I am going to attempt to explain the disconnect the community and the developers have over effective health.

When I first learned to tank, long before I came to Blizzard, I learned that effective health is a measurement of your stamina in relationship to your armor. This is a pretty easy number to generate. It’s reasonable to include say shield block and other simple forms of mitigation into the calculation.

However, you cannot directly translate effective health into best tank. Avoidance matters. If it didn’t, we would have no reason to nerf it in Icecrown. Good tanks don’t depend too much on avoidance, but great tanks understand its value.

Furthermore, your estimations of effective health become less and less accurate the more variables you try to factor in. Most saliently, you can’t easily account for cooldowns. You can’t compare a short duration that reduces damage by 80% to a long duration that reduces damage by 10%. Mathematically they might generate the same effective health number, but in reality they work pretty differently and each has their own benefits in certain situations, which vary depending on boss mechanics. (I’d generally take the first one though.)

We purposely made the cooldowns difficult to compare from class to class. You shouldn’t then be surprised when we take your effective health calculations based on direct comparisons of said cooldowns with a grain of salt.

It’s fine to compare health, armor, avoidance or cooldowns. I would not recommend putting too much faith in one ubernumber that you generate by combining all of them.

What I was trying to say above was that a half strength cooldown that is available twice as often is not the same. You can multiply the numbers to convert them to the same relative uptime but that doesn’t mean they are of the same utility.Would you rather have a cooldown that prevented 90% of incoming damage once a fight or a cooldown that prevented 5% of damage eighteen times a fight? You might be able to make arguments for both, probably depending on encounter specifics, but I find it hard to argue that the decision is irrelevant.

High avoidance is important because it makes the bosses hit hard. If avoidance was nigh irrelevant then bosses would not need to hit harder. They hit harder because you are avoiding so much damage. I’m not sure how many other ways to say it. I’m not arguing you should gem parry, but I find the arguments that avoidance is irrelevant and only health matters to be specious.

Q:

1. Can we expect more unavoidable, devastating melee-based attacks in ICC?
2. If so, what point do they serve from a design stand point?Are we finally going to get away from the 2 hits back to back will gib the tank situation we are in now?
3. How are you going to handle Chill of the Throne when the pre-Cata patch comes out and we lose another ~18% avoidance when Defense goes away as a stat?

A:

1. Probably. They serve as challenges that your group needs to overcome by making sure enough healers are focused on the big damage spike and cooldowns (the tank’s or external ones) are used appropriately. If you could avoid those attacks they would need to hit for even harder to compensate. If you could avoid those hits, then sometimes you would just let lucky and make it through the encounter unscathed and other times you’d get gibbed. Believe me; you want those to be unavoidable.
2. The idea behind the Chill is to lower boss damage per hit but keep damage per time the same overall. The reason I caveat that statement so much is I know that we’re going to see lots of tanks that die in Icecrown and then ask us to nerf the encounters or buff their tanks. The purpose of these changes is not to prevent tank deaths. You will die. Probably a lot. You are going into the Lich King’s home after all.
3. I dunno. I imagine we will just drop it. We’ll have a lot of fixin’ up to do before Cat is ready to go live.

Well, to be fair most theorycrafting tanks on Tankspot and other places who really understand the concept of EH won’t tell you it’s the only thing that matters, just that it is very important. The problem is that some players who perhaps don’t understand the theorycrafting as well try to take the notion to illogical extremes.

If we buffed DK parry to 99% and dropped their health 5K then it would be a huge EH nerf but they would probably be the best tanks by a wide margin. Now that’s a very contrived scenario but you can’t argue that just because tank avoidance happens to be close right now that avoidance is irrelevant as a stat when determining survivability.

(Source)