH damage, but the number is not completely useless, just less good at accounting for spikes due to alignment of MH + OH attack timing.

Net Swing Speed, Sn, is quantified as:

Sn = length of fight (in seconds) / A

note that this number will likely be slightly slower than it 'should' be, as any movement, knockback effect, or other reason that the boss may not be getting 100% from swing timer would mean less attacks over the course of the fight.

Thoughts on how to eliminate this, without a very complicated program parsing out exact hit times and parry times from combat logs?

Yes, in ZA, CoH is fabulous, as it has its uses on every boss except Bear Boss. And it is nearly invaluable if one of your three healers is not a Shaman.


Side note on OverHealing: Druids have always been OP on low overhealing since a large chunk of there healing is from HoTs. And as mentioned above, HoTs don't record in the log anything if the target is already at full health. So while its a wasted tick in no healing done, its also no over healing done.

Hey guys, I could use some help.

Although I have never really played a priest, I'm a big fan of theorycraft, skill use optimization, and the like. Not just for hunters, but for other classes as well, which is why I spend alot of time on EJ.

I have been discussing alot with my guilds healers about what skills they use, gems, strategy, and whatnot. One thing I found is that our holy priest doesn't like to use GH. In fact, she uses flash heal and renew almost exclusively. I'm looking for some advice to help her realize that GH is almost always the better choice (we don't really have a problem with tanks dripping low). I read the whole first post, but It kind of assumes we all know GH is usually better, and doesn't really go into the "why's" much.

Thanks in advance.