Well, you're not going to be seeing me in T6 anytime soon (if ever). So, I'll just have to look at the math, since there's no way I can produce raw data (as preferable as that might be).

As HL 9 gets 100% of +heal, I'm going to assume it's being used in preference to HL 10 or 11. The base mana cost of HL 9 is 660. Crits thus give 396 mana. Casting time is 2s as long as Light's Grace is in effect. Chain casting with no lag therefore yields the equivalent of 990 mp5 with a 100% crit rate, or 9.9 mp5 per +1% crit. Switching now to item value, crit gives 1 mp5 per 2.23 and mana regen 1 mp5 per 2.5. Thus, if you're casting HL 9 over 89% of the available casting time, spellcrit gives as or more mana than an equivalent amount of mana regen. And if you have, for example, 4pc T5 or spellhaste (and using the shorter casting time to cast more HLs), or are casting HL 10 or 11, this weighs things even more in favor of spellcrit.

Of course, that's not the only effect spellcrit has. In situations where the extra healing from a crit is not solely overheal, you can also be getting some amount of extra healing, up to 50%. Potentially, this could let you use downranked HLs (thus further lowering mana costs), or let you heal through a situation you might normally not have been able to.

I'm not saying spellcrit is the best stat or anything. I'm just saying that I can see that it might become more useful to stack than mp5, depending on the situation. Personally, I stack mana regen as much as I can and go with FoL spam with the occasional HLs. And that works great for KZ/Maulgar, which is as far as I've gotten.

Well, the middleish of this thread got pretty horrible and almost unbearable. That discussion and huge tangent on various T6 setups and the 'amazing value' of stamina and intellect and finally Saviour's Grasp... Help.

Anyway, this is basically what has been established, but I'll just say it again incase some people are just really not understanding it:

If you are in a shadow priest group as a paladin, it doesn't matter what gear you have or your healing strategy is, because you have infinite mana.

_any_ level 70 holy paladin in Tier 3 (yes, 3) has the baseline spells to put out enough throughput (HP/S) for any fight in the game so far. That's because the highest ranks of Holy Light on a BOL target are just absurdly huge heals and in the vast majority of cases are going to be overheal and wildly inefficient.

Semi related to the above point; I can only see spell haste ever being useful if you were always given a shadow priest, and thus, wanted to utilize any stat you could to throw away your mana quicker.

At the extreme upper echelons of spell crit percentages, Holy Light is the most efficient MT heal that combines both HP/S and longevity. (It has the most talents, coefficients, and other modifiers, this can't be disputed). By extreme I mean when you WILL be sitting at 50% holy light crit raid buffed. There is a wide range of valid HL ranks and they will be from a standard 3000 range to the 8000 range. There is no need to ever go larger, currently. (Or anywhere even above 5k)

Also, with reliably high HL crit percentages (50%), you can count on your 'heals of choice' being quite a bit larger than their noncrit heal value. If your HL is healing for 3k noncrit but it's a 50% crit, you may as well assume you're casting a ~3700 heal each time and not just a 3k, 4.5k, 3k, 4.5k...

Comparatively, if you somehow achieved 50% Flash of Light crit, even using your max rank flash with the ~2200 healing you'd have (not possible and still have 50% flash crit, but for the sake of theory lets assume so), you're still capped at your HP/S because flash is inherently small.

The closer you get to a 100% spell crit rate, the closer you are to almost never running out of mana. For this to be true, though, you have to account for the baseline MP5 that is on all of the best gear out there, which gives you a very nice pool regardless of how else you gear up. This is because spell crit gives the best average mana regenerative boost per itemization point spent on it compared to the alternatives. Compare 10 spell crit gems to 4 MOT, assuming HL, etc. Tiffara actually mentions this, it isn't even a hard concept to grasp.

+Healing is a statistic that diminished as you use lower ranks of spells, or Flash of Light. If you are trying to maximize your healing statistic gain on your gear per itemization point (say, a socket), why would you choose something that is already going to be fractionalized several times? Both MP5 and +spell crit rating do not diminish and will be constant benefits.

I don't do much special thinking or formulaic research when I gear up, I just look at in the above cases. And this is me:

I never get a shadow priest, by choice. Also, the other paladins rarely get shadow priests.
I always have an MT type healing role.
I never find myself wishing for more gain from my lower ranks of HL, or my Flash, to want more +heal. If I ever find that situation arising, I can use a higher rank.

Also, for whatever it's worth, I don't HL exclusively on every fight ever, or something. I flash on Illidan p1 and P3+ for example, but HL P2. I MT and raid heal on Council and use a mix of both, etc.

And to mention it one more time: If you always have a shadow priest, throw all of this out the window, and just get spell haste and +heal, and spam anything. It wont matter

I just don't see this. Do you have a WWS parse up where your crit was worth more than mp5?

Seems to me that a 4mp5 gem will beat a 10 crit gem in terms of pure regen power in all cases.

There is not an encounter in the game right now where you need to have the regen gained from straight mp5 gems (or hybrid ones). Between flasks/elixirs, and mana oils (and food for that matter), you gain enough passive regen that even without group support I can not see it possible to actually run out of mana. That combined with using mana potions on each cooldown means that having extra throughput on your heals to be that much more important.

The only times that I see myself even drop below 25% mana is during Gurtogg where I wear full spell haste gear (Dawnsteel Shoulders and bracers, Belt from teron, and 4pc Tier 5) and that is only right after a fel rage phase when my mana gets instantly refilled via bloodboil healing, or when I lay on hands during Mother or other fights when I have timers up just to make further healing that much easier.

Extra mana can always be converted to extra healing by upranking, at least until we can chain cast R11 indefinitely. I heal much the same way Yilona does with 4 pc T5, spell haste, and Libram of Absolute Truth. I try to do as much healing as possible with my hyper efficient R4, then when I need more HPS and/or can afford the mana I cast R11 (sometimes canceling, sometimes not, depending on the encounter and my current mana). For a random 7 min encounter without a shadow priest my ratio is about 85% R4 and 15% R11. Thats going through about 50k mana over 7 mins with my base mana, super mana pots, bow, shaman totems, mp5, etc (not counting illumination - I factor that in to the average mana cost of a spell for my calculations). The more mana regen I have access to translates into more healing because I can cast a larger percent of R11. Of course R11 has a greater chance of overhealing, but other than that it is very easy to model which stat yields more raw healing over a set amount of time given various ratios of R4 to R11.

For my gear at 90% R4 / 10% R11 mp5 gives the most raw healing over 7 minutes. Spell crit passes mp5 (factoring in both the mana gained and the extra healing from crits) at about 70% R4 / 30% R11. And +healing finally passes mp5 at around 30% R4 / 70% R11, though it is the +healing that is making the other two stats so good. At 0% R4 / 100% R11 spell crit is just ungodly which makes sense because you're multiplying a bigger heal and gaining huge chunks of mana back every time you crit. The 100% R11 actually occurs in raiding in a few encounters like P3 Reliquary and P2 Illidan (after which there is a lot of downtime and low healing time which you easily can get you back to full mana).

But all this raw healing talk isn't everything when concerned with MT healing. Getting your spammed heal as high HPS as possible also does wonders for tank safety, because you can't automatically uprank your healing mid cast. The stats that effect R4's HPS in order of importance are spell haste, +healing, and lastly spell crit.

So for most encounters, my gear setup, and being in the MT group (meaning I usually get a shaman and not a shadow priest), ranking the stats in raw healing over 7 minutes puts mp5 well ahead, followed by spell crit, followed by +healing. Ranking the stats for tank saftey vs burst damage I go with spell haste, +healing, followed by spell crit.

With the level of heal redundancy you have to have for MT healing, the raw healing is almost never an issue. Because of this, I place a huge importance on spell haste. Being able to land your big heal 1/2 a second faster saves many a MT. I'm even thinking about replacing Girdle of Hope with Girdle of Lordaeron's Fallen after the spell haste patch goes live. Dunno, will have to see how it goes.

I think the ideal gear set for MT healing in almost all situations would be to wear 4-piece Crystalforge, but if a Paladin does not have access to it, their healing options become a lot more muddled.


With 4-piece T5 and a Libram of Absolute Truth, R4 Holy Light on a BoL target surpasses the HPS and efficiency of R7 FoL. However, without those 2 bonuses, any situation that would call for R4 healing can easily be replaced just by Flash spam for an increase in output with a slight decrease in efficiency. This is one of the reasons why I personally utilize FoL a lot more in my healing, and just cast a R4 every 15 seconds in order to keep Light's Grace up. It's also why I do not value Spell Haste that high, since it's returns on pure Flash spam are almost negligible with proper use of /stopcasting.

Even in Black Temple and Hyjal, there are a lot of opportunities where FoL spam is appropriate. A lot of what Tpyo said resonates with what I've experienced as well, and that is if everyone upranks to a max rank "big heal" most likely the person with the least latency will land their heal and everyone else will overheal. In my previous guild, I've always assigned a few people as dedicated small heal spammers (usually 1 Paladin and maybe 1 Shaman for AC buff) and have others utilize larger heals and cancel them halfway through if the tank hasn't taken damage and immediately start a new one. While this style of healing isn't applicable in all BT/Hyjal situations, it is still useful in some fights where the MT damage taken comes in predictable spikes (Azgalor comes specifically to mind).

On other fights with more constant damage though, I typically use R7 Holy Light provided that FoL does not output enough healing to cover the needs of the encounter, with upranking to R9 spam in situations where mana isn't as big of an issue, swapping to max rank as needed for short bursts of added healing. This may change once I get the 4-piece Lightbringer bonus to Flash of Light, but I definitely think that for a Paladin without the 4-piece T5 bonus, R4 is no where near as efficient for output or mana usage as proper use of FoL and higher ranks of Holy Light.

Hi Tiffara,

you said :

Sorry to upset but your downranking penalty is also incorrect...

and you use : ((spell lvl) + 11)/(player lvl))

source : WoW Forums -> Change to Coefficient Bonuses on Spells

So unless what this Bliz guy says is wrong or not up to date... All your calculation that derides from this formula is incorrect :s

Neha.

Like I said, "unless this bliz guy is wrong", and he turned out to be so indeed.
Greeeeeeeeeeeat, you just cannot trust official sources any longer .

thx for the advice, I ll look on wow wiki from now on.

Neha.

PS : btw neha is my nick hey ^^

Sadly, WoWwiki is not my friend. I used to use that formula, but after testing it for myself, I found it to be incorrect. According to my results, (spell_level+11)/70 is the correct formula for a level 70 character. I'll post the data when I get home. I tested with all ranks of FoL and HL (although HL 1-3 also gets the under 20 penalty, WoWwiki's formula for that worked for HL 2-3; HL 1 was just off).

You seem to be right! I only tested rank 6 FoL but it really seems to get a penalty!( My rank 7 FoL did 1805-1866 without any buffs and rank 6 did 1672-1717. I might not have gotten all the end points but it seems pretty close to your 69/70 penalty AFAICT.

If you are using next spell levels instead of just spell levels (since if you get your last rank at level 1 you won't get penalized with a (1+11)/70 multiplier) the formula becomes

( next spell level + 3 ) / player level

However, didn't someone actually test the WoWWiki formula on spells with ranks 6 levels apart and found it correct? Then the formula becomes

(next spell level + 11 - ( next spell level - spell level ) ) / player level

equaling your

( spell level + 11 ) / player level

or 1 when there is no higher spell rank. Or maybe they are just using ad hoc lookup tables.

Just for anyone curious about this, I got the [Libram of Souls Redeemed] just the other night. It turns out to behave exactly as a straight bonus to Blessing of Light, in that the bonus healing is not affected by the casting time coefficient or the downranking penalty, and it does get the benefit of the 3/3 Healing Light talent.