MoP In Review, Part 2: Mana Matters

This opinion-laden post follows on from my previous MoP In Review topic, Healer Representation, and builds on themes I started to explore in the introduction. Go back and catch up there if you’d like 🙂

Remember when we had to do this?

Remember when we had to do this?

In the interests of full disclosure, I have to admit something. When I first started writing this series, I was of the opinion that mana regeneration in Mists was completely bananas – that despite the developers’ stated intentions and best efforts, they had created a scenario in which mana is unlimited, scaling way out of tune with player power and content, and that the overabundance of mana was the root cause of all my other issues in the game. I was ready to rail against just about every mana regen mechanic in the game, to vilify the legendary meta gem, and to burn Spirit at the stake.

And then I had the pleasure of talking about this topic to the inimitable Hamlet while he worked on this post, which is a far more elegant and insightful analysis of the mana regen issue in this expansion than I could possibly have ever made, and the conclusion he came to at the end shifted my perspective. (This is one of my favourite parts of the theorycrafting-slash-commentating process, incidentally, and why it is incredibly valuable that there is a commentator community and not just a bunch of individual voices. It’s great when we all agree, but there’s so much more value in disagreement and coming to understand why we each hold the positions we do. It’s always awesome when someone can change my mind with well-supported logic. But I’m getting off-topic!)

Anyway, as I was saying, after reading that post, I realised that it’s not that bad. In fact, it’s pretty good. We do have more mana than we should, we are less reliant on Spirit than we ought to be, and it has a definitive effect on the endgame. But I can no longer maintain that mana is the entire, or even the primary problem, and that meant I had to rethink exactly what I wanted to say about mana regen in this series; hence the delay. (Why, yes, this is just a long-winded apology! What else would you expect?)

Regardless, Hamlet’s post, and my own experiences and thoughts on mana regen, highlight issues that are relevant enough to us now that they are worth discussing while we are anticipating more Warlords information, so let’s get into it, starting out with a discussion of the evolution of mana regeneration mechanics from Vanilla to today.

Regenerative Design

Disclaimer: Some of this will necessarily be fuzzy, because I didn’t play much in Vanilla, and didn’t start thinking seriously about healing mechanics until the Mists beta, so I’m pulling information from contemporaneous sources to illustrate a point: that throughout the game’s history, mana management for healers has been – or at least, was intended to be – a core component of healing gameplay.

Back in the Dark Ages of WoW, mana regeneration used to be governed by a “five-second rule”, wherein your Spirit-based mana regen was deactivated for five seconds after spending mana. The idea was that this rule would cause players to throttle their own healing throughput based on damage patterns. You could heal full-bore during heavy periods of damage, but only to a point, and when incoming damage was low, you were better off not casting than casting cheap, small heals. Healers were encouraged to evaluate whether a healing spell was really necessary at any given time, and if not, to wait a little longer before casting. The goal was to feel rewarded – with better mana regen – for understanding the damage pattern of an encounter and choosing your behaviour accordingly.

The five-second rule sounds shallow and arcane to us now, but this was the very first incarnation of mana management as a means to control throughput. For healing specs whose mana regeneration was affected by Spirit – everyone except Restoration Shamans and Holy Paladins, I believe – there was some depth to it. Skilful play for these specs involved thoughtful use of clearcasting procs, clever timing of instant casts and channeled spells, and informed decisions on when it was safe to take a little “regen nap”. And these specs had other sources of mana regeneration – e.g. Shadowfiend, Innervate – to prop them up when the five-second rule wasn’t enough.

However, gear (and gems, introduced during The Burning Crusade expansion) with the “mp5” stat would provide regen throughout an encounter regardless of your casting-or-not status, smoothing the effects of the five-second rule for Spirit-loving specs and allowing Shaman and Paladin healers to ignore the rule’s existence entirely. In fact, I was shocked to find that the five-second rule actually persisted from Vanilla all the way until Cataclysm’s pre-patch, because the rule’s existence never impacted my play at all. I’m not even sure I knew about it (despite raiding on my Restoration Druid pretty heavily throughout Wrath)!

With Cataclysm came a revamp of mana mechanics, allowing DPS to have essentially infinite mana so long as they played their class properly, and eliminating entirely the five-second rule and the mp5 stat. In its place, healers received passive abilities that allowed them to regenerate a reduced amount of their out-of-combat mana regen during combat (whether casting or not). The out-of-combat mana regen was based (as it always had been?) on Spirit and the square root of Intellect, so there were two ways to increase your in-combat mana return in Cataclysm – improving your Spirit, and improving your Intellect.

Unfortunately, there were many other mana-return mechanics that scaled off your maximum mana pool – which Intellect also increased – and no healer specs received Spell Power benefit from Spirit, so Intellect-stacking tended to be the way to go for both mana regeneration and throughput purposes. Especially for classes with “active” mana regeneration – Paladins with Seal of Insight melee, Discipline Priests with Rapture, and Shamans with the Telluric Currents talent – the expansion was incredibly lopsided.

At the start of Cataclysm, because we had low Spirit and low Intellect values, our mana regeneration was quite limited, and many players felt frustrated by the expansion’s early healing challenges (especially given the combination of mechanically challenging Heroic dungeons and the LFD tool, and some early balance issues with 10-player Heroic raiding). For me, though, this was the absolute best time to be a healer in my entire WoW career – it was the first time I ever had to think about anything besides pumping out as many Chain Heals as possible (ah, Mount Hyjal…) and encounters seemed designed to create interesting choices in spell selection and targeting.

I used to Lightning Bolt this guy so much.

I used to Lightning Bolt this guy so much.

By the end of the expansion, with our gear improvements – from 346 heroic dungeon blues to 410/416 heroic Dragon Soul gear – we were all swimming in mana. Intellect and Spirit on our gear increased by 80% just from item level inflation, leading to a nearly 150% increase in our Spirit-based mana regeneration, expanding our mana pool significantly, and increasing the amount of mana we regained from mana-pool-based effects.

These Int-based sources of mana regen are quite difficult to quantify in retrospect like this, but we probably had at least double the amount of mana available to us throughout an encounter in Dragon Soul than we had in early tier 11 content. (With Telluric Currents, in fact, I’d say I had at least three times as much mana to spend in a Dragon Soul fight than I had against, say, Heroic Maloriak in t11. On Heroic Spine, I was able to refill my mana bar from empty 4 times by Lightning Bolting the first appearance of each Burning Tendon alone.) This led to a strange imbalance where healers just starting out had a lot of trouble in the content available to them, while geared healers doing their end-game content found mana too easy to manage, such that the game got easier (in that regard, at least) as you progressed to harder and harder content.

The developers recognised the strangeness of this situation, and that led us to Mists, where mana regeneration was revamped yet again. Intellect now plays no role in mana regeneration. All characters have a base in-combat mana regen of 6000 mp5, Spirit adds to that, and most mana regeneration abilities operate on Spirit rather than maximum mana pool (Hymn of Hope/Shadowfiend being the two exceptions).

These changes were made with the intent of giving healers in entry-level gear an easier time handling their intended content (Heroic dungeons and LFR) while preserving the mana management game for well-geared raiders.  And in retrospect, they did all right; as Hamlet notes in his post, our stat inflation from gear this expansion granted us a 200% increase in Spirit, but we saw less than a 100% increase in our total available mana over the course of an encounter from tier 14 to tier 16. But mana management still is a bit too easy in t16 content – at least, in 25-player mode, which I’ve been arguing is the format we should be looking at when speculating about Warlords and Mythic raiding.


Why is it so important for healers to manage their mana? What is wrong with properly-played healers just having infinite mana, much in the way a properly-played damage-dealer has these days?

I’d argue that it is because the healing game is quite different to the DPS game. Damage dealt is always useful in an encounter, but not every encounter feels the same; the variation in DPS playstyle comes in with mechanics that must be worked around and technical skills like kiting, performing utility functions whilst still maintaining high throughput, target-switching, etc. There are very, very few situations in which there no targets to whom damage dealt is useful. Even consider this tier’s General Nazgrim, during Defensive Stance – yes, there are negative effects that arise from dealing damage to the boss, but if your raid can withstand those negative effects, the damage you deal to the boss is still valuable. He still dies sooner.

In comparison to the complexity of coping with all the mechanics – not just the encounter itself, but also each class’s abilities, both of which are getting more convoluted with each expansion – having to deal with mana management would be just an additional chore. And since damage dealt is always useful, there’s no way to spin “I can’t deal damage right now because I’m worried about my mana” into an interesting mini-game. It’s never going to be fun for a DPS player to choose doing nothing over dealing damage, and this is why so many DPS hate Nazgrim’s Defensive Stance – their raid demands that they make this choice, and it isn’t pleasant.

(I recognise there are Energy- and Rage-using DPS classes that are frowning at me a little bit right now, because they do occasionally have to pool their resources, and only auto-attacking is pretty similar to doing nothing, but the rate of regen of these resources is much higher than it is for mana users, so you’re not as screwed for as long if you mess it up. Now hush, I’m trying to write a healing blog here 😉 )

On the contrary, unless every encounter has an identical damage profile – constant AoE – then healing throughput is not always useful, nor should it be. An expansion of “all Garalon all the time” would be extremely boring. The healing game is, and should be, kept interesting by giving us varying damage patterns to heal back up, which sometimes means throttling our throughput now to be able to burst later. And the best way to accomplish this is by making our mana management an important, attention-demanding part of our play.

Here’s the problems I can see with granting healers infinite mana:

  • Our healing choices become solely dictated by throughput;
  • We would settle into a max-throughput rotation very quickly, and not have much reason to deviate from it;
  • Classes with overheal recovery (e.g. Divine Aegis, Illuminated Healing, and to a much lesser extent, Wild Mushroom: Bloom) would always do better than classes without it, when mana is infinite;
  • We reduce our ability to modify our playstyle via our gearing strategy.

Nearly all of these come down to removing elements of choice and decision-making from our role. At the heart of it, healing is an optimisation and pattern-matching puzzle game. Good play should involve assessing the current threat to the raid, thinking about what is coming up, reacting to what has just happened, and choosing the right spells, all within this resource management framework (whether that be mana alone, or secondary resources like Holy Power and Chi, or the quasi-secondary resources like Tidal Waves or Evangelism). And like a puzzle game, if the puzzle never varies, where is the game?

This is not a game

Infinite mana: This is not a game. P.S.: I-blocks are clearly the Tetris equivalent of Prayer of Healing 😛

Because this is difficult to illustrate in WoW, I’ll make an analogy to my favourite non-MMO: Tetris. It, like healing, is unmistakably a pattern-matching and optimisation game, in which the varying pieces are similar – in my tortured simile – to the various spells in our toolkit.

When we heal with infinite mana, all we must do is choose the highest-throughput spells and cast them repeatedly. It looks rather a lot like this Tetris game to the right — high “score”, but low on “puzzle” and “game” quotient. If the right answer is always the I-block, the only way you can go wrong is by lying the I-block down on its side instead of filling the very, very obvious gap – which is a lot like saying “Healing Rain takes a lot of skill to use because sometimes you can accidentally cast it on the ceiling.” Yes, there’s a modicum of skill in not casting Ceiling Rain, but that is a pretty low bar to hurdle 🙂

Mana starvation: Toolkit limitations make for slow and unsatisfying gameplay

Mana starvation: Toolkit limitations make for slow and unsatisfying gameplay

Mind you, I’m not trying to advocate that we should be mana-starved, either. When we are mana-starved, we resort to casting only our most efficient spells, and these are often difficult to adapt to the ever-changing landscape that is an encounter’s damage pattern. With limited and inadequate tools, healing starts to feel more like the picture to the left here – yes, you can technically get things done, but it’s slow and frustrating, and while skill can be a serious determinant of your “score”, for many players we are losing out on the “fun” factor, which is still a very important part of gaming.

Ideal gameplay: Varied ability usage and clever optimisation leading to high "scores"

Ideal gameplay: Varied ability usage and clever optimisation leading to high “scores”

What I want to see out of the healing game is more like this picture: A place in the game for all the tools in our toolkit, where encounter knowledge (which I guess in this example is analogous to the “next” column) can be combined with pattern-matching and resource optimisation skills to produce results that are more strongly predicated on player capability. The reason we don’t see this in 25-player raids today is linked in some part to mana abundance – when you can afford to cast your AoE spells even when only one or two players are injured, there’s little incentive to cast single-target heals instead.

So that’s the kind of game I want to play. It has existed for me twice so far in my 25-player raiding career – tier 11 and tier 14. I could wax poetic for hours over the pure joy of healing Heroic Maloriak, while anything I have pleasant to say about healing Heroic Norushen, for example, is related strictly to the magnitude of the numbers on my healing meters. It’s no surprise that we have to use more of our abilities more intelligently in early tiers of an expansion, but it is a bit counterintuitive that as our gear improves, and as we move into harder and more demanding content, we enter a strange dimension where our spell selection matters less. Right now, the only way I can go OOM is by pouring single-target heals into a player who is in dire need of healing – and doesn’t that just seem a little off?

And let’s not forget that excess mana and encounter design are linked. In order to keep encounters challenging for healers as they gear up, damage has to keep increasing. And our healing throughput doesn’t increase as a simple linear effect as our gear improves – secondary stats are multiplicative with one another, and with Spell Power, and as we have more mana available to us, we can spend more mana on additional spells or on replacing efficient-but-weak heals with inefficient-but-strong heals.

So our gear improves, and the damage output of bosses increases to compensate. But because player health is finite, there’s not that many degrees of freedom for developers to play with here. Increasing the damage of single-target attacks can only go so far, before it becomes impossible to heal through them. So attacks instead choose more targets, which drives more AoE healing, which drives heavier AoE damage into the next tier, or the attacks occur more frequently, which mandates more use of damage reduction and healing throughput cooldowns, which breeds resentment amongst classes that don’t have them, which leads to cooldown proliferation, which leads to NUCLEAR LAUNCH DETECTED (Dayani begins to cast Healing Rain).

Everything is connected. There is no one problem, and no one solution. But having the mana to be able to cast literally whatever we want, whenever we want it, is certainly a part of it. And while this has been better in Mists, an expansion that started off with developers redesigning mana regeneration specifically for that purpose, than it was in Cataclysm, I do want to note that there is one large misstep – the legendary meta gem (LMG).

Legendary Mana Gem

I like active mana regen. I like being rewarded for paying attention to something. So you’d think that I would like the LMG, given that it is a proc to which you can respond in order to gain mana-free healing or directly generate a chunk of mana. But I don’t actually like it – and it took me a long time to figure out why.

Basically, because the LMG operates on a proc basis, it is pretty much agnostic to encounter design. What I mean by this is, it often procs at times where you need neither the mana, nor the healing. This is a pretty shallow mechanic to me. There’s nothing you can do in this situation but either waste the proc or generate meaningless healing. Granted, some classes have mechanics by which that “meaningless” healing becomes partially meaningful – Divine Aegis/Power Word: Shield and Illuminated Healing, for example, ensure that Disc Priests and Paladins who cast during their meta gem procs will see some benefit to their throughput even if the actual healing dealt was 100% overheal. For Restoration Druids, an LMG proc whilst the raid is on full health is simply mana-free fodder for your Wild Mushroom. But for the most part, the LMG’s behaviour inelegantly encourages us to spam our heals when they are not needed, on the off-chance that they will be needed by the time we’re done with our casts (or, for lingering effects like IH/DA/PW:S and Rejuvenation, shortly thereafter).

The only healer for whom I do not feel this is particularly problematic is the Mistweaver. It’s actually quite awesome how much flexibility this spec has in choosing to respond to its LMG proc. Need mana? Jab/Tiger Palm/Jab. Need Chi? Soothing/Surging spam. Need emergency healing? Spam that Healing Sphere. Don’t need any of this, just want to DPS? Crackling Jade Lightning. Since the damage that you can deal during this latter scenario actually feels worthwhile, and will continue to heal should unexpected damage occur while you’re channeling the spell, this feels like a solid slate of choices. (By comparison, a single Wrath, Lightning Bolt, or Denounce, which is about all you could fit in the window, is extremely lackluster.)

So as Hamlet mentions in his post, the LMG isn’t necessarily – and doesn’t have to be – overpowered. The same effect could have been tuned such that it wasn’t worth quite so much of our overall mana pool. But it certainly is not an interesting method of mana conservation for 5 of the 6 healing specs, and it threatens the delicate balance the developers worked so hard to create in revamping our mana mechanics.

Passive Aggression

Something you’ll also notice from reading Hamlet’s post is that the effect of Spirit from gear on our overall mana pool is pretty low. 26% of the mana we could spend over the course of a 6-minute fight originated from Spirit on gear (under his model of a 12k-Spirit-wearing healer). That’s not because there isn’t a lot of Spirit on our gear – there’s heaps! It’s because the Spirit-to-mana-regen conversion rate is quite low. This is by design, so that healers in early stages of the gearing process do not feel so limited by mana regen as they did at the beginning of Cataclysm, and so that healers at the later stages are not swimming in Spirit-based mana regen.

But what it does result in is your Spirit gear making less difference than the other sources of passive mana regeneration you have. For example, the base mana regen of 6,000 mp5 is the equivalent of a little more than 10,600 Spirit. A trinket such as Heroic Dysmorphic Samophlange of Discontinuity provides an average of (2750 Spirit times your Haste), and for everyone except Disc Priests, we’re running at least 30% Haste, so you’re looking at around a 3500 Spirit benefit. At 12k Spirit, the healer Amp trinket is another 1080. The legendary meta gem is worth around 7,000 Spirit. And if you have a similarly geared Resto Shaman in your raid, you’re getting 2100 Spirit equivalence from their Mana Tide Totem.

I used to be a contender :-(

I used to be a contender 😦

Against this backdrop of more than 24000 “Spirit equivalence” just from passive effects alone, Spirit on your gear is of diminished importance, and since we can’t very well opt out of our base regen, our legendary meta gem (I am aware that some do, but in general healers use the healing LMG), or external mana regen sources, we opt out of Spirit on gear and reforge it into throughput stats instead.

Just as a little thought experiment, let’s say that we took our 12,000-Spirit-having Druid healer and gave him 50% more Spirit on his gear. His Spirit-based regen would now account for 34% of his overall mana budget, up from 26%. His total mana budget increases by around 12% – 230k more mana to spend over that 6 minute fight. But in order to achieve the extra 6,000 Spirit, we have to take away 6,000 in other secondary stats, and we’re likely to find that makes a larger difference in our throughput (especially since 12% additional mana is not even remotely correlated to a 12% increase in throughput, but -12.5% Mastery or -10% Crit for that Resto Druid would be pretty huge). This is why more Spirit is not the answer to your healing problems, most definitely not now 😉

In fact, the picture could be made even bleaker if we convert class-specific mana regen or refund abilities into Spirit equivalence. Considering each class running 12k Spirit (for comparison purposes), I’ve calculated the Spirit equivalence of each class’s mana conservation techniques. I won’t expand too much on how I got these numbers, because it’s well beyond the scope of this article, but basically:

Restoration-Druid-IconInnervate is equivalent to 2955 Spirit when used on CD.

Omen of Clarity procs are trickier. At 95% chance to proc from a 16-tick Lifebloom, we can estimate 1 proc per 15 seconds.

Assume every proc is used on a Regrowth; there are two extremes. The conservative estimate is that each Regrowth’s healing itself is useless – and thus, Regrowth is not a spell you normally would have cast, so you’re not really “saving” any mana by casting a free one. The only reason you’re casting the Regrowth is to refresh Lifebloom, so you’re really only saving the cost of one Lifebloom every 15 seconds. This is equivalent to 2100 Spirit.

If you always use your OoC proc to cast a Regrowth on a target for whom the entire spell is valuable, well, Regrowth still probably wasn’t the spell you were going to use, it was probably going to be Rejuvenation. So you’re replacing a Rejuvenation cast with a free Regrowth, saving 8700 mana every 15 seconds. Clearcasting used in this way is equivalent to 5141 Spirit.

If you always use your OoC proc to cast a Regrowth on a tank who is taking a lot of damage, and you therefore would have been casting a Regrowth on them anyway, then Clearcasting is worth 10500 Spirit. In practice, sometimes your Regrowth will be fully effective and something you were going to cast anyway – on the tank – but most of the time, it will be cast solely to refresh Lifebloom or as a mana-free replacement for a Rejuvenation cast, and the real value is going to lie much closer to 2100 Spirit than 10500 Spirit.

mistweaver monk iconMana Tea is equivalent to up to 10600 Spirit (same as base mana regen) if the Monk uses one charge at least every 10 seconds. This estimate was taken from my raid’s Mistweaver on Heroic Malkorok (a situation where constant Chi generation and expenditure is encouraged).

holy-paladin-iconDivine Plea for a Paladin with 12,000 Spirit is equivalent to 3600 Spirit on average when used on CD.

Notice that classes with secondary resources have fewer of these class-specific sources of mana regeneration – that’s because heals relying on secondaries are properly considered “part of” the mana-costing casts, making the mana-costing heals far more efficient than they would appear just from looking at their tooltip, and allowing these classes to ‘store’ parts of that healing for later, similar to how healers used to store mana during lulls so they could spend it more freely during burst periods.

discipline-priest-iconMindbender used on CD (without Hymn of Hope overlap included) is worth 7100 Spirit.

Rapture once every 20 seconds – which is a fairly poor Rapture rate – is the equivalent of 6100 Spirit. (Every 15 seconds, which ought to be achievable, is equivalent to 8100 Spirit.)

Evangelism’s mana discount is difficult to evaluate because on some encounters, there’s a lot of Smite spam, and on other encounters, there is less. I’ll look at the worst-case scenario for Evangelism – only enough damage spells cast to achieve 5 stacks every 30 sec for the purposes of using Archangel, with Holy Fire and Penance being the primary stack generators. In this situation, Evangelism is equivalent to 6360 Spirit.

restoration-shaman-iconWater Shield’s passive mp5 is the equivalent of approximately 3790 Spirit.

Resurgence (estimated from a Heroic Norushen log) is worth approximately 3910 Spirit.

Glyphed Totemic Recall of Healing Stream Totems when no Elementals are active, over the course of a 6 minute fight, is equivalent to 4335 Spirit.

If you consider these sources of mana conservation and their Spirit equivalence, you can see pretty readily that Spirit on gear is a weak link in your mana allotment. It’s not entirely kosher to imagine some of these sources of mana regen in this way – particularly Resurgence and Mana Tea where the rate of return depends upon the rate of casting, and thus on mana expenditure – but these are just examples of the strength of our class mechanics compared to our regen from gear alone.

Again, this fits into the plan of making us less gear-dependent for our mana regeneration, so that entry-level healers did not find content too taxing. But I think it’s possible that the pendulum may have swung too far in that direction, as a bit of an over-reaction to the very vocal complaints about the difficulty of early Cataclysm healing (and particularly, undergeared healers venturing into challenging Heroic 5-player content with the random match-making system). As it stands now, using your class’s mana-refunding abilities well and maximising your LMG usage pretty much dwarfs the benefits of adding an extra piece of Spirit gear or changing your reforging, and I think that’s a bit of a wasted opportunity.

While a rare few of these refund or mechanics are more fun than passive Spirit-based regen, particularly Rapture, Evangelism, and Mana Tea, most are equally shallow. Nobody ever got excited about Water Shield’s passive bonus to mp5, for example, and Divine Plea is pretty dull now that you don’t even have to worry about using it during downtimes so as not to lose out on throughput. For many of these mechanics there is no fundamental difference between them and Spirit-based mana regen, and it begs the question of why so many functionally identical but semantically separate regeneration abilities need to exist.

Genuine, Bona-fide, Electrified, 6.0 Mana-rail

Okay, we know that there are some gear changes coming to us in Warlords of Draenor, so before we get too far down the rabbit-hole of my hopes and dreams, let’s continue the train of thought from above and talk about Spirit on gear and how it’s going to work in the upcoming expansion.

The gear philosophy for WoD is laid out pretty well in this forum post; for healers, the most notable information to take away is that Spirit is going to be available on a vastly reduced number of items. Instead of us feeling like we’re stuck with heaps of the stat on every item we pick up, Spirit will now be more rare, and ideally, more coveted and valuable for healers. This may mean that Spirit gems and enchants are more important for healers in WoD, and with the Spirit-to-mp5 conversion increasing, I hope it will address my reservations about the diminished effect of gearing on playstyle. I recognise it’s going to take a fine balance to make Spirit on gear meaningful without tipping the scales in the other direction, and making mana too plentiful at the end of the expansion, though, so perhaps the answer to this is to provide fewer Spirit-bearing items in later tiers.

With that out of the way, I just want to briefly touch on a few things I’d like to see from mana regeneration concepts in Warlords. Since we don’t have a lot of information about how things are changing in the upcoming expansion, I won’t spend too much time agonising over the details here – just some friendly little dot points:

  • No LMG-type boost to mana regeneration in the middle of the expansion, ensuring a smoother and less drastic increase in our spendable mana as we gear up.
  • Many mana regeneration abilities need some sort of … I don’t know. Oomph? Je ne sais quoi? Sparkleponies? Something to make them less dull.
  • AoE healing spells should be less mana-efficient than they are now.
  • Single-target healing spells should be more mana-efficient than they are now.
  • Total available mana over the course of the expansion should increase perhaps by half of what our stats-on-gear do, to keep mana a constant concern, without ever necessarily being a constraint.

On these last three points, I’d like to again return to Hamlet’s discussion in the Mana Economy post I linked earlier. A central theme to his analysis of the mana situation is the mana curve (see here if you’ve accidentally closed the tab or something) – the abstract representation of how your throughput increases as you gain more mana to spend. In his idealised scenario, at low mana levels healers should still be able to lean on their core, CD-bearing, rotational heals – using the examples he uses throughout that post, these are spells like Wild Growth and Swiftmend – and Spirit becomes fun and healthy when we’re able to cast those CD-bearing spells as close to on CD as is required by the encounter we’re attempting.

He and I differ in opinion here, and I won’t attempt to reflect the content of our discussions here as I’ll likely do him an injustice with my portrayal. I’ll just say that I, personally, disagree somewhat. I found MoP to be the most fun in tier 14, when I literally couldn’t cast Healing Rain on CD, even if the encounter wanted it (like Garalon did). Here, decisions mattered, cooldowns were planned more carefully, and I simply felt more like I was playing a puzzle game than a “game” of I-block-organisation. And while I recognise that my experiences and opinions are coloured by a lot of baggage, because 25-player Resto Shaman playstyle is regrettably shallow and Healing Rain’s strength is a large part of this, I just don’t believe that there should be rotational AoE healing at all. (I have absolutely no problem with single-target rotational healing like Riptide, Penance, Swiftmend, Expel Harm, or Holy Shock – spells which can require targeting and thought to use.)

Part of what can restrain us from having rotational AoE spells is their mana cost. And I don’t believe that expensive, CD-limited AoE healing is a bar to entry-level healers the way expensive single-target heals can be, so long as there are also mana-efficient ways to achieve similar outcomes. After all, entry-level healers are mostly going to be doing content like normal or Heroic dungeons – where there are only five party members, so AoE healing is not as important as it is in raids, and we can make do with Chain Heal, or glyphed Riptide, without needing to be able to cast Healing Rain for every trash pull – or LFR, which we recently learned is going to be tuned extremely forgivingly for players in normal dungeon gear.

do believe that cheap AoE healing leads to its overuse in raid situations – and particularly so in the large-format sorts of raiding that are going to be prevalent in WoD. I would hope for an expansion where your decision to cast your CD-limited AoE heals (CoH, Healing Rain, Wild Growth, and those limited by ersatz CDs in the form of secondary resource requirements, like Uplift or Light of Dawn) is influenced more by the damage pattern of the encounter than by the simple fact that it’s more efficient to do so than to bother with single-target healing. I’d like to see us being able to use these spells literally on CD by the end of the expansion, not somewhere in the midst of it.


There are many ways to achieve the goal I’m after, and not all of them (perhaps not even most) are linked to mana regen. I’ll be discussing the other aspects of healing that contribute to this overall 25-player landscape of AoE healing spam and the degradation of decision-making in future posts in this series (although next up is a segment on Absorption healing, since I already have it mostly-written, and I’m hoping that we’ll get more information about Warlords’  healing paradigm before I post about spell design).

For those of you who raid in 10-player formats, a lot of this might sound pretty foreign to you. I raid 10s on a Monk and a Druid, and on my Druid I do occasionally not cast something if the current raid status doesn’t call for it. My Monk has no concerns whatsoever with mana, but then, as I noted above, Mana Tea is pretty strong, plus the Chi system allows me to ‘generate’ healing during periods of mana surplus and ‘store’ it for times of mana drought, and the LMG’s interaction with my class mechanics are actually interesting. I’m not at all advancing the argument that mana doesn’t matter in 10-player raiding – the core problem here is with AoE spells and their increased importance in 25-player raids, and the whole reason I’m reminding you of this is that with all raiding transitioning to either a flexible or 20-player format, you’re likely to see my issues crop up in your own raiding future.

But I hope that I’ve convinced you somewhat that mana management ought to be an important part of healing gameplay, that it isn’t terribly demanding on our attention right now in larger raids, and that a return to the mana-conscious, spell-diverse healing style of early Cataclysm and early Mists doesn’t have to feel punitive. I’d love to hear from you on what you think – about mana in general, about the LMG, about my silly opinions, and what you hope for in Warlords to make healing a rewarding and enjoyable experience for you 🙂


About Dedralie

Stuff about me!
This entry was posted in Cataclysm, Discipline, Druid, Dungeons, Holy, Mists of Pandaria, Monk, Paladin, Philosophy, Priests, Raids, Shaman, Theorycrafting, Warlords of Draenor and tagged . Bookmark the permalink.

12 Responses to MoP In Review, Part 2: Mana Matters

  1. Volladon says:

    Why does mana regeneration have to change due to gear at all? Why does spell selection have to change depending on what gear you have? If the challenge comes from limited mana, why can’t that remain the same throughout an expansion? Throughput (int+secondaries) can absolutely be the way we “get stronger” throughout an xpac.

    I do agree with you on this, and it’s a big reason why I’ve left 25 man healing, but I really enjoy picking the right spell for the right occasion. Mindlessly spamming rain/stream on CD is pretty boring.

    It’s actually a huge pet peeve of mine for rage tanking as well. The game completely changes because we have a completely different amount of resources comparing T14-16. I would much prefer picking a middle ground (not so starved as T14 and not so abundant as T16) and sticking with that the whole time.

    I think that’s why ToT felt so good for me!

  2. Ashunera says:

    The 5-second rule was arguably more important to Shaman in Vanilla than other healers. Priests and Druids had a talent that allowed 15% of mana regeneration to continue while casting (casting meaning within 5 seconds of spending mana, you could channel spells over and over as long as you didn’t allow them to complete – useful for tank healing) and both classes also had a tier 2 set bonus that allowed an additional 15% to continue while casting. Because of that, they’d see less of a delta between managing it and ignoring it than a Shaman might (MP5 gear was pretty rare and very weak until later on). Aside from a low-level crafted mail set (Green Dragonscale), Shaman had nothing like this, and was basically required to maximize the 5-second rule to have similar longevity as other healers. Paladins, on the other hand, could gear for intellect and crit due to Illumination (which was 100% mana return until Ulduar) and spam a super efficient flash of light which had ultra low base healing but normal “spell power” scaling (at the time) and basically never go out of mana… if their guilds would allow them to take the necessary cloth and leather gear.

    I think it’s important to emphasize that tier 11’s damage, healing, and mana models received a bad rap from a lot of healers without understanding that the healing model wasn’t what they were actually upset about. The content’s difficulty being higher than people were used to is entirely to blame, even if mana mattering was an adjustment from the ICC era. People had a hard time healing and maintaining mana **because the content was hard**, not because there was anything wrong with how healing worked at the time (though I’d argue Moonkin Tranquility was far too important for heroic raiding).

    As I’ve been saying for, literally, years, single-target heals should be your go-to healing buttons. They should be more efficient than AoE spells, period. AoE spells should sacrifice a significant amount of efficiency for higher throughput. Sources of healing from damage dealers (which are basically “free” as far as mana pools go) need to either not exist, or come with a significant DPS cost so it doesn’t just automatically happen.

  3. Samantha says:

    Another great post about healing, I’ve thoroughly enjoyed myself.
    Like you said on the last post, it’s interesting to see what others’ experiences are. It might be because of my Penance glyph that I never remove, but I’ve been fairly uncomfortable with my mana pool the entire expansion. That might not be indicative of a problem but just that I like to have a better cushion than I do, but it has certainly made the healing game more interesting and that I think has made this expansion better overall for it. I was a little frustrated with mana cap at the beginning of the expansion (for raiding, not for dungeons), but I also started the expansion on a holy paladin and switched back to my discipline priest for ToT. Depending on the way they change abilities and mana regen in WoD, I’m excited to see how it will feel. I’m assuming that since I take that Penance glyph that increases the mana cost of a spell I use quite a bit, that it has made my mana pool overall less infinite, but not so much that it affects me negatively.

    I agree with Ashunera above about the early Cata raiding content and it not being a problem with healing but a problem with people not liking the difficulty of the content. I really enjoyed healing tier 11 raids, and it was an interesting jump since my raid group never made it to Firelands before tier 13 came out (and didn’t finish it until well after completing Dragon Soul). I’m interested to see how they work with it, especially with some of the racial ability changes (for instance, I play a gnome priest, so she has maybe about 15k more mana than anyone else, humans have extra Spirit, etc.) and how it feels once we start testing in beta.

  4. sebastian says:

    Hi, i don’t really know where the following questions, so i ll just post them here :)) i would find it really useful if you could answer.

    1) HST over Ancestral guidance or does it really depend from boss to boss,
    2)I’ve noticed you use Elemental Mastery instead of AS is this because of mastery build?
    3)Legendary meta really worth loss of 3% from primal?
    4) here is my armoury any other tips would be very much appreciated.

    Thank you for reading.

    • Dedralie says:

      Hi Sebastian, sorry for the delay.

      1) Depends on what your goal is. Mine is always to deliver healing in a manner that matches the outgoing damage of an encounter. Sometimes I achieve that by using Rushing Streams (particularly on fights like Iron Juggernaut, Paragons of the Klaxxi, and Sha of Pride) and sometimes I achieve it by using Ancestral Guidance (particularly on fights like Norushen, Garrosh, and Thok).

      Ancestral Guidance is very difficult to use to its maximum potential and requires a burst damage event or long bouts of sustained AoE damage in order to be close. Whilst it can theoretically deal more healing than RS in a given 2-minute time slice, it’s rare that it will.

      If your goal is more to buff your HPS and look “better” on the meters, then you can pretty much stick with RS for everything. Personally, I am not interested in meter-based performance evaluations, and the type of healing that RS does is what I call “maintenance” healing, healing up of small amounts of damage that aren’t lethal and never would be (because some other smart or splash healing effect would grab that little chunk of HP if your HST wasn’t there doing it). This is why for fights that have burst damage during stacked, stationary phases I prefer AG – it may heal for less in practice, but the healing is in response to life-threatening damage, and I will always always prefer to heal harder during the scary parts. 🙂

      2) It’s because I use EM during the Garrosh encounter to buff my CDs (see above paragraph XD). But I run at 10118 Haste (the 30% breakpoint without AS) so that I can swap between EM and Echo of the Elements on a per-fight basis. AS is great, but with the Amplification trinket I just don’t feel I need the passive Haste boost, and in my 25-player raiding group the instant heal from AS just isn’t as valuable to me as it could be – it’s almost always sniped by a stupid disc priest’s stupid Atonement 😉

      I gear Mastery for progression content and because I believe my raid role is to provide strong healing CDs that save the raid when they need saving. While I’ve finished the tier, I am just too lazy to go back and re-gem for haste and crit. If you’re not doing progression content where players are regularly dropping below around 50% HP, then you will probably be better off sticking with a high haste build, meaning you should still use AS.

      3) Absolutely. LMG is worth around 7,000 Spirit, meaning you can drop quite a bit and reforge it into throughput stats. Meanwhile, the 3% crit healing from the Burning Primal Diamond – which actually leads to a 6% improvement in the size of your critical heals – is equivalent to around 3,200 crit rating. Plus it gives more Intellect than the BPD does, so it is a clear winner. Use the LMG and reforge that Spirit into Crit 🙂

      Hope this helps, sorry for lots of words, but I’m sure you’ve noticed that “lots of words” is what I do best 😉

  5. Jasyla says:

    I agree with pretty much everything you say here. I actually do kind of miss the 5 second rule. Deciding not to heal might not seem like a fun option for many (like deciding not to dps sucks), but having to make decisions is something I miss. I think an excess of mana can make for lazy players too (who cares if my LB isn’t up? I don’t need any more OoC procs).

    However, when it comes to laying blame for the current healing game being dull and thoughtless, I rarely look at the mana abundance situation. For me, the first problem is encounter design. 95% of fights are all raid-wide damage, all the time. Of course people are going to lean on their mindless multi-target healing spells. Tank healing is practically a thing of the past, as even those who get specifically assigned to it (if that assignment is even needed) rarely cast actual heals on the tanks.There aren’t really fights anymore like Brutallus where the tank needed to be bombed with heals, and dedicated healers were needed to take care of Burns (plus, there was still raid Cleave damage too), or nasty debuffs like Doomfire that caused healers to pour heals into a few people while the rest of the raid’s health was relatively stable. The last time dispels required thought was T11 (and the dispel cooldown really didn’t add anything to the game).

    The second thing I blame is healer toolkits. Overabundance of absorbs, spells that require more action on the rest of the raid’s part than our own (stand in the blue/green/gold circle, go grab that monk ball, click the lightwell), and, of course, smart heals, which make dumb players. I’m sure you’ll talk about this stuff in later posts – can’t wait to read them.

    Another thought (tangent) that just came to mind – how do we measure how powerful a healer feels? A big part of the game is acquiring new gear that makes us stronger. I remember playing Oblivion and quitting because though I leveled up and got new gear, the monsters leveled up with me so I never felt any more powerful than I did at the beginning of the game. There was no sense of progression. In WoW tanks get more health and avoidance, and can live longer on their own (in spite of the healers), DPS see their numbers get much bigger. When I upgraded my Hunter’s weapon from a normal ToT weapon to a normal Siege weapon, my dps skyrocketed. Healers don’t really get that though. An increase in ilvl or a shiny new staff does not necessarily make a healer’s numbers any better, as the amount available to be healed is not just based on our performance. As a tier wears on, bosses die faster, people take less damage and have more health, a healer can actually feel less powerful as there is less to do (unless half the healers get benched). If I think about it, the only way I’ve felt more powerful as a healer is that I can now toss around high cost healing spells with wild abandon, whereas I could not at the beginning of the xpac. If that gets taken away will healers ever feel more powerful, like their gear upgrades and experience are doing anything, over the course of the game?

    • Dedralie says:

      Hey Jasyla, thanks for your thoughts! I will cover those questions you brought up in your first two paragraphs in future posts, just wanted to get this mana one out of the way while I was still reeling from reading Hamlet’s analysis 😛

      I guess this is the same for all specs, whether they are DPS or healers – as you gear up, you not only make your spells/abilities more powerful (through added primary stats/AP/SP, as well as Crit and Mastery) but you generate more resources, and are thus able to use your harder-hitting spells more often. The problem with this for healers is that there is a finite amount of damage that can be dealt to the raid in a fight, so once you have “enough” stats to cope with that, there’s nowhere else to go. And I think I mentioned this above, that it ties in to encounter design somewhat – if they have to challenge our mana bars, after they have given us more and more mana, they’ve gotta resort to raid-wide AoE damage spam.

      I don’t want to eliminate the feeling of progression in gear, and I do think it is important that we feel we can use our better abilities more frequently as we progress through an expansion. But I think it’s important to note that I hit that point somewhere mid-ToT, where I could use my big exciting spells on CD without suffering from mana problems. As someone who really doesn’t care to monitor my progress by meters, and as a class who never needs to think about who to target, I don’t have a lot that keeps me engaged in fights.

  6. Rust says:

    Continuing the Twitter discussion:

    As I understand it, the issues are that AoE, Absorbs and Smart Heals are undercosted, leading to a scenario where Dayani asks “Right now, the only way I can go OOM is by pouring single-target heals into a player who is in dire need of healing – and doesn’t that just seem a little off?” I may be playing a bit of Devil’s Advocate but I see this as a plausible design choice in a world where Blizzard has decided that pure triage healing isn’t what they want, because it leads to a Whac-a-Mole playstyle that they don’t particularly like.

    In this world, Blizzard wants healers to adopt a rotational playstyle some of the time, much like how DPS have a rotation or set priority and react to procs or changes in resources, etc. Fight mechanics are structured then, in a way to make use of the common healer toolkit – raidwide damage goes out, AoE healing, absorbs, HoTs, and smart heals are the counter. All of the rotational tools have a few things in common – cooldowns, durations, efficiency. Dropping 2xHR in one area doesn’t help you obviously, nor does Riptiding the same person twice.

    Why does AoE need to be efficient? Well, I don’t see any other choice for it, really. For me, ‘low efficiency AoE’ in general is hard to imagine. When would you use it? To get the absolute maximum 6s throughput? Sure, but that gap is so narrow (either it’s less than 1 GHW gained or the penalty for ‘misusing’ HR becomes prohibitive) that as gear improves the unfortunate consequence might be that HR gets completely ignored. I’d love to be convinced otherwise, but I just don’t see Lifes’ situation of ‘cast HR or your raid dies’ holding up throughout an entire tier.

    My other point is that healing can be made more interesting not by revamping how mana changes throughout a tier (aka Spirit and trinkets) but via small scaling tweaks and more varied fight mechanics. We have an issue where AoE healing overwhelms ST healing, to the point where healers just drop the smart/AoE heals and absorbs and call it a day. If we lower SP ratios on these 3 categories of heals, they’ll still need to be used rotationally (because of efficiency), and at the end of an expansion their effects will not reduce the need for ST heals substantially. The next part is to introduce more mechanics that require ST healing – even now in this end of expac/BiS state intense triage healing strains healer mana pools, so the potential is there. To keep it balanced from start to finish of a tier, Blizzard might consider mechanics that allow the player to take the damage for a benefit.

    This is the best I could do for now, my brain is fried.

    • Dedralie says:

      “To keep it balanced from start to finish of a tier, Blizzard might consider mechanics that allow the player to take the damage for a benefit.” — You just want warlocks to get Vengeance right? XD

      Okay, but for serious – since this is a thoughtful comment that requires more than just me making jokes – the main thing I want to see about spells in WOD is for it NOT to be a better idea to use an AoE heal to deal with single-target damage. I’m going to get into this a lot in the 4th post in this series, so I don’t want to jump too far ahead of myself and lay out the entire 4th post in this reply!

      I don’t know that it’s necessarily true that Blizzard doesn’t like a triage/Whack-a-mole healing style. Their commentary when the Intellect/Spirit changes were first announced – linked in the post above, right after the Zon’ozz picture – seems to call for healers to do less overhealing, and you simply cannot have a “less overhealing” model in the same space as a “rotational AoE healing” model. I’m not sure if it is just a pervasive creep from the fact that, well, they had to come up with exciting talents and exciting new spells for MoP, and it’s difficult to invent a single-target heal that is significantly more interesting than what we already have, or if it was their conscious design choice, but in either case, I do not prefer it and am arguing for it to be toned down.

      There are a lot of solutions to the shallowness problem – most of them are “redesign the Shaman class please” – and what you suggest is fairly in line with what I want. But I don’t think that rotational AoE tools should be an element of the healing game. I would like to see AoE used when AoE healing is called for, and otherwise not. I’d like to see situations where 2-5 targets are injured be handled more by single-target healing, and situations where 6+ targets are injured starting to call for AoEs.

      Part of the reason I want this is that AoE healing just doesn’t feel skilful. I know like nothing about warlocks – so I can’t make a very good argument based on your class – but ask anyone if it feels skilful to be an Elemental Shaman and have Chain Lightning so strong that it’s all you cast in AoE situations. I’d hope they would say no. It feels powerful, yes, but not at all like there is any skill involved. (My support for my thesis: when I am Ele, I can match Moshne’s output on AoE fights, adjusted for gear differences. There is clearly no skill involved in Elemental AoE or else I would not be able to do that :P) Now imagine if Chain Lightning was also so strong that it was clearly better than other spells for single-target damage as well. That would be an incredibly shallow playstyle. That is what healing feels like to me, except in the place of “Chain Lightning” we have “all of our AoE toolkit”, so it is a little less shallow than my analogy, but not by much (and especially not for my class, where so little of my tools require any thought).

  7. Navimie says:

    I actually found your “pattern-matching and optimisation game” analogy with tetris amusingly astute, but I have felt that this last expansion has made healing feel like i can still fit crooked pieces into long straight holes in terms of spell choices withouth much consequence. And the LMG has meant very little to me. When it procs I don’t need it, in fact, I am so apathetic about it I don’t even use it to charge up my mushrooms because they seem to be charged all the time anyway – I am sure in a 25 man they charge up even faster. One could argue why I bother with it at all – but the truth is I paid so much for the damn thing, I’m too cheap to bother taking it out and putting something else instead.

    Great post Dayani, I can’t wait to read your next installment.

  8. Pingback: My Endless Journey: Proving Grounds | healiocentric

  9. Pingback: Warlords of Draenor Mini-Analyses: Healing Gameplay | healiocentric

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s