There is a common opinion that healing meters are worthless. The first question that arises here is what the hell these meters do on this blog? This is simple to answer: because it's a perfect example for accounting.In simple situations, like the gold/hour ratio of daily quests or herbalism people see the point of accounting and do it. In more complicated situations where several factors are in effect, people tend to come up with the idea that it's "impossible" to account for, so it does not worth trying. Typical example is the crafting tradeskill, where not only gold from the sold products, but time saved by not seeking crafters and "good reputation" within the guild for crafting needed items also count. Yet these factors can and should be counted somehow and accounted. Not doing it inevitably lead to underperformance. Remember, people naturally overestimate their success, so they won't notice their underperformance without numerical proof.
Everyone accepts DPS meters as the measurement of worth of a DPS player. The warlock damaging more than the mage is more skilled or geared than him, period. The reason is simple, every damage you cause to the boss is counted.
"Healers have the problem of overhealing. If you heal someone a milisecond after the other heal, your job is wasted. Still your action is worthy since you would heal your target if the other healer couldn't. Such overhealing is not counted in the meters therefore they cannot measure it." - this is the common reasoning with "healing meters are bad". However this reasoning is completely wrong. It is just valid for "being a healer is harder than being DPS".
The healer is there to heal taken damage. The healer who heal no damage is obviously worthless. The healer who heal all damage alone is a very good healer. Since there is a continous line between 0% and 100% it's obvious to assume that the healer who healed 38% is better than the 28% healer. "Better" can mean skill, gear or spec. Of course it's easier to claim "meter suxx", than finding out how can you be a better healer, just like it's easier to claim that "accounting is boring and does not work" than finding the ways to make more money.
Stop whining, start counting:
- If the accounting is imperfect, fix the problems. For example PW:S does not show up in the "healing done" section. Find them in the "buffs" section and multiply them with their effectivity and add them to the healing done. Your spec has a buff that no one else could give like Improved divine spirit or tree form: calculate it's effect (as extra mana regen and damage/healing done by others) and add it to healing done. If you spend 20% of the bossfight on decursing GCD, divide your healing done by 0.8.
- Why am I here? To heal! So the explanations for "why did my work went to overheal?" or "why can't I do my job?" bothers nobody but you. Your job is to find a way to do your job. No one cares how, just do it. If you don't you slack. If you can't you are dumb.
- Is my work needed at all or does the raid has too many healers (the enchanting market too many competitors)? If yes you must make a choice: fight for your spot, being better than the competitors or change your market by respecing shadow or finding other raid group.
- Is the quantity of my work is enough? Is my heal+overheal is more than the competitor's heal+overheal? (Or do I post more auctions than the competitor?) If not, start working harder. Get better gear, stop watching TV while raiding, read elitist jerks.
- OK, I do enough work, it's just ineffective. Welcome to the real world Neo! Competitors are part of the world, they want your spot, they want your heals. You must find a way to make your heals effective. This is your product, this is what the market wants from you. Find a way to do it, or find a way out! No one needs a loser! (though some tolerates them)
- "OMFG my HoT-s are overwritten by pladanins falsh of light, druids r ruind": well, HoT-s are not heals my little friend, their purpose is not to heal damaged targets but to create a buffer on undamaged ones who will be damaged. That paladin did the right thing healing the damaged player instead of risking that he dies because he get another damage before your HoT ticks down. Post-damage HoT-s only work if the direct healers are overburdened or slacking. Your job is not to heal the damaged ones, your job is to find out who will be damaged. Tip: the main tank, the warlock on low mana and the DPS warrior in the face of the cleaving boss are sure winners.
- "Fkn priests eat nerf, now you can't always CoH the people instantly before my heal could hit PalasFTW!" If you wasted cast time on someone who was on 90%, you deserved your fate. Direct heals are here to heal big damage. If someone is on 50%, he won't be topped by a CoH and need your direct heal much more than others.
- "Damn X had less lag, his Gheal hit before mine on the main tank, than on the offtank too. Damn the luck! ((C): Telestra) No, damn your coordination skills! Maybe you could have talked to him before the fight that "I heal OT, you heal MT OK?" It's called distributing the market and exactly serve the purpose of avoiding such collisions. There are also addons that show incoming heals, like Grid. If you see incoming heal on a target that is big enough, cancel the cast and heal someone else.
Edit: lot of people mentioned something that was so obvious for me that I forgot to mention. They say "tank healers could heal more by healing the raid but it would wipe the raid". And he should! By sticking to his tank he save the current fight at the cost of the long term improvement of the group.
If he is capable of doing efficient raid heal, that means that the raid healers slack or incompetent. Next to a good shaman I cannot increase my healing done by raid healing because he heals faster than my HoTs would tick. Actually I decrease my healing since I waste the GCD, risking that the tank fall low, allowing him to increase his healing done by healing my tank.
If my HoTs can tick that means he did not do his job. So if I can increase my healing by raid healing, I should, even if it kills the tank, because it forces the RL to do something with the shaman. Either teach him to be useful or replace him with a good one.
My class, build and gear set me to a niche. I'll never be as good raid healer as a good shaman. I'll never be as good single target healer as a good priest or pally. And they will never be as good on a 2-3 tank situation as a good tree. Invading their niche is an uphill fight unless they are M&S. I do a great service to the group by invading it. Practically I'm saying: "watch the noob shaman, even a poor tree could outheal him on raid".
24 comments:
You completely miss the point of the "healing meters are worthless" argument. A well organized raid has healing assignments, where it is the job of the healer to keep his target alive. So how much he can heal depends very much on how much his targets are damaged.
For example take the guy who is charged with general raid healing, for example at Sartharion. If the other raid members perfectly avoid the flame wall, he has very little to do, except for adding a hot on the tanks. If the other raid members get hit constantly by fire, he'll be on top of the healing meter at the end of the fight.
Number of points healed is meaningless. Everyone alive at the end of the fight is the only goal.
I have to disagree strongly here. I think you're right in the idea that people need to learn and improve their play instead of just saying "meters are worthless".
The problem is that a pure, simple healing meter that only measures healing done and overhealing IS worthless because it does not tell you the most important thing: who is doing their job correctly.
Healer A who heals 38% is NOT better than Healer B who heals 28%, because you don't know that their job is exactly the same.
When we were doing Vashj, there was a retarded paladin healer who was supposed to heal elemental dps (me). Instead of cleansing the poison from elementals, he spammed flash of light. When he got bored, he ran off to heal random people who got hit by chain lightning instead of staying close to me and other elemental dps, so we sometimes just died from poison because he was too far away. If you look at healing meters, he is on top of other elemental dps healers but he is still a retarded player who caused quite a few wipes.
Position on healing meter is almost completely worthless, you need a more detailed tool like wws or recount that breaks down what each player is doing. Unless 2 people are playing the exact same class and healing the exact same person, it is meaningless to compare healing done.
@tobold: "Boss is down at the end of the fight is the only goal" by rouge_below_the_tank.
The question is: is everyone are alive because of your effort or because someone's else's effort and you just tagged along.
@pockie: I wrote that healing done should be modified by decurses. The paladin was dumb because he did not made this adjustment, believed that decurse is worthless and went to do something else. BAD booking is useless.
You're still missing the point though Gevlon. Let's look at a example from the rare bit of 'hard' content in this game: Sarth 3D.
On Sarth 3D, the Main tank will, during the fights most dangerous segment, eat breath attacks that can hit for upwards of 40,000 damage. Yes, this is enough to ONE-SHOT most tanks with the -25% max HP debuff(Hell, even without it in many cases) without the use of significant defensive abilities and the rotation of 'tank saving' cooldowns. Immediately following this breath, he'll often swing with his ordinary melee attack for another 5-10k. An MT healer in Sarth 3D needs to be doing 2 things. 1) Mitigating tank damage (Sacred/PW:S) - you account for how this is to be measured in healing meters, but in this case, it's not enough to mitigate X Raw amount of damage and have it be more than the other guy - a PW:S that absorbs a melee autoattack from sarth is worthless. He hits like a wimp and such damage is easily healed. It needs to absorb THE RIGHT damage (the front end of that 40K breath), not THE MOST damage (think of it as working in lower volume in a premium market if you will.
2) Keeping that MT topped because a breath could occur at anytime, which often means sitting around doing nothing but avoiding environmental damage, while waiting for the MT to take damage - wasting a GCD on trying to heal someone else could very easily allow the MT to die.
Meanwhile, as this is going on, the ENTIRE RAID is taking 3-5k shadow damage hits every 2-3 seconds or so, due to a debuff called Twilight Torment. That means that your Resto Shamans and Resto Druids, who are on raid healing, are going to be pushing through an enormous amount of throughput.
Are your MT healers worse because they are doing less healing than your raid healers?
No, they are operating in a different market as it were, and healing meters are incapable of reflecting that.
The only true metric of healer effectiveness is: Did their assignment die. Now certainly, when two healers share an assignment, you can use meters to compare them, but to simply say "Paladin X did the most healing in this raid, he's our best healer!" Even if you adjust for shields and decursing, is going to miss the boat on understanding your healers relative worth.
Healing is an art of Reaction. When there is nothing to react to, it is impossible to objectively measure performance.
To put it in other terms:
An ambulance company that has an average response time of 5 minutes is not inherently worse than an ambulance company with an average response time of 3 minutes, if the average distance of Ambulance Company A's calls is twice the distance of the average distance of ambulance company B's calls.
Ambulance Company A might be able to lower their response time if they stopped accepting calls from further out, but then they wouldn't be doing their job.
Healing meters often do not depict some essential aspects of the encounter. Particularly time. When you do a heal X is often more important than how much you heal. However, overheal is not a clear depiction of when simply because there are variables which can not be easily modelled. Paladins are an excellent example. Take judgement of light. In cases where the damage is minimal most of the procs result in overheal. Should the paladin not use it? Glyph of Holy Light can also result in overheals.
What it comes down to is how much of that overheal is attributed to spent mana. Take healer (1) that spent F(x) mana and has X overheal and healer (1) that spent 1/2 F(x) mana and got 2X overheal (often the result of passive effects as the one described above); it does not mean necessarily that the healer that overhealed by 2X is worse than the other.
Couple that with assignments. If I am to heal the MT and I stay focused on that then chances are I might have less heal than others that heal multiple targets.
Healing is about reaction, anticipation, exploiting ability mechanics, encounter mechanics assignments and the ever important healer-target vector.
Still, you make a valid point that healers ought to try to improve themselves. Trying to reduce your overheal from wasted heals is a different thing from hiding behind cases like the aforementioned. There are encounters that Class A, using Spell A can top the meters and others where they can waste their mana for nothing. Rectifying this comes with experience and effort and adaptability. Particularly experience may lead you in many cases to allow overhealing to compensate for various issues that might crop during an encounter.
If healing could be easily and accurately measured and modelled we would not have this `debate' :-p
Even if you modify by decurse time, that paladin will still be on top of meters. Because depending on random targetting of Vashj's chain lightning and poison from tainted elementals, your assigned elemental dps can take 0 damage. But if you move away because you are bored, like that paladin, then suddenly dps can eat a "combo" of chain lightning plus poison dot that kills the player before you can get back in range. There is no way you can "book" that correctly because the correct action is to keep in cleanse/healing range of your assigned targets and do absolutely nothing unless its required. A good healer in this situation will always lose on the meters to Mr Paladumb no matter how you adjust it.
Another example is Tidewalker in SSC (sorry I'm using BC examples because I don't raid much now). After he does earthquake AoE and summons murloc adds, a warlock will spam lifetap down until very low and a pally will spam holy light to heal crazy amounts and get healing aggro, so the murlocs all go to the pally and can be AoE to death. Other healers can spam aoe heals madly on the raid to top healing meter, but then they will pull healing aggro and the raid will wipe.
Healing meter is useful in flagging out potential problems i.e. 1 healer is very very very low compared to the rest, meaning you need to take a good detailed look at what he is doing. But whether you are 1,2,3 or 4 doesn't mean anything.
Yeh to be honest you are totally wrong, healing meters ARE totally useless.
people assigned to raid healing will always be on top of them, that doesnt make raid healers better than tank healers.
In a raid you will be assigned to do a certain job and asked to keep a certain person alive, if you do that job then that is the only meter any healer needs.
Healing Meters ARE worthless. FACT.
P.S. I play a shadow priest so I'm not a annoyed holy pally or anything.
Gevlon, you are contradicting yourself when you say people are wrong to say healing meter sucks.
If I have (not that I do, mind you, I'm a mage:) to manually multiply my buffs by their effectiveness and add to healing done (which misses the point of whether or not they were needed) and other such things it *does* mean that healing meter sucks. It would be even better to say that the healing *metric* sucks, as it is inaccurate and doesn't reflect my individual contribution to keeping the raid or particular targets alive. What it does *not* mean is that healing *metering* sucks. Healing metering is perhaps good, but if the metric and tools make me do it by hand, on paper, they are worthless pieces of poo.
Also, even more to the point. You completely miss the point of a team. I know you will hate this :)
The whole point of a team is that they have a common objective and are *not* in competition.
Let me give an example: if I am a healer attempting to maximise my healing done and standing among my guild peers by riding the healing meter I will always spam CoH, even if it's not the best spell to keep people alive, because it generates more healing done. If some people die because I didn't -say- cast a greater heal on one of them, then it's not my business as nobody else can heal them after death and I've done my better healing even if the raid wipes.
The case of dps is the same. Have you noticed how many people still hit the boss instead of wiping when they are just 3 or 4 left and there's no chance to down the boss. They are wasting time (they should wipe and try again with enough people alive to have a chance at the boss) to squeeze in some more sweet dmg.
The fact is that the extra mile some people go as a result of internal competition is not worth losing the many other miles people stop going for the very same reason.
If you feel that your guildies must be tested competitively to check for slackers/dumb ones, do it with a dummy target. But not in a raid environment. There you have to work as one, not as one against the other.
@those who believe healing meters are useless:
You surely believe that there are good and bad healers. It's also obvious that the bad ones must be identified for improvement or removal.
Do you know a better method than the meters? Or do you say: let's not poke that issue as long as we don't wipe and hope it stays that way?
About the tank healer: the text is changed to reflect this idea.
A reliable method for measuring heal? No. But what can be done is to investigate the meters in depth bearing in mind assignements, encounters, clas mechanics etc.
Its no easy task and can not be formulated. But it can (and should) be investigated. Of course doing that or not seperates the true enthusiasts of the game from the indefferent ones (or acording to the vague leet vernacular hardcore/pro from slackers/noobs).
WWS reports over a large number of raids have worked for us and we more or less know where we are merely effective and where truly efficient. But with current content there are many things you can get away with.
You mention that players need to look into these things and improve. And that its their job to do so. I can not agree more. That is exactly what they should do. But truly appreciating the available info - particularly when opinions vary amnong for example guild members - is no easy task.
As all others, I'm going to add to the choir of "you're completely off-base".
The first and foremost reason healing meters are useless is that they are not only inaccurate but also miss several heals. You don't do book-keeping by rounding figures to the closest 10g number and discounting parts of your income and expenses - if you do that in the real world you end up broke in the best case, in a courtroom otherwise. And that's where your entire point already falls flat.
The second reason healing meters are essentially a worthless tool to compare different healers, as others have pointed out already, is that you can't compare healing performance across different healing roles.
The third reason is that healing meters have no way to account for the damage prevented, eg through PW:S. This is also part of a healer's performance, but you can't figure that one out by looking at your recount window.
The fourth reason is that the numbers in themselves are meaningless. To evaluate a healer's performance, regardless of whether you look at it standalone or in comparison to others, the key metrics is to understand who healed whom when for how much. Understanding the circumstances which lead to selecting a healing spell and measuring the impact that particular choice had is indispensable.
The only way is to go through the timeline of events in a combat log and evaluate every spell cast by a healer based on the circumstances leading to his choice of actions and their consequences. A healing meter doesn't show this.
Using healing meters as the comparison tool is akin to just holding books by checking the amount of gold you have at the end of every session. I believe we will both agree that this isn't something you'd advocate.
I think they are useful, I do trust our healers though, I give them a moment at the start of the raid to discuss whats best in healing rotations and jobs, though the meters are useful, you can work it all out easy from checking their mana and how well they use it compared to their healing done/overhealed... as a raid leader its things like this that give me the chance to think "if i pick him and her, this will work".
So even if we have a few healer that are not up to it or geared enough I can balence it out with others.
Healing meters are certainly not useless.
They are certainly not a ladder of skill either.
They are a guide to how much healing has been done by each player, nothing more and nothing less.
It's up to those analysing the meter to take into account the other variables in the situation.
Other abilities or jobs not influencing healing meters need to be considered and are generally of high importance but hard or impossible to measure against healing done.
@ gevlon (your latest comment)
we raid very simply... you do not compensate for another healer... because you have a job and if you are busy compensating for someone else... you are doing 2 things wrong... first - you are taking time away from your assignment, and 2nd - you are making it seem like that person is pulling their weight when they are not...
If a tank dies, and we wipe... we figure out why the tank died... we consult grim reaper and the players and figure out what happened...
if the healer let the tank die - he is told to do his job correctly - and we figure out if it's possible for him to do so. If it is and he's just not doing it for some reason (laziness, stupidity, wasn't paying attention, etc) then we can replace him
you can't fix something if you don't know if it's wrong because someone else is compensating...
this is why our tanks don't blow shield wall just because they get low... can u imagine if a healer stopped healing me, I blow shield wall then later the boss (which for the case of example is maexxna) hits her soft enrage, and I have no shield wall to survive with... it is my fault, not the healers...
you can't fix a problem when you don't know if it exists... and if you are using "band-aids" - the problems will come out in time - better to address them sooner
Tobold's right, the overall point is did the assignments survive and the boss die? The only other point I'd add is did I have enough mana at the end?
Lance is right. The other night I was in a 25 man raid for 4 hours. I had nearly 50% overheals, but that's because my JoL healed for 4.5 million second only to my HL and 80% of it was overheals. Should I not have used it? I'd be stupid not to!
Healing meters are NOT worthless. You just have to read them differently. It's also a great tool to help you fine tune your healing by seeing what others are doing.
Players who aren’t competing for the top spot on healing meters are often the ones that shout from the mountaintops how worthless they are. Competitive healing can and does work. Any healer that blindly sticks to their assignments and stands around with ample mana and watches some other healer’s assignment die is a bad healer.
Being able to handle your assignment and help out others who are weaker players is what makes good healers great. They earn their spot on the top of the meters. Don’t discount their efforts and blame them for healing your assignment because you were too slow.
I would go most of your way with "healing meters have flaws and more of them than damage meters, but still better than not measuring / benchmarking / using actual numbers."
However, I think there is an aspect of overhealing you leave out: insurance. There are times where the boss/debuff/tank gear mean that there is a chance for Sudden Tank Death Symptom. E.g., If you spend US$5k on health/life/car insurance (pls no politics) last year and didn't get sick, then it was wasted. But it was still the right thing to do. Similarly, there are times when looking better on meters (not overhealing) is bad for the raid (increased chance of wipe.)
I don't raid at your level or skill, and all the tanks in our guild love (/sigh prefer) druids, but I suspect that Holy Paladins might have an even larger advantage on 2 tanks due to Bacon of Light.
Holy crap you guys are getting waaaay too close to CQI (continuous quality improvement) and cost-effectiveness/efficiency analysis of health care which is what I do in RL. I read WoW blogs on my break to GET AWAY from this crap.
But, nice to know there is some application in game life.
And there really is not way to assess QUALITY of healing by just looking at numbers of patients seen, or number of prescriptions written, ie using healing meters. You must spend vast amounts of time/resources reviewing charts and assessing why mistakes/misadventures happened, and deconstructing systems (shudder). And even that is limited by good documentation.
In a WoW raid wipe, look at who died first, ask their healer if they know what went wrong, maybe reassign healing targets, and try again. Meters are for bragging rights & motivation, not for /gkick. Good healers don't do it for the loot, anyway. They heal because it makes them feel like a God.
Perhaps your raid healers should be encouraged to work among themselves to improve and give each other feedback. Occasional "rounds" or "M&M" (morbidity and mortality) meetings among themselves (and of course administrators) helps spread good skills and communication habits. And they should be aware of the concept of "triage", prioritizing targets based on importance and acuity of damage taken. You can ask a Blizzard Corp. Rep. to bring lunch to your meetings. Maybe they'll sponsor one in the Bahamas!
A warning, though. Under no circumstance should a healer EVER admit in writing that his actions may have caused a wipe. Ever.
Note : when I say meter in the following I use it with the idea that meter stands for having a ranking of your raid members.
Healing meter mean squat when looked at with no knowledge of the context. I'd go further to say that DPS meter is not either an appropriate ranking tool, let's take a quick example : on Maexxna there are one or two DPS who are on cocoon duty, which means they will lose a ton of DPS getting in range and bursting down those cocoons.
Healing meters have the same problem two orders of magnitude greater. You say it's the "booking" system which is the problem, I agree up to a point, but to have any meaningful ranking system you would have to come up with equivalences. For example you would say one dispell equals 1000 HP in healing ? 2000 HP in healing ? How would you make it not arbitrary ? Maybe you would equal it to the healing required if it were not dispelled ? How would you do that on non damaging debuffs ? It would require a combat analysis which is far too complicated to do.
Moreover some things are simply not doable in terms of code required (I'm a fairly good lua coder) : there is no way in the combat log for example to link the absorb of a PW:S / divine aegis / sacred shield to a given healer, except heuristically and even then there are limits.
Healing meters are only good insofar as they give you some tools to understand what went wrong. In that respect the "if his targets did not die the healer did his job" is very true and maybe the only meaningful way to turn that into useful information that is through death analysis (Death window in Recount or Acheron addon).
There is however something that is not done in those addons that would be relevant : record for each deaths what were each healers doing at the time.
The way we used to measure healers in bc was to have them 2 heal a za run. The smaller scale raid puts more pressure to perform on the healer and reduces the chance that somebody can cover up for their shortcomings.
At the end of a moderately challenging (for your gear level) 2 healer run of a 10 man instance you will know the skill level of your healers.
The way we used to measure healers in bc was to have them 2 heal a za run. The smaller scale raid puts more pressure to perform on the healer and reduces the chance that somebody can cover up for their shortcomings.
At the end of a moderately challenging (for your gear level) 2 healer run of a 10 man instance you will know the skill level of your healers.
You've got some interesting points about economics, but when it comes to healing you sound like one of the M&S.
As a healer, your job is to make sure YOUR assigned target is alive. That's how you're evaluated.
If you're assigned to MT heals and you decide to raid heal instead, you're a bad. If you're assigned to cleanse and you feel like spamming Healing Touch instead, you're a bad. And if you then try to post healing meters as "proof" of your "skill," you will get laughed at because it doesn't take skill to ignore directions.
I know you are fishing for ways to assure yourself that you're not one of the bads, but if your heal assignments routinely die then you're failing as a healer no matter what numbers you put up. By intentionally failing, the way you brag about doing, you're admitting that you are being boosted by the other healers who are having to pick up slack for you.
I do agree that there are good and bad healers. I do not agree that the currently existing healing meters are useful to determine this in any way, shape or form.
Metrics that work:
- Count the dead people at the end of the fight.
- Check which players died first.
Problem with both:
- It's often not the healers fault, but the dying player's: stood in the fire.
- It is incredibly low-grained. 3 casters dead means what exactly?
With dps, you can easily compare, because with a mob having 20 million HP, it really does not matter when you use which attack, as long as the total is big enough. With healing, it's the complete opposite. Healing someone from 12k to 13k just before the second 12k hit comes in is worth A LOT. Healing someone from 20k to 24k in a fight where the biggest hit is 9k is zero difference (death in three hits in both cases), but will show up on the meters.
It is actually a good idea to *not* snipe off the top 10% of targets so CoH and WG can top them off for free, and you can spam the tank a bit more.
Hots and Shields not getting counted is only a slight problem compared to the uselessness of "healing data".
Or would you really think that the ret pala is the best healer we have, just because he heals for more than anyone else on saphiron?
I disagree with the point you made about HoTs.
Speceificaly a claim that 'HoTs are not heals'
Given that some are instant casts, they are a perfect way to heal AoE burst damage to the raid.
Also druid HoTs can be burst consumed with another instant.
Finally a point about healing accounting. I do not stress about getting on top of healing meters. My job is to keep the raid alive not to score. Every time I shoot for the top spot, I overheal to buggery. I have many raid heal addons and one of them tells me who targets my heal targets.
In a raid with good healer, 2-3 OTHER healers will target my target. I ordinarly just throw a quick instant HoT and switch to an untargetted secondary heal target. I do not the bonus points but raid gets better overall healing and mana preserved. All around smiles with the exception of a poor raid leader who comments on my 'low healing'.
Post a Comment