For me, the issue of gearing for throughput instead of regen comes down to this: will gearing for throughput change the way I heal? And the answer that I've always found is, no.

Say I'm raid healing, and one of the DPS is down 5k HP. I cast FHeal on them, healing them up to full. If they aren't at full health, they'll be pretty damn close, making casting another FHeal a huge waste of mana. Whether or not I had geared for full throughput, the target would still be at or very near full health. So the extra throughput would do nothing in this case.

Now let's say that DPS is down 10k health. In most cases I would still use FHeal, as using GHeal on squishy DPS is often a very real gamble. If my first FHeal hits for ~5k, then I'd still need to cast a second FHeal to top them up. Assuming that 5k heal was with regen gearing, let's look at throughput. I don't know realistically how much a FHeal would go up by, but let's say it will hit for ~5.5k. Meaning after my first FHeal, the targe would still be down 4.5k HP instead of 5K. Since 4.5k is still a lot of missing health, even with gearing for throughput, I would still need to cast a second FHeal. And that would most likely result in overhealing.

Of course, you could also take the case where the target was down enough HP for two 5.5k FHeals exactly. Then, if that was the case, the target would still be down ~1k health after being healed by two FHeals if the healer was using the regen gear. So the worst-case scenario for regen is that your target will have a slight health defecit (1k is generally nothing), and the worst-case scenario for throughput is that your extra healing will be overheal and wasted. Both cases are nothing of major consequence; but with regen gear, you're gaining a lot more mana to work with, while having basically the same healing style as someone who geared for throughput.

To sum all of that up (which I may not have typed clearly) - for me, throughput only becomes important when it can actually save me a spell cast.

But this will generally never happen, because especially these days, pretty much all of the healing gear a Priest will take is balanced between regen and throughput. Gone are the days when you could really go all-out in gearing for one particular stat (I remember running DM endlessly for MP5 gear when I first hit 60).

What that means is, the only places where we really have choices in terms of throughput or regen are gemming and enchanting. And while those add non-trivial stats to our gear, I don't think you can realistically get enough throughput from just gems and enchants alone to significantly alter the strength of our heals. Which again means you wind up casting the same number of heals as someone who geared for regen, except they're getting more mana back.

There are other reasons why I prefer regen as well, but others have touched on most of them already. For me, regen will always be my primary objective of gearing unless something drastic changes.

I should note, however, that when I say throughput, I'm really talking about spellpower. Haste is another story, and does indeed have its own tangible benefits. Regen and spellpower become meaningless when a target dies because it didn't get a heal in time. While in an ideal world healers would be prepared and haste generally wouldn't be necessary, it can help a lot if your ping is a little high, or if you're just a little slow sometimes.

So, I can understand healers (well, maybe non-Druids) that go for haste. I personally still prefer regen, but haste is also a pretty good choice. What I don't really understand is healers going for all-out spellpower. I understood it back when downranking was king, but these days? It just doesn't make much sense to me.

Isn't regen vs throughput just a manifestation of the very old risk vs performance problem of raiding? After having healed in raids for more than four years I've always geared for regen if I was uncertain on the risk of the pull. I.e, most of our harder progression pulls has been done in regen heavy gear as opposed to throughput. It allows for a far superiour flexibility when something doesn't go as planned - and it very often does on the harder bosses. Fellow healers might die or disconnect, DPS might fall off which will make the pull much longer - and in both cases you might be able to salvage the pull much more easily with a manabar that lasts as opposed to heals that still overheal.

You might provide math that quantifies spellpower as the single most potent stat. But the math of being OOM is rather easy to fathom as well and of a much more absolute value in terms of the end result of the pull.

Mana versus output really just depends on the fight, which is why having or gemming for balanced gear is probably the best route. What's that mana regen going to do for you against General Vexas, are you going to have mana, or will your heals be potent enough for hard mode fights? Versatility is everything for a priest. Playing on the priest strengths and not being sloppy will take care of your raid and your mana bar. When you're at the point that you're not running out of mana then output should become important. What good is lots of mana when you or half the raid is dead. (as of now there really isn't any limping across the finish line. Much of uld is all or nothing when it comes to people standing. Hard mode is all or nothing.)

As long as we're (or were) on the topic of trinkets, it looks like there is yet another Je'Tze's Bell-esque trinket dropping from Yogg-Saron on Hard Mode called Show of Faith. This is obviously a direct upgrade from [Sif's Remembrance], assuming similar uptime on the proc.

In terms of spell power, the new [Titan-Forged Rune of Audacity] from Wintergrasp Marks of Honor equals the [Forethought Talisman] yet has an (arguably) better effect with it, since the Talisman's HoT is effectively useless in a raid environment. I considered using the Rune for Yogg-Saron to remove his "soft" mind control (read: not the 0% sanity one) when Will of the Forsaken was on cooldown and there were no dispels available. I'm sure there are other Ulduar 25 encounters in which it could be useful. A "your mileage may vary" thing, but could have some PvE use if you are out of luck trinket-wise, or could benefit from the PvP dispel.

[e] Broken link fixed.

I feel like a lot of this discussion is getting astray in "general healing theorycraft", and isn't going to be that useful to a typical priest reading the thread. There's also an awful lot of anecdotal evidence and stories, and very little data. To summarize, people are saying three different things, either through math or examples. They are:

• I don't benefit much from marginal additional spell power
• I don't benefit much from marginal additional haste
• I don't benefit much from marginal additional regeneration

And often that claim is follow by "and that's why I stack spell power / haste / regeneration instead".

However, all of these stats are useful in large quantities, and if they were totally useless in small quantities, we wouldn't care about gear upgrades either. Clearly the small upgrades matter too. Additionally, all of these stats have theoretical effective caps on usefulness, but we will not reach any of these caps with current gear. Those caps are:

• Spell Power: When all heals are overheals
• Haste: When all spells cast faster than the minimum global cooldown
• Regen: When you can chain cast any spell and not run out of mana

None of those situations will ever happen (barring strange fight mechanics like Vaelastraz). It's crucial to note that when people say "spell power is worthless because it's just overhealing", the real argument is that sometimes the extra healing is just overhealing. Other times its not. Similarly, sometimes extra haste lets you cast a heal in time and other times it wasn't needed. Sometimes you can spend the extra mana you regenerate and sometimes you can't. The real dispute is over how often a marginal increase in a stat actually prevents a raid death, and the numbers are so low for all stats, it's hard to say which is "best", even if you know both the fight mechanics and duration.

However, I think everyone can agree that some gear trade offs are always worth making. Even if you think regeneration beats throughput, everyone would give up 1 spirit for 100 spell power, or probably even 1 spirit for 10 spell power. Many people would trade it for 1 spell power but not everyone. This calls to mind the immortal words of Captain Jack Sparrow:

At this point, we are haggling over the price of stats. Is it worth trading 100 spell power for 200 int? 100 int? 50 int? There isn't an optimal solution for this explicitly because some fights favor regeneration more and others favor throughput more.

I acknowledge that there are also feedback loops between the stats. For example, the more mana you have, the better spell power is because it makes each point of mana buy more healing. But the reverse is true as well. If you have a bunch of spell power, you get greater returns from an increased mana pool than if you had less spell power. However, if you crunch the numbers on feedback, it's never enough to substantially change your stat valuation. In other words, if you pick one stat for your yellow gem stat, all yellow gems you use should provide that stat.

Rather than try continuing to argue over whether its better to stack throughput or regeneration, perhaps we should try to derive three basic stat equivalences, one for maximizing throughput, one for maximizing regeneration, and one for balancing the two. It's still up to the individual priest to decide which gear set is appropriate for which fight. That's a useful mathematical thing to model, and we can leave the decision of which model to use up to your own discretion.

Glyphs: Because its still missing.... :>

Right now I only can refer to Holy-Raiding because thats what I'm doing all the time.

Glyph: Flash Heal
I think its a must have because even if you don't use it that often its one of the main spells as Holy.

Glyph: Guardian Spirit
There are so many occaions where your tank needs extra protection or someone is getting too much dmg...
Every Minute you are able to protect someone form dying and therfore its really helpful in every boss fight....



That leaves one slot and two Glyphs you have to choose from.

Glyph: Circle of Healing
I do not have the exact data but in Ulduar you do not stand so much next too each other and therefore one additional Target isn't that good....

In my last Raid it did a avarage of 2746 on one Target.

Glyph: Prayer of Healing
I am pretty suprised about the date I am getting from WWS every Raid... I usually cast it after Flame Jets(Ignis) ore Earthquake(XT-002) and it only produces about 4% overheal.....

It does singel Target about 5537 on 5 Players and leaves 5*20% hot --> like one extra target you get from coh....

So in the end:
Better Range and Higher Heal....

You could argue that its only a group heal and coh is smart target and so on, but I think you cast CoH in the same situations where you need to cast a PoH.... If you first cast CoH you save thoose with low health and then you cast PoH and overall you get more heal.


Glyph: Power Word Shiled
I think we need to do the math on this Glyph once we figure the T8 bonus out

A spell is not one of the main spells of a spec "even if you don't use it that often." For something to be a "main spell," it has to be used often. For me, greater heal is turning into a spell that I "don't use that often." Flash heal is not.

I'm not sure what fights you're doing in Ulduar where you "don't stand next to each other" enough that circle of healing couldn't find a sixth target. The only fights where circle of healing shouldn't be a significant portion of your healing are the ones with little to no aoe damage.

You're misinterpreting the data you're getting on glyph of prayer of healing overheal. HoT tics that don't heal for anything don't show up in the combat log, and aren't counted for overhealing. For example on my Ignis kill last week, I healed for 505k effective healing with prayer of healing with 28% overheal. That should mean I had 129k in glyph healing, but the actual number was 66k. That's almost 50% overhealing from the glyph, but the combat log parse says I had 3.2% overhealing with the glyph.

The problem I have is that most caster gear of the same ilvl is very similar: it has roughly the same int and spellpower with a choice of crit/mp5/spirit/haste so you're only going to be able to min/max certain stats with gear that actually exists. The only notable exceptions would be trinkets and a choice of gem slots. This is not to say that theorycrafting this wouldn't be useful (it would be very useful as it shows what is possible), but I'm just worried that people coming here for information would end up wondering just how feasible it is to carry around 3-4 pieces of gear in the same slot just to have a "haste" set, a "spirit" set and a "crit" set, for example.

Still, a good approach might be to try and build sets of gear that max each desired stat and see just how much you gain or lose. If they show more than marginal differences, then we can go from there and see what those differences are. If they don't actually amount to much that is different, then at least we'll know for sure that, other than choosing a few obvious best in slot items, the choice about whether to stack any one stat after a base amount is more of a personal preference than anything.

A note on this in the spreadsheet I have included 226 iLvl sets fo mix maxing SPR SP INT Crit and Haste.

Stat sums are in the bottom.

Yes, I'm looking at it now. It's very helpful for this discussion (though there's alot to go through still). Also one possible error in the sheet is that Hood of Rationality seems to be listed on both lines 5 and 7. I don't think it matters though.

I've been looking over the gear available to us and virtually all items come down to "Pick two of Spirit, Haste, Crit". In general, it appears that items with m/5 instead of spirit should be valued at a tier lower than they are. In other words, a 213 item with spirit is roughly equivalent to a 226 item with m/5, but an otherwise similar distribution of stats.

If there are two pieces with the same kinds of stats but different numbers, the one with the lowest stamina is almost always the best. For example, [Conqueror's Leggings of Sanctification] versus [Leggings of Lost Love]. The t8 piece is the better of the two (though I personally think other pieces are better than both).

So to make gearing choices, we only need to decide on an ordering between spirit, haste, and crit. The value of Spell Power and Intellect is rather meaningless, as you don't have many gearing options regarding these, though it does matter for gemming. The math I've run says Crit > Haste > Spirit, since I haven't had mana issues healing. This means my first choice is Crit + Haste, second choice is Crit + Spirit, third choice is Haste + Spirit. Pretty much all gearing choices fall out automatically from this ranking system.

For the record, I'd suggest that both Crit > Spirit > Haste and Haste > Spirit > Crit are logically implausible options. Haste and Crit are primarily throughput stats while spirit is primarily a regen stat. If you think throughput matters more than regen, then Spirit has to be the worst stat of these three. If you think regen matters more than throughput, then Spirit has to be the best.

Any of the other four options have credible arguments behind them, generally depending on when you think you might run out of mana.

On the subject of the gear choices put forth by that spreadsheet, I'd suggest that [Conqueror's Mantle of Sanctification] is flat out superior to [Mantle of Dissemination]. You gain 6 spell power, 2 crit, 6 int, and lose 3 spirit. Yes, it's the shadow set piece, but it's really good. The holy piece might be good depending on how useful the set bonuses are. I'm still undecided.