Life in Group 5 – A Resto Shaman Blog
A resto shaman perspective on raiding


August 4, 2010

Just in Time vs. Just in Case Healing

More articles by »
Written by: Vixsin
Tags: , , , ,

We’ve been hearing the fuss for months—healing is intended to be a “triage” affair come Cataclysm. Health pools will be up while healing numbers will increase incrementally, and having players at less than full health will become acceptable again. (Hey, we might actually be able to convince dps to use … um, what were they called …. popular back in the day … bandages!) In all, a healing mana recession is poised to be released on Group 5 residents, with serious ramifications on healing strategy and methodology. But as the excitement builds over this new improved healing model, which decries “wasteful” and “mindless” healing spam, I got to wondering, are overhealing and button-mashing really the enemies we think they are?


The Origins of Healing Waste

Oh the joys of healing in Vanilla … Although largely touted as the pinnacle of “intelligent” and “challenging” healing, the fact remains that fights in those days were quite a bit different than the encounters to be found in ICC and Ulduar. Raid teams were 20- or 40-person groups, comprised of anywhere from 10 – 16 healers. And fight mechanics assumed that at any given time, maybe 50% of your healing team would be focused on regen. (Hell, they probably also assumed that ~10% of your raid was afk!) Druids Healing Touched between Innervates, Shamans decried the lack of Chain Heal, Pallies cleansed and buffed, and Dwarf Priests, the only priests with Fear Ward, were the de facto healing standard.

There was no such thing as “replenishment”, nor was mp5/regen at such a level where you could assure your own longevity. So, frequent regen breaks, innervate staffs, JoW wanding, downranking, stopcasting, and the OOR 5-sec rule were all common work-arounds for the mana-strapped healer. Healing assignments were set in stone to assure that you didn’t have overlap, and tank deaths were progressive not instant. Healing was a game within a game, more akin to the micro-management found in RTS than the cause-and-effect exchange in an MMO. It was in this type of environment where healers fussed about mana like penny-pinchers, and determined overhealing to be the worst kind of evil imaginable—wasteful.


“TPS Reports” (not just a line from Office Space)

In thinking about the Vanilla healing model, with its narrow definition of excess, I couldn’t help but think of another model out there which views waste in a similar (although more broad) perspective—the Toyota Production System (TPS). (Some of you might remember me mentioning this system when I wrote about the 5 Whys of Wiping.) For those of you who aren’t familiar with TPS, here’s the crash course (bear with it, it’s only a paragraph, and it’ll have immediate relevance!) …

The second major principle in the Toyota Way, TPS seeks to redesign processes to eliminate overburden, inconsistency, and waste such that a smooth flow of work and production are achieved. According to TPS, “waste” can be classified as any one of the following:

  1. over-production – eg: making 5 widgets even though you need 3
  2. motion (of operator or machine) – eg: needing to walk the 2×4 from the table saw to the belt sander across the room, instead of having the two machines right next to each other
  3. waiting (of operator or machine) – eg: waiting for someone to finish assembling the widget that you’re going to paint
  4. transportation – eg: having parts sent to someone else who then brings them to you, instead of just having them shipped to you directly
  5. over-processing – eg: designing the widget so that it works counterclockwise and clockwise when the customer stated they only needed it to work counterclockwise
  6. (unused) inventory – eg: raw materials which are sitting and waiting to be used
  7. Repair and rework – eg: installing a widget incorrectly the first time and needing to reinstall it again later so that the machine functions properly

As you can see, some of the above “wastes” derive from a need to do something “just in case” while others address bad system design. Beyond understanding that the two are intimately intertwined (who hasn’t left their house early “just in case” they hit traffic? Is that you wasting time or bad traffic engineering at play?) there’s something else to consider–even in the most perfect system, waste will always exist. And so, the system designer or participant attempting to apply TPS theory to reality, eventually faces a question–what constitutes acceptable waste?


The Wasteful Debate

With that explanation of TPS, I hope you’re starting to get an inkling about what I’m getting at—the healing game within WoW has a lot in common with a production system in which the users are striving to eliminate multiple kinds of waste. (Yes, it’s a nerdy perspective, but hey, what did you expect from a nerd like me?!) Healing is, at its very core, a production model / delivery system, focused on meeting a precise and finite need at the moment it occurs. Anything that occurs outside of that very specific solution is seen as waste. And while many players would argue that the wastes found littering the healing landscape– overhealing, spamming, healing the wrong target, running unnecessarily about the room, excessive mp5, bringing too many healers or too many of the same type–are all user-based, I don’t think that’s entirely correct. In fact, I think the production system itself, which tasks players to deliver heals which fit the incoming damage, is more to blame for wastes than lazy play.

While I concede that many of the aforementioned elements are in fact user-controlled wastes (excessive movement, incorrect targeting, excess mp5, and bad healing team composition, to name a few), when it comes to overhealing and spamming, I have a hard time viewing them in the same light. Yes, bad and/or an inattentive healers fall prey to these wasteful habits quite quickly, but a correlation between “bad” healers and these practices does not necessarily mean that the practices themselves are “bad”. In an encounter where damage is random, unpredictable (to an extent), and demonstrates significant fluctuation of effect, spamming and overhealing are not wastes by themselves, but rather wasteful practices which were developed to address variation in system design. (Remember the example I gave about leaving early because you think you might hit traffic on the way to your destination? Same thing. You’re technically wasting time if you arrive early BUT it is the only way to mitigate the potential impact of heavy traffic, which represents a flawed transportation system.)

So, in an environment like ICC, where an offtank on Heroic LK can die in 1 second, (a death which cannot be mitigated via CD or a pre-cast heal because the kill is simply a matter of RNG alignment), a healer doesn’t have the luxury they were afforded in Vanilla WoW encounters. You simply cannot play a micro-management game because decision and reaction time have been reduced to a minimal window while variation and inconsistency have been exponentially increased. The only way to give your tanks the best possible uptime is to provide them with a continual stream of healing, regardless of the waste. Thus, overhealing and spamming are *necessary* to the encounter.


Applying “Just in Time” Healing

In TPS, the inherent goal of streamlining a process is to eliminate variation and inconsistency (known as “mura”) so that an even flow can be achieved. Good healing practice, in essence, is about achieving this same flow; delivering the right heals in the right quantity to the right person at the right time. Thus, becoming a good healer, one that can quickly master fights, is about moving from one type of healing perspective and practice to another:

  1. Just in Case Healing – letting a heal land without stopcasting “just in case” the tank takes spike damage or casting shields/hots on targets “just in case” they take splash damage.
  2. Just in Time Healing – pre-casting a heal so that it will land “just in time” to bring a tank back up from 20% HP or casting a shield/hots on a target “just in time” for them to be immediately effective.

But, what is implied in the two above philosophies is a variation in the knowledge the healer has of the encounter and its damage mechanics. The more a healer is able to predict encounter effects, and develop an appropriate solution, the more he can practice “Just in Time” Healing and the more healing waste he can eliminate from his delivery system. While some of this information needs to be incorporated into the encounter (because a fight with pure RNG would be absolute hell), it is really up to the player to bridge that knowledge gap by any means necessary.

So why does this matter now? Because with Cataclysm, Blizzard is touting a return to the Vanilla micro-management model for healer mana, and is thus promoting (again) that overhealing and spamming are wastes which good healers will need to manage and strive to eliminate. If fights maintain their level of complexity and immediacy of damage,  healers will have an even more limited number of options to address the problems in front of us. Much like someone using TPS to refine a process, and its inherent wastes, so too will we be tasked to become better at our healing delivery. And, while we won’t be able to address the encounter design mechanics (which, to date, have made overhealing and spamming necessary strategy), we can address the knowledge gap between “Just in Case” and “Just in Time” healing.

In the months before Cataclysm, we healers have an opportunity to hone our skills, to start developing practices which will let us achieve better delivery, and enable us to exert better control of the tools we do have at our disposal. Hopefully Blizzard will offer us some help along the way, by keeping our tanks from being 1-shot and not punishing us too much for the reactionary (and wasteful) healing that is sometimes necessary. But we have to meet them halfway, and that, friends, means a change for us all.


  1. […] I started pondering whether the larger mana pool wouldn’t actually be better in the announced Cataclysm healing. I guess we will see. Right now we can only hope that all the crit dependant mana returns and the […]

  2. It’s in post like these that a person’s background immediately shines through. Are you in the business of production management Vixsin or ever had a course on it? (if you didn’t then that makes this comment quite stupid, so fingers crossed).

    An interesting read and no doubt something that the entire ‘good healer’-community has been sensing/hoping for. I remember the days when healer channels were actually used to debate healing rotations, cool down management, phase transitions, etc instead of “you heal A, I heal B, hots/shields on all” tactics that seem to be the case nowadays.

    To an extent, the “just in time” mentality has never really been gone. Dragon’s Breath’s, like Sindragosa, Halion, Onyxia and the like, combined with a Hodirlike raidwide damage spike have served to keep healers on their toes. Sad thing is that now that gear has become so good, that most fights can get by with the “just in case” mentality.



  3. Suffice to say that some of the graduate topics I studied (and really enjoyed) make appearances in my posts. (heh) 😛

    In all, I’m approaching this “return to Vanilla” healing model with a good amount of skepticism, because although I appreciate a good challenge, there’s nothing more frustrating for me as a healer than having to ration my healing like it’s a precious resource. But I suppose some would argue, that’s the whole point.

  4. Another nice post Vixsin, you’ve definitely moved into my favorite blogs to read lately. The differences in delivery involved in “Just in Case” versus “Just in Time” healing can also be addressed by knowledge of the situation, an extension of the TPS model and “Industrial Management”. In fact I have been crafting a bog post on the Addon vs. Default UI debate based on the result of research from what grew out of this field. Now I’ve got somewhere to point to as an example of how the concepts pertain to healing in general.

  5. […] Just in time versus Just in case healing from Life in Group 5.  An excellent analysis of the current healing style and what the cataclysm healing style might be. […]

  6. I’m anticipating a rough transition here. Even among good healers, i just don’t see this going smoothly. We’ve been so used to a world where spamming is fine, and mana management is glorified cooldown management, that going to a “ah, he’s at 50%, no worries” model is kind of a scary thought. I’m really hoping to get back to the proactive healing model of old instead of the reactive model of late, but a large amount of encounter design would have to change.

    I suppose in a guild situation, I just see this as requiring your healing team to work more like a team to avoid the waste than a pack of wild dogs scrambling for heal snipes, but what I’m worried about is pugging. Back in vanilla, pug raids weren’t necessarily common. The kind of coordination that proactive timing and minimal waste requires also requires a higher level of skill and teamwork amongst healers. In current content, if I’m pugging and see a healing that’s struggling, it’s really not an issue to pick up some of the slack. When mana is a problem again, how much more difficult will this be?

    Don’t get me wrong, I’m excited, and I’m enjoying the triage required even for 5 mans in the beta, but i guess my concern, and my question to you, is where do you see waste management fitting into a group situation, particularly pugs?
    Aunna´s last post ..Mana Management – Tips to Avoid Going OOM

  7. […] Vixsin of Life in Group 5 gives us a final Restoration BiS! Shopping List and discusses Just in time vs just in case healing […]

  8. […] it yet? Ok, I’ll just tell you. Life in Group 5, a great healer blog, wrote a piece called Just in Time vs Just in Case Healing. Sumanhi stole this title and referenced life in Group 5, the very site she stole it from! […]

Leave a Reply

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

CommentLuv badge