Tobold's Blog
Friday, January 30, 2009
 
The importance of healing meters

A week ago greedy goblin Gevlon posted why he thought that healing meters are important to measure a healer's performance, to violent disagreement of the healers in his comment section. Of course he was talking about an ideal healing meter, which would take all of a healers actions into account, including things like decursing or shielding. Unfortunately such a healing meter doesn't exist, and there are good arguments why it isn't even technically possible to build one. The real existing healing meters can be gamed, which is the subject of a famous post on how to top the healing meter; that post is extremely funny for a real healer, because it shows how by maximizing your position on the healing meter, you are actually minimizing your real worth as a healer. The worse you heal, the higher you score.

Nevertheless healing meters are a problem, because they exist. People look at them, and judge you according to them, often not knowing how little they actually tell you. Thus when I let Recount run for a whole 25-man raid with 7 healers (2 druids, 2 paladins, and 3 priests) and the three priests ended up on places 5, 6, and 7, I was't happy. How can a class specialized in healing end up bottom of a healing meter, behind the druids and paladins?

The answer is easy: The positions on the healing meter not only depend on the performance of the healers, but also on the performance of all the other classes, and the difficulty of the encounters. If your raid is going somewhere really hard, is somewhat short on healers, and the tanks and dps are undergeared or otherwise underperforming, all the healers can heal to the maximum of their capacity. In that comparison holy priests score quite nicely on a healing meter, and even discipline priests do well, in spite of the healing meter not counting their shields. But if you are in a farming raid, with more than enough healers, and the tanks and dps just rolling over the enemies, the nature of healing changes. There just isn't enough damage to heal, and healing becomes a competition of who heals the fastest. In that situation shorter casting times and heals over times easily outperform slower, more mana-efficient heals. Thus druids and paladins top the healing meter.

The danger is of course that if a raid leader realizes that he has too much healing, he will be tempted to invite one less healer and one more dps. And if he isn't very well versed in healing mechanics, he might well cut the guy at the bottom of the healing meter. Which explains why healing meters are taken so seriously even by people who know that those meters can't measure real performance. Even a damage meter isn't perfect, but at least there is no limit to how much damage you can deal; healing is always limited to how hurt the raid is, and "avoiding overhealing" is often neither possible nor even desirable. So when Ghostcrawler announces "We have some exciting changes planned for priests", I sure hope that these changes make my priest score higher on the healing meters. Not because my *real* performance can be measured that way, but because other people will judge me by that. Getting kicked out by Mr. Stupid from a pickup raid for underperforming hurts, even if it was just him being wrong.
Comments:
I'm scared how my reaction to my first real contact with healing meters as holy priest on raids will be.
I just leveled my secret priest alt to 80 and rejoined with my guild that I raided with in TBC as a mage. As mage I could easily judge my performance based on the numbers, the boss, my rotation, spec, etc.
If the healing meter will tempt me to score higher numbers I'm doomed. But it also depends on how the meter is used and perceived by the other healers I guess.
 
As a raid officer and raid healer, I only look at the meters to determine one of a few things:
1.) is one person doing all the dispelling/decursing/disease/poison removal?
2.) are people healing the right targets? (single target healers on assigned roles, raid healers healing the raid, etc)
3.) if someone is mana starved are they casting to expensive of a spell all the time, or are they just spam casting a spell without paying attention to the actual damage that needs to be healed?

-Ghaar
 
That link was pretty funny to read (Even for someone who just did 5-man instances).

The entire post does get me wondering, though, how there would be so many people who don't understand healing enough to get even some of the problems with a heal meter. Even just thinking about healing mechanics for a bit, and knowing how overheals and cast times work, it's pretty simple for me to figure out possible issues with healing meters (and how they could be gamed), and playing some 5-mans and guild wars (as a healer), quickly showed enough of how healing tend so to work to clue me in on some basic healing meter issues.

Are there really so many people leading raids who have never healed, looked at the healing spells and thought in their head about healing mechanics, or at least understood that healing meters, like anything really, are imperfect in some way and looked around somewhat to find the limitations, to better understand how they work and use them?
 
@ Dillon -

There are only so many people who read the internet. There are plenty of good RLs in Wow, but there are plenty of bad ones too. Besides, as Tobold notes, when a RL has to pick a healer to be kicked and replaced by a DPS, there isn't a lot of information to go on. People point to healing meters as "objective proof" to justify their decision which, unfortunately, is probably going to be pretty random otherwise.
 
Gevlon just put up ANOTHER post on the topic of healing meters:

http://greedygoblin.blogspot.com/2009/01/damage-more-than-tank.html

In this case, the 2nd healer is a Discipline priest and is "outhealed" by the main tank, a prot specced Paladin. This happens to me routinely as well and I'm better geared/specced than the Disc priest shown in Gevlon's post.

In his post from a week ago, Gevlon says something to the effect that raid leaders should "book" the healing done from shields and PoM and credit it to the priest. A noble sentiment, but 1) how many raid leaders are going to actually make a habit of doing this, and 2) it's impossible to do this in a raid with more than 1 priest because shields/PoM are not linked to their caster in the combat logs.

Until this happens (in all likelihood, never), Discipline priests will always be at the bottom of the healing meters and we will be only be grouping with people/guilds who "pick the player, not the class/spec". Personally I don't have a problem with this, the more min-maxers I run into in WoW who try to tell me how to play the game, the more I get ready to go on hiatus from WoW for a while again. The ignorance is only tolerable for so long...
 
Tobold, even though recount excels at a lot of things, one of the things that it fails at hugely in this iteration of it is measuring *EFFECTIVE HEALING*, which is the amount of healing done, less overhealing. You currently have to manually do the math on that in recount, and it is probably the worst meter available to base healing on. It *IS* fantastic to so many other ways, that I am really sad they haven't added an effective healing option into the meter as they had at the end of TBC. It's really unfortunate, since it is currently the most common meter in use.

I am currently opting to use a low memory meter called Violation, that to my knowledge is the only meter currently calculating effective healing.

I would take what you saw on recount with a grain of salt. Did you look at the amount of the paladin's overheal in your raid? I'm guessing it was probably between 40 and 60 percent, where as your three priests were probably around 20-30 percent overheal. For recount to be a viable healing meter you really have to look at what the overheal is. I would guess that after you factored in your overheal, it probably went something like this: Druid 1, Druid 2/Priest1, Priest1/Druid 2, Priest 2, Paladin 1, Priest 3, Paladin 2.

As long as your raid leaders and your guild leadership know the pitfalls of a healing meter, and how to properly evaluate from them, I think they are fine to use. But your leadership also needs to know the strengths and weaknesses of the classes healing as well. Paladins will never see the top of the meter if they are tank healing and you are in a fight with lots of AoE damage for the priests/druids/shaman to pick up. However, take a fight like Instructor Razuvious where there is massive, consistent, single target damage and I would almost guarantee that your paladins are probably rocking the top of your effective heal meters. We have one guy that will outheal every other healer in that fight by a good 15% every week.

With all that being said...the old addage of "as long as people aren't dying, the healers are doing their job" should be viewed. If everyone is living, I really don't care which of my healers is number 1 and which is number 7. =) It does stink that random pug 1 probably doesn't get the intricacies though =(
 
Oh, and I forgot. I do beleive that recount IS crediting PoMs to the priest that cast it and the bloom on lifebloom to the druid that cast it now as well (at least it was when I switched from recount to violation). I'm not positive what Violation picks up on that front, as it doesn't provide the same slew of information that recount does (being a very basic, low memory option).
 
All of those healing effects, PoM, Lifebloom, earth shield, and similar effects had their threat generation amount reduced very very low, 'and' had all the healing done by them credited to the healer who cast them. This change was implemented in 3.0 (pre-Wrath release).

The wow combat log is how all those addons (recount, volition) get their information.

The paladin outhealing the priest is because judgement of light is kinda broken I believe. helps to create massive threat gen 'and' keeps the paladin healed.
 
Post a Comment

<< Home
Newer›  ‹Older

  Powered by Blogger   Free Page Rank Tool