If you've tanked at all over the course of Wrath, you've probably become familiar with the phrase "effective health." It's a concept that's cropped up with increasing frequency on the tanking forums, and not necessarily in a good way. If you knew nothing of the idea beyond how players tend to use it, you'd be forgiven for thinking that "effective health" is the only metric by which all tanks are measured, and proof that Blizzard either can't (or won't) balance the game. There are very real differences between the tanking classes when it comes to average EH, and this has resulted in some angry discussion when the term is thrown around by players who either don't really understand what it means, or don't know that it was meant to be used in context. Consequently, "effective health" as used on the tanking forums has become an endlessly parroted phrase that's not only starting to lose all meaning, but is also guaranteed to derail a thread once it makes its inevitable appearance.
When I say that effective health needs to die, I don't mean that the concept itself is intrinsically wrong. It's not. But the twisted version of it so frequently used to bludgeon players over class differences is getting more ridiculous by the day, and it prevents or distorts more reasonable commentary on things that are much more likely to kill tanks on hard-mode content.
Why effective health is important
Effective health was a concept first discussed by Ciderhelm of TankSpot in a May 2007 article when pre-nerf Karazhan was progression content. It was something he'd toyed with since tanking the original 40-man Naxxramas, and more particularly Patchwerk, when it became obvious that the off-tanks had to reach a certain armor and stamina threshold to avoid being one-shot by Hateful Strike. Under those circumstances, he observed, it was pointless to encourage a tank to stack avoidance; when they inevitably failed to dodge or parry a Hateful Strike, they still had to survive eating the blow. A raid that wanted to get past Patchwerk had to be realistic about the amount of damage the offtanks were likely to absorb from a Hateful Strike and gear them accordingly.
The general principle is simple; tanks, regardless of how high their avoidance is, will eventually get hit. Ideally, you want that hit to be as small a percentage of the tank's life as possible, because healing a tank through a blow that hits for 15% of his/her life is much easier than healing through a blow that hits for 40%. Because most other contributions to tank survivability are procs (e.g. Ancestral Fortitude) or RNG (avoidance), the only way to ensure that an incoming blow will always hit for less is to gear for as much armor and stamina as possible. More armor reduces the damage of the blow; more stamina guarantees that what does land will be a smaller percentage of the tank's overall health. Good tanking is about avoiding damage, but it's also about being able to smooth incoming damage into a manageable rate for the heal team.
Isn't that basically the druid tanking model?
Yes -- to a certain point. The BC-era druid tank was essentially a waddling pile of effective health, because we were designed around the ability to mitigate (rather than avoid) crushing blows. With a 15% chance per hit to receive a 150% damage attack with no way to dodge it, druids stacked armor and stamina to turn monstrous hits into manageable ones. Warriors and paladins were hit less often and could avoid almost all crushing blows, but the hits they did take were proportionately larger than the ones that fell on a druid. Over the course of a given encounter, the druid would take more cumulative damage but less per hit; warriors and paladins would take less cumulative damage but more per hit.
Although it made us overpowered versus basic melee damage, the druid's armor and health advantages were not considered a significant balance problem. The need to absorb crushing blows, our vulnerability to magic damage, our lack of worthwhile cooldowns, and inconsistent itemization all made the druid a less attractive choice than we might otherwise have been. However, it was the inability to deal with mechanics like Pyroblast (Kael'thas), Shear (Illidan), Deaden (Reliquary of Souls), and Fear (Nightbane, Archimonde) that made these encounters virtually impossible for a druid to main-tank.
To put it very simply, the druid's effective health advantage didn't make much difference on encounters where tank death was overwhelmingly the result of mechanics that ignored armor and health. Thus, the ultimate lesson of the BC tank experience is that encounter design is capable of bypassing a particular tank's strengths, regardless of how massive their advantage is -- and the druid's armor and health, relative to a plate tank's, were considerably higher than they are today.
What happened when crushing blows were removed from the game?
This is an issue we've discussed previously, so I won't repeat it at length here, but Shifting Perspectives: Tanks, Wrath, and crushing blows illustrates the transition between Burning Crusade and early Wrath tanking. As a gloss on that March 2009 article, the introduction of the death knight tank forced Blizzard to eliminate crushing blows from boss' hit tables, as PvE and PvP balance problems would have resulted from a 2H-weapon tank being able to avoid or absorb them. However, death knights still became the progression tank of choice on hard modes, mostly due to the introduction of giant magic-based hits and cooldown-based tanking in general (e.g. Sartharion 3D, hard-mode Vezax), which was uniquely suited to the death knight's ability to chain strong defensive cooldowns with help from a 25-man heal team.
In another nod to the influence of encounter design, warriors and paladins -- by no means weak tanks, but still designed around avoiding the now-obsolete crushing blow -- were hit hardest by the introduction of "crushing blows" that ignored their avoidance, +block, and armor. Druids, by contrast, were an acceptable second choice if you didn't have a death knight; having been designed around swallowing melee crushing blows gave us the option of stacking stamina in order to absorb the magic "crushing blow." While the druid was significantly harder to heal than the death knight on content that made use of this damage, a large health buffer provided a welcome margin for healer error.
Now and then
Roughly 8 months after writing that article, several things have changed:
- Death knights have been nerfed, mostly in the form of their ability to chain percentage-based damage reduction cooldowns. Right now they are popularly considered the weakest tank for progression raiding, although (as always) much of the community discussion concerning this is completely overblown.
- Paladins have been buffed, most notably in the form of the new Ardent Defender. In an interesting reversal of their usual fortunes, they are popularly considered the best tank for progression raiding.
- Warriors are roughly the same, although Glyph of Shield Wall (introduced in patch 3.1) has given them the opportunity to tailor Shield Wall to specific encounters, particularly in conjunction with Improved Disciplines.
- Druids have been nerfed, but are still popularly considered the second-best or best tank for progression raiding due to an effective health advantage (should they choose to stack stamina over avoidance, and most do). In another curious note, especially given the tanking community's strong feelings on the matter, druids have also been confirmed to be the least popular raid tank.
Every week, Shifting Perspectives treks across Azeroth in pursuit of truth, beauty, and insight concerning the druid class. Sometimes it finds the latter, or something good enough for government work. Whether you're a Bear, Cat, Moonkin, Tree, or -- for some unaccountable reason -- stuck in caster form, we've got the skinny on druid changes in patch 3.3, a close look at the disappearance of the bear tank, and thoughts on why you should be playing the class (or why not).