Terminology:
HPS - Healing per second. The output you can get by using only this spell. Or how much output per second you get by casting that spell.
HPM - Healing per mana. The amount of healing done per point of mana spent. Not counting overheal.
Over the last few weeks, a repeated claim on the forums as well as on the official boards is that Heal is worthless. I've been one of the propagators for this claim. And since easter means I had a ton of time to burn, I've been doing some spreadsheeting to verify the validity of this claim. The answer surprised me.. but not in the way I .. uh... didn't expect.
Since I don't have beta access, all this math is based on the status of the live servers. That means this post is cata specific, and its relevance for MoP may or may not be good. Don't be applying these findings to MoP without verification, that's all I'm saying. I will remake this spreadsheet using Beta information once spell costs on the Beta servers finish the mana cost rebalance pass, and we can see what the state of the spell would be in MoP. But for this thread, it's all about Cataclysm!
This is the spreadsheet I used as a base for these arguments. Feel free to play around with it.
Base problem with Heal
Heal heals for roughly a third of GHeal, and cost roughly a third of the mana. Using these ballmarks, Heal is indeed pointless, as instead of casting 3 Heals, you can cast 1 GHeal and idle around for 5 seconds, achieving the exact same result. The only reason Heal is worthwhile is when you are hovering around 0 mana, and Heal allows you to cast that heal sooner than GHeal allows. But at that point, you're royally screwed anyway.
In addition, Heal doesn't heal for remotely enough anywhere, unless you overgear content badly. This was true in early cata heroics, and it is true in the dragon soul. You can use the heal to stem the tide, but you can never use it to stop it. The next time any of us are likely to actually use Heal is in MoP, but that GHeal is equivalent still stands as a big counterargument to even hotkeying the spell.
Claim: Heal is always redundant for a disc priest
For a disc priest, Heal is competing with Penance, Greater Heal, Flash Heal and Power Word: Shield. For completeness I'll also list Renew and Atonement though neither are really considered proper healing alternatives as Disc.
For the comparison, I will be using a standard Disc build for "tank healing", and a (currently) common 10k spellpower + 18 mastery gear setup. The same results are actually present with a 6k spellpower 6 mastery setup as seen in early Cata. Feel free to play around with the spreadsheet and see if you can find some levels of spellpower/mastery that makes my claims invalid.I ignored Haste and Crit in this comparison, and assume this won't change the relative balance between each spell.
Explanation: "Real Output" include tooltip damage, spellpower contribution and most common talents and effects. More details in the spreadsheet.
I did not consider Divine Aegis in this comparison. But as this is applying equally for the spells we care about, it shouldn't matter. Crit tends to double-dip on Atonement, but I'll live with that inaccurracy as well. The biggest problem is that my values for PW: Shield are quite off as I do not count the rather massive mana gains from Rapture, as well as the HPS gain from Borrowed Time. Keep in mind that the value of PW:Shield in the spreadsheet is as such valued notably less than in reality.
But - judging by these results, Heal is very bad when it comes to HPS, being dramatically outdone by anything including Atonement - and more importantly, dramatically outdone by the damage from even the most trivial non-elite level-appropriate quest mobs on the world map. This is not really any surprise, we all knew the heal sucked. The question is really whether it is efficient.
Surprisingly, the HPM value show that it is. But not by a whole lot. Heal is 12% more efficient than GHeal, to be precise. This was far more than I expected. Ultimately however, this is irrelevant, as Atonement is also more efficient than Heal. If you ever want to consider using Heal, you should instead pull an Atonement Holy Fire.
More importantly, these numbers do not account for the rather huge mana saves you will get by abusing Inner Focus. Inner Focus can proc every 45 seconds (every 18th GHeal for free in a spam scenario) and even faster if abusing Train of Thought. Of course, you're not likely to spam GHeal - it's the heal you add in between Penance and shields if you have to. So that means far more than every 18th GHeal is actually free. The number above lists only every 18th heal as free though. And more importantly, Since these GHeals have a larger chance to crit, the HPS and thus also HPM is also going to be massively better than what heal can offer (roughly, 0.5*0.25 extra throughput).
As you can see GHeal + IF's efficiency is actually managing to surpass Heal. A combination of Penance, Shields and Atonement healing - or Penance, Shields and GHeal - will thus always offer both superior efficiency and throughput to ever casting Heal as Disc. Heal is as such one of the worst heals in the disc arsenal! Stop casting it!
I will conclude that Heal is indeed redundant for disc priests. The only actual usecase is as a cheap way to refresh grace. And honestly, that's not a very good validation for a button since Penance really does the same thing, better.
Claim: Heal is always redundant for a Holy priest
On the Holy Side, Heal is competing with GHeal, Flash Heal, Serenity and Renew. I'll be using the same gear setups as we saw in the disc comparison. I'll also list a few AOE Heals to make a point.
I did not consider Serendipity or Test of Faith. But I guess that's okay for the same reasons as not covering all the disc talents. They should act the same for most all of these heals and thus won't impact much when it comes to the final conclusion.
Analysis then. The same results as seen in the Disc side is apparent; Heal is actually efficient compared to GHeal. But as with disc, it's only 12% more so. Since Holy don't have Inner Focus to shift the balance, the simple conclusion is: you should use Heal instead of GHeal if you can manage the risk of doing so. 12% isn't a lot, but at the end of the day it may mean the difference between a wipe and a kill when mana is tight. As long as you avoid the scenario where Heal leads to rampant use of Flash Heal to undo panic you created by casting Heal in the first place, at least.
But...
In raids, a holypriest is rarely going to cast neither GHeal nor Heal, as both are pretty inefficient and mana always is a prime asset to worry about. Most holypriests stay in the Sanctuary chakra state for raiding, where as you can see AoE healing spells are notably more efficient than most anything in the single target arsenal can offer. This is of course also true for a disc priest, but few Holypriests are tank healing outside of heroics these days simply because it's just not an efficient use of the spec. To get the most bang for the buck out the limited Holy manapool, you go for the aoe heals, and (outside of random renew bombs) leave the tank healing to other specs unless it's an emergency. At which point Flash Heal is the only real card available.
So to say that Heal is an efficient heal is a big lie. Sure, it's slightly more efficient than GHeal, and notably more efficient than FHeal and PWS. But that's it. Every other heal in Holy's arsenal is more efficient than Heal. Most are vastly more efficient. The reason Holy even want to use this spell is out of desperation in a pure single target healing scenario where mana is so tight that you cannot even afford to waste or overheal 1/8th of it. That is and will always be a scenario you can outgrow by getting better gear.
I could here make an argument that Surge of Light will have more proc chances on 3x Heal than 1x GHeal, and thus increase Heal's relative efficiency to GHeal. But as we all know, SoL is such a sorry talent that they had to more than triple its effect to make it worthwhile as a MoP talent, and most holypriests avoid it in favour of "solid HPS boosters" like improved death and rapid renewal (note sarcasm). So I'm not gonna bother.
I cannot claim that Heal is useless for Holy. But I will claim that it is not efficient, and I will claim it is not special enough compared to GHeal!
Claim that Heal is useless is invalid for holy then?
Theoretically, yes. But...
A relevant question for Holy: Does Heal need a little something to make it more worth the hotkey slot?
In my opinion, yes. The difference between Heal and GHeal is far too small to even bother arguing about, and as a result almost everyone ends up dropping the spell from their hotbars after decking out in purple gear.
A serious suggestion is to make it completely free of charge. It's still going to be useless in 99% of all content, but at least you won't feel it's not special anymore. Or at least cut the manacost by a notable amount so that it efficient compared to every other heal in the spellbook except Lightwell.
Ultimately though, I don't think that the spell is worth the hassle. The cast time is too long and the spell itself is too boring. The hell-bent design idea that we're supposed to enjoy being restricted to this one spell for the duration of the first content patch is imo a mistake, and placing healers in a situation where they have to stick to Heal simply for the benefit of fulfilling this design idea is causing nothing but frustrations. Healing as Holy is the most fun when you get to select and abuse all your heals, and limiting us to be a one-button healer goes against anything the holypriest stands for. I wouldn't mind scrapping the spell altogether. Reduce the cost of GHeal by 12% and presto - Heal is utterly useless. It's not like Holy lacks casting options, and if there really was any spell that didn't fit in at all in the holy toolkit, it's Heal.