Tobold's Blog
Friday, December 26, 2008
Healer competition

In the recent Blame the Healer thread, Markash linked to an interesting discussion that originally happened on the WoW forums, and was archived at MMO-Champion. And what I found especially interesting was the first part, quote: "Healer Competition: While most of PvE is cooperative, raid healing is stupidly competetive with poor support mechanics for multiple healers healing the same damage. It also creates an environment where faster spells are artificially more valued because it means your heal actually works. To overcome this with communication and mods takes an unproportional amount of effort."

In other words, if 3 dps classes hit the boss mob simultaneously for 10k damage each, it all adds up to 30k of damage. If 3 healers hit the main tank simultaneously with 10k heals, chances are that half of that is lost in "overhealing". The reason for that is basic math: A raid boss mob in Wrath of the Lich King has several million health. A main tank only has somewhere between 30k and 40k health. My priest currently heals for up to 9k with a Greater Heal, up to 14k on a crit. In other words, my crits heal a main tank for between a third and half of his total life. But with a casting time of over 2 seconds, I'm not waiting for the main tank to be below half before starting to cast. And if the main tank *was* down to half, another healer in the raid would probably start healing the main tank too, as even with healing assignments you never know what happened to the other healers. So overhealing happens all of the time, even in the best raid groups. Damage comes in irregular chunks, so healers need to keep everyone at or close full health, and that means that a good part of all heals is wasted. DPS is rarely wasted, because the best damage ability anyone has is still doing far less than 1% of the boss mobs health.

It would be wrong to judge healers on how much or little overhealing they dealt. Overhealing is the lesser evil, much better than letting the main tank die. As holy spec priest I'm often on raid healing duty nowadays, where I can use the "smart" spell circle of healing (until the nerf next patch), which automatically chooses wounded targets if in range. Way less overhealing, and much higher up on any healing meter than lets say a discipline priest keeping the main tank alive. But it would be silly to believe I would be doing a better job because of those better score. Healing meters are misleading. The only important thing is whether everyone is alive at the end of the fight. Avoiding overhealing ultimately falls under the category of mana conservation, which is important for long fights, so somebody who is overhealing far too much risks letting his target die later when he is out of mana. But trying to heal in a way which completely avoids any overhealing is ultimately the bigger risk. The 2 to 3 seconds a priest's greater heal needs as casting time is a long time in a raid, and from the moment you decide to start casting to the time the spell lands, a lot of things can happen. Using faster heals is less mana efficient, so avoiding overhealing that way won't help either.

I'm not sure how Blizzard could redesign healing to be less competitive, to avoid several healers working at cross-purposes to each other in a raid. I'd love a talent that gives me back part of my wasted mana when I overheal, but that would take away a lot of the skill currently needed for healing. Having some sort of display of which healer is casting what spell on what target sounds nice, but is probably too confusing in the heat of the battle. Maybe one interesting option would be to convert overhealing into temporary health. Lets say half of the points I overheal, up to a certain limit based on the total health of the target, remain on the target as damage shield, lasting 3 seconds. That way the points aren't totally wasted, because at least they enable all the healers to heal a bit less for the next 3 seconds on that target.

How would you redesign healing, or add new talents and abilities, to avoid overhealing?
Mandatory raid mods take care of:

Healing stats (healing numbers, overheal numbers, total healing attempted).

Stat queries. How much mana does everyone have? (with correlated healing stats, you can see who's the weak link in terms of gear or performance)

Target/ability announcement. Who is healing who, broadcast through a subscription/listener mechanic so only the people who are interested get the message, and they get it in a format most useful to them (simple text, as a screen alert, whatever).

The mods we had included Warrior abilities being announced for finer control over healing ranks.
A communicative tank works just as well.

Progression healing is inherently spiky, making high end encounters near-random. This is a weakness of design, in WoW.
There aleady is talent that gives you back mana on overheals (only on overheals from Flash Heal and Greater Heal) - Serendipity.
And discipline spec can already do damage shields on tank (not from overhealing, instead it gives the shield from any crit) - Divine Aegis.

Your point still remains valid, neither of these does not solve the competitivness of healing.

One possible solution could be to rework healing spells into "buffs" that increase effectivness of self heals, and let each character heal himself. Or make healing spells like lightwell. Healing on tanks could be done with damage shields. But such changes would most likely upset a lot of people.
It is a basic issue with healing that there's no point healing someone past max health, whereas as long as a mob is still alive, more damage is always useful.

The best solution probably is either a talent for allowing overheal to be saved up and used by the target later, or letting healers switch seamlessly between healing and dps, depending on what is needed.
This comment has been removed by the author.
I like your idea of allowing overheals to actually have an effect on the target. This works well in Team Fortress 2. I think it would be ideal to actually make this a quality of the tank's gear though, so they could choose to gear up for less wasted healing. Then there'd also be the possibility to let dps classes convert their overheal into a damage buff etc, depending on how they were geared. You might even allow healers to overheal each other (but not themselves), to increase heal potency.

I think fight pacing has a lot to do with this as well. If there's a nice 2 second windup for a huge attack, it's a little bit more time to make sure that a healer also has a chance to get their cast started up too. Not that attacks can't be fast at times for more difficulty, a slower rhythm on average could be helpful.
Does over-healing really matter anymore considering healers have endless mana pools now. It is a freakin' joke.
It has been a few years since I played WoW, but does not healers have any "heal target when hit", "convert part of dmg to heal", "reflect dmg", "share part of dmg" type of spells?
One solution would be to add graphical elements to the healing spells themselves.

For instance, all healing spells would immediately upon the start of casting create a visible connection (whispy strands of ether, etc) between the healer and intended target(s). The graphics for the connecting strands might be color coded so that deep red tendrils signal a large heal while blue ones are lesser heals. Moreover, one might do something like adjust the opacity of the connecting strands as a function of the healer's mana. Partially translucent strands hitting your main tank? You might need to start picking up the slack because that healer is now close to going oom. Perhaps one could also incorporate some sort of measure of the progress of the casting of the heal into the graphic (a "bulge" in the strands moving out from the healer as a function of cast progress).

I think something like this could help make healing more dynamic, reactive, and less tied to UI mods. It would also help to stave off healer competition as one would immediately see who the others healers are healing and how they're doing.
The whole game mechanic of a boss mob hitting for half the tank's health, and then the tank being healed that amount in one spell is just silly for reasons you stated in addition to others. I think there should be no heals that heal for 50-100% of a MT's health. Then with smaller heals the dev's don't need to create bosses that can hit the MT for 50+% of the MT's health.
Why wouldn't simply organizing raid healing help out with this problem? I would just assign the healer some targets to focus their healing on and if their targets are fine they could check to see if the group next to them needs help, and if they do not need help, why can't they just throw in some DPS if they don't have anyone to heal, would help beat the rage timer.
Easy fix - make any overheal hitpoints convert to damage on the target's target!

That way healers will not only be doing there normal healing job, but contributing to dps also !!
I don't like the comparison to the DPS role. Healers are playing a defensive role, while DPS are playing an offensive one. Therefore, they face different challenges. It seems like complaining about how other healers can cause you to waste your healing is like a tank complaining that he can't hold aggro on a mob - only because another tank is already doing it! Sure, it might seem like a waste of the healer's (or the tank's) talents, but all that should really matter is (as you said) that the job gets done. It's not like DPS, which has essentially no upper limit, always being able to be improved upon in order to bring the boss down faster. Healer's play a totally different role. The real way to judge your skill, or another healer's, shouldn't be a healing meter - rather, you should just look around to see how many people are standing at the end of the fight.
There already is a talent that refunds mana on overheal, Tobold. You don't have it? ;)


That aside, I agree that healing's implementation has seriously outgrown its design. But it's way too late in the game's lifecycle to expect it to be fixed.
I have seen a lot of good ideas on here, but I tend to stay on the side that thinks that healing and healers should never be looked at by numbers, but rather, did the entire group/raid survive? Were there deaths that were avoidable? I agree it becomes annoying sometimes as a healer when multiple healers focus on the same target, but honestly I attribute this to lack of knowledge. There are several add-ons which will show a target is being healed. I also agree with the notion that healers should be assigned targets. For instance, my pali has no true AOE type heals, so generally I will take the main tank or off tank and a class who can heal several targets will focus on the group/raid party. Is it perfect? Not always, but it does give some direction and of course if I see someone is extremely low and the other healers are busy with another member, I will try to cast a heal on them to pull them through. I think it all comes down to the skill of the player. There should be some sort or challenge in healing. Had I desired to do DPS I would have brought my hunter, lock, or mage. I also like the fact that each healing class is a little different in what the excel at healing wise and what other options they have available. There have been times where a tank may go down and if we are running a 10 man with three healers, I always can jump in and try to pick up the slack of the downed tank. Am I great at it? Hell no, but I am able to be utilized in that way. Anyway, that is what I think on the subject.

Hope all had great holidays!
I blame mods and people for the "stat craze" rampant in WoW. Extensive modding offers a lot of improvements and many mods have been incorporated in the Blizzard UI.

But mods often make things too easy, take away from the gameplay aspect by automating things to a certain degree. Plus DPS and Heal meters indeed cause discrimination. EVEN if you do more useful things than the top DD/healer: Spreading Curse of Elements or other curses on more than your target helps the group, but lowers dps for you. In longer fights it still pays off, but if you want to lead the damage meters in 5 man instances, Rain of Fire for Warlocks all the way e.g. with destro spec. Not even Seed comes close. A healer conserving his mana and healing when necessary is good on the overheal meter, but maybe worse on the overall heal meter. A Rogue sucks in 5 mans, as he cannot AoE as good as other classes. But does this take in account his interruption abilities or endurance, not being dependant on mana?

But most raid leaders, even good ones, just go by their WEBSTATS, RECOUNT, whatever meters and a few ingame observations. We, the players, take these meters too serious.

I did some instances with good players that hab rather basic equipment, and our tank asked if I could not ask for another DD, as they were far behind me (pimped to the extreme) in damage. He did not overaggro, he played smart, had some nice tricks and good crowd control. And we had no single dead player after the Mechanar. The tank got lead, and kicked the Mage. I left rather pissed, the healer, too. Then we suddenly got all re-invited...^^ The tank apologized, but said the Mage should work on his damage output.

We should really get rid of all the various meters and mods... but for WoW, it is already too late. We can only hope that common opinion regarding the interpretation of the results becomes more sane.

Damage races between two DDs that overtax the tank just in order to lead the meter were already my pet peeve in Burning Crusade, they are damaging to the party.
If all healers in a raid use healbot it shows nicely what incoming heals are expected on a target.

The problem still persists however as many bosses do imense damage and spam healing is called for (Patchwork for example).

The stats are so high at the moment that the only stat healers should worry about is mana regeneration. A priest for example with high mana reg but lower +heal is more worth than a +2k heal priest. Sad but true.
The main issue, as mentioned already, is that a boss hits a tank for 50% of his HP, and then a crit heal brings it all back. If a boss at best could take away say 15% of HP, and heals bring back 10% or so, overhealing would not be an issue, and timing/mana would play a bigger factor.

I think TBC and upping staming was an attempt to come closer to this, but it was not enough. If I recall correctly, LoTRO is a bit closer to this, and WAR is as well. Tanks have a ton of HP, and healing generally does not have the huge spikes it does in WoW.
...If healers werent bottlenecked into such a mundane wack-a-mole heal task maybe...

The problem with how blizzard balanced the plate of characters and abilities, is that there isnt much care put into anything besides the dps role. Tanking, well needless to say it is pve and it has a much higher importance to a group as a whole in comparison to how much dps a rogue can do. The problem with putting so much shift on a role is that a tank is not even useful standalone and requires heals and backup damage to even function in a group. And unlike a dps char always brings a asset to the raid, too many tanks or healers is a burdon. Blizzard fails pretty bad with this overlapping of needs for the class roles an infact with certain pve content requires very specific bottlenecks of classes to move forward.

Now for healers, the problem definately is with the overlapping role and lack of versatility. Ive never understood why blizzard made tanks soo wussy with their damage, and healers lacking either crowdcontrol or damage ontop of their main role. Warhammer does something right in terms of pvp and pve when it comes to designing chars and blizzard doesnt. Overhealing is a non issue because having access to big heals is great, if i were to contrast a healer class that had to constantly spam a smaller and more efficient heal (like a paladin) then you bring the element of forcing a char to constantly be spamming button vs pressing a heal and then having free space for doing something else.

As you say, not overhealing is one of the only ways for healing to be 'skillful'. I don't think it would be good to design it out of the game. Plus, healing in MMO's in general is more efficient than damage (ie heal spells do more healing than damage spells do damage), which helps to alleviate the issue.
The solution is better raid design in general.

There is no real solution to this problem until raid mobs are not able to kill a tank in 2-3 lucky hits in a row.

Muckbeast - Game Design and Virtual Worlds
not sure how to make it better in the first place, but it reminds me of a legendary thread from November 2006 on the wow boards entitled "How to Top Healing Meters", including how to avoid overhealing. using such tips as "don't waste mana rezzing, and use huge heals the instant after someone is rezzed", and other gems such as "If you macro some questions , like "I would to talk to you about your spec for a moment" and use them as an officer you can also interrupt their healing. ";jsessionid=1BCF2A9AF777A526674713D7DA5008AC.app25_04?topicId=51005333&sid=1&pageNo=1

pure win.
Math answer - let overheals temporarily give the tank more hp than his max. And gradually make it geometrically less efficient.

e.g. Tank with 1000 hp. First 50hp (5%) of overheals gives 75% (4 hp of healing gives +3 hp). Next 200 hp (20%) gives 50% (4 hp of healing gives +2 hp). Next 250 hp (25%) gives 25% (4 hp gives +1 hp).
Why even have healers? who is it that actually *WANTS* to heal?
Post a Comment

Links to this post:

Create a Link

<< Home
Newer›  ‹Older

  Powered by Blogger   Free Page Rank Tool