Healing Theory, Part 3: Mana and other Resources

In the first two posts in the sequence, we started building a foundation of how to think about the task of healing, and conducted a basic survey of how all the various stats might impact your performance. As promised in the last post, this is going to be a whole piece focusing on how to make decisions concerning mana (and secondary resources if your class has them). It will also finally bring us back around to the issue that started me down this whole train of thought in Mists.

Beyond Spirituality

Everyone accepts what the purposes of Int, mastery, crit, and haste are: to do more healing. You can do more healing in a given amount of time, you can do more healing with a given amount of mana (haste doesn’t actually do this, but that’s not for this post), and ultimately you can keep more people alive over the course of an entire encounter. The first premise of this article is that Spirit is no different. If you’re using a stat, it must be for the purpose of doing more healing, and its value is determined by how much more you can do (usual disclaimer applies throughout–”more healing” doesn’t necessarily mean more meter healing, it means fulfulling your healing tasks more consistently). In order to be worth using, Spirit has to pull its weight by allowing you do more than you could do with an equal amount of crit or mastery. I want to stress how different this is from viewing it as an independent requirement of some kind, a sort of “you must be this tall to ride” minimum to survive any encounter, before you can worry about other stats. It’s a stat like any other, and if it doesn’t pay its dues in terms of added performance, you’re free to replace it with a stat that does.

So what does Spirit do for you? It lets you use your non-cooldown heals more frequently. I’ll only briefly recapitulate the basic dichotomy between cooldown and non-cooldown heals here; it’s appeared in every one of my MoP healing posts thus far. Remember from the previous post that well over half, possibly as much as 3/4 depending on class, of your total available mana is from sources other than Spirit. Even if you had 0 Spirit, you’d be just fine casting your core short-cooldown heals as much as you wanted (Wild Growth, Penance, Holy Shock, Renewing Mist, Riptide, etc.). These heals are cheap and powerful, and form a sort of healing “baseline” that’s mostly unchanged by added mana reserves beyond what you start with. The most important point is that if you find you’re coming up short to cast these at the end of a fight, it is not because of insufficient Spirit. You budgeted your mana poorly and spent too much on less-efficient no-cooldown heals earlier in the fight.

This brings us to the first fundamental result: the goals of mana management.
1) Solve problems with efficient cooldown-bearing heals whenever possible. This very often means knowing when you can wait a few seconds to whack a particular mole (from the ABC of healing in the first post: Anticipate Damage). Without that anticipation, you have no sense of whether a particular person is at immediate risk of death and can’t make efficient decisions. When you spend mana on a Flash Heal when you could have Penanced 4 seconds later instead (or simply let the target top off from the background noise of smart heals in any raid), what you’re doing is overhealing, even though no meter will say so.
2) Whatever’s left after (1) is your allocation for spammable heals. Only experience can teach you to not bust that budget. Since it would take some clairvoyance to know how much you’re actually going to spend on core heals for the rest of the fight, this is an iterative process where you learn to adjust your spellcasting habits. But if you don’t consciously make this effort, it will never happen, and you’ll be one of the many healers who burns all that mana without even realizing it (partially because this isn’t picked up by any common metric like overhealing percentage). Think about this next time you’re in a conversation with someone who’s running well over 10,000 Spirit and is convinced they can’t cast their heals enough.

The way Spirit, or other talents or bonuses that produce mana, plays into all this is clear–it increases the excess mana you have to spend on the extra heals beyond your usual rotational casts. Without, at the moment, getting into an appraisal of the value of those added heals for each class, we have the conceptual framework. Int, crit, and mastery (in general) make all your heals stronger, while Spirit allows you to add casts of certain types of heals. The most important factor affecting its value is the efficiency of those filler heals. If it’s high, adding points of Spirit will add casts that might output more total healing than proportionally increasing all your heals with another stat (albeit with a loss of burst healing ability, which is significant). But if the filler heal is low HPM, then Spirit will be a weak stat because it doesn’t buy you as much healing.

Note that we’ve reached a completely opposite result from people with a “you must have this much Spirit to ride” concept of things. In their view, having more efficient heals (say, due to a buff like Rejuvenation got in 5.2, or to a set bonus) reduces your Spirit “requirement” because you need less mana to cast the same heals–you hear people say this all the time. But it’s a logical trap; we’ve just shown that a change improving the HPM of a spell like Rejuvenation increases the value of Spirit as a stat, so the intuition to wear less of it is wrong. The conclusion that Spirit became a stronger stat for Druids in 5.2, and that we should therefore value it higher in gearing choices, is a good practical application of this post so far.

Better Living Through Chemistry

Now an interlude about classes with secondary resources (Paladins and Monks). It adds a twist to the basic analysis above, but not any drastic change. The key is that since every Chi or Holy Power spender was paid for by some Chi or Holy Power generators, and you can lump those together into a sort of defined molecule of healing that has a certain mana cost, more or less like any other heal. For purposes of the above discussion, an Eternal Flame funded by Holy Shocks is a core cooldown heal, but an Eternal Flame funded by Flashes of Light on the tank is an expensive filler heal. The rest follows exactly as described above.

This is why both secondary resource classes are set up the same way. Some cheap, inexpensive generators that are rate-limited by cooldown or otherwise: Holy Shock, Renewing Mist*, Expel Harm, Soothing Mist (Renewing is actually not cheap, but unlike the others it does a large amount of healing on its own, so it plays a dual role as a conventional core heal and an on-cooldown generator). These provide baseline resource intake that is independent of mana supply. If you want to generate more than what you’re given by those, you pay a lot to use a non-cooldown generator (tank Flash, Surging Mist, or Jab), the equivalent of another class’s spammable heals. If this paragraph makes sense to you, you should understand why Holy Power generation from Holy Light, or Chi generation from the cheap 5.0 Jab, were both design errors that had to be fixed.

So it’s a little confusing at first since the same heal (usually Eternal Flame or Uplift) can be either in a cooldown or non-cooldown column. It’s clearer when you think of the whole molecules of spenders and generators together, each having a defined output and efficiency. The force of the chemistry analogy is that each molecule should have 0 net Chi or Holy Power use. Shock-Shock-Shock-EF and ReM-EH-Uplift can be examined through conventional HPM analysis just like any normal single heal.

Where it gets a little more confusing is that in real healing, you don’t just spam out all parts of a molecule in a row. A mixed chain of spenders and generators is a sort of soup from which you have to tease out what’s really happening. If you’re using Surging Mist to enable more frequent Uplifts (anticipating that Surging will replace Jab for this in 5.2), and you cast ReM-Surging-Uplift-EH-Soothing(2 Chi)-ReM-Uplift-Surging-TFT-Uplift, what matters is that mixed in all that you have two Surgings and the one Uplift they pay for. The rest of it is normal use of your cheap Chi generators and the spenders they support. So the analysis should be that you added one heavy mana expenditure (a Surging-Surging-Uplift group) to your baseline healing over that period.

Next Steps

Perhaps by now you’ve noticed that I’ve gone another whole post with no (explicit) math in it. But that is largely the point–this is all the sort of discussion that should be had before putting detailed numbers to paper. Both because understanding these strategic points can already help your healing without the heavy investment in complicated math work, and also because we shouldn’t be doing any computation without a detailed understanding of what we’re computing and why. I think the latter has been a problem with a lot of healing theorycraft over the years. My own is no exception here–much of these posts are new ideas I’ve had since MoP arrived, and I’d been analyzing healing for years before.

I’m expecting that the next post in this sequence will be a numerical comparison of various healing spells, informed by all of the discussion so far. This isn’t cast in stone since I’ll still have to do a bunch of research, but the most prominent things I’m thinking of are:

  • What is the raw HPM of each class’s best no-cooldown heal? (Rejuvenation, PoH, Surging-Surging-Uplift, etc.). Keep in mind other factors–some are HoTs, some are smart. What if we restrict to a single target?
  • What’s the total mana output on each class’s core heals if used on cooldown? (i.e. how does the total average MP5 of WG/SM/LB on cooldown compare to Penance/Solace/PoM/RapturePWS on cooldown?)

I hope this has been helpful for you so far, and I’m looking forward to seeing where it goes.

10 thoughts on “Healing Theory, Part 3: Mana and other Resources

  1. Hey Hamlet, I’m from Ask Mr. Robot. I love these healing series posts and I’m always looking for well-articulated resources to point our users to. Would you mind if I ran a little promo spot on our site that points people to this series of posts?

  2. Pingback: Why More Spirit is not the Answer to your Healing Problems | It's Dangerous to Go Alone

  3. I’m interested in your methodology here. However, I think that some healing classes don’t operate rotationally like druids do, in particular Holy priests. Is your analysis still applicable for classes that are less rotational? Do you think you could go over the premises you’re operating on? They may not be true for all specs.

    • Holy priests are a good example, although it’s tricky into get into in detail because they’re likely to be changed significantly before becoming very popular again. Note how Disc priests were overpowered before 5.2, and the changes brought them much more into the usual cooldown+filler territory (Penance and the new Solace became very powerful spells, and the spammable PoH was drastically weakened). I’d think the first spell reworked on the Holy side is likely Holy Word: Sanctuary, which will almost certainly be some effective AoE spell with a cooldown. Combined with Circle at 8 or 10 seconds, and PoM at 10, I’d expect these concepts to fall into place in a similar way to other specs.

      The post I still want to do next (although it requires a good bit of research) is to mathematically compare “filler” spells of each class and post specific comments.

  4. I just found this blog today… Super stuff. I just wanted to add my two cents to the Spirit/healing/thing. I think the statement that the only healing meter that matters is if everyone (or the majority of the group) is still alive is 100% accurate. I think that is a spot on assessment. In my opinion.

    My question is this… how do you account for variables that are involved in healing? In my mind healing is subjective because it’s not just your errors that are going to hit your mana pool, but also the mistakes of others. I think thats the muddy part of this whole thing. Healers stating that they need to add more spirit could have nothing to do with their ineffciency, but rather the group as a whole. I think that is especially true when it comes to new content.

    but, then again… I could be completely wrong (probably am).

  5. Pingback: Healing Theory, Part 4: Comparison of No-Cooldown Heals | It's Dangerous to Go Alone

  6. Pingback: [Resto] Mists of Pandaria 5.3 - Page 55 - Elitist Jerks

  7. Pingback: [Resto] Mists of Pandaria 5.4 - Page 72 - Elitist Jerks

  8. Pingback: Healing Theory, Part 6: The Mana Economy | It's Dangerous to Go Alone

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current day month ye@r *