Sympathy For the Devs: MMO Healer Tuning
I was watching the most recent FFXIV Live Letter, and this question brought up a lot of memories from tuning raid bosses in WoW:
https://youtu.be/WRpdIL7_NII?t=17440
Essentially, the question boils down to balancing healing output to damage done. Of course the answer that Yoshida gave makes sense and is correct, especially from the perspective of communicating with the community. But I thought it might also be a good jumping off point to discuss some of the detailed difficulties in tuning content in MMOs.
Bring the Player, not the Class
This is something Ghostcrawler said a while back with regards to World of Warcraft, and outside of extreme balance outliers in specific situations, it has generally been true that the best players of the worst class perform better than the worst players of the best class. Some examples from FFXIV:
Here you can see several phenomena that are interesting to both raid and class tuning - First, that if you were to stack rank by 50th percentile damage done (the central line in the fat bar), the stack ranking would not be the same as if you stack ranked by upper quintile (the rightmost “whisker”). Or, in other words, the average Reaper does more damage than the average Dragoon, but the best Dragoon does more damage than the best Reaper. This is mostly interesting from a class design perspective, but from the perspective of a content designer and the overall tuning of content with respect to the community, the important thing to note here is that the gap in output due to player skill (the width of each individual box) is much greater than the distance from the weakest class to the strongest class if you fix the player’s position within the respective classes. Again, this has always basically been true in my experience in MMO tuning, outside of extreme system balance problems.
Additionally - and I will have to provide my anecdotal evidence here because gathering data on this is quite complex - but it is my experience that it is very difficult for players and groups to improve their performance at the game. While it is true that very dedicated players will focus on optimizing their class and output, in the majority of cases it is my experience that, when faced with a situation where a player’s only option is to “push their buttons harder”, outside of the most dedicated players, the response of many players is to give up. Especially when you must take into account that some players are dealing with poor internet connections, and/or physical or mental issues that make playing at an optimal level very difficult. Additionally, imagine you were to select a quartile of players, and select the lowest performing class (or the average of all classes at that quartile). That would give you a Damage or Healing number you could then use to set the minimum requirement for your encounter. However, what you are effectively saying is that, outside of improving gear, most of the players below the line you select will never clear the content unless they improve their performance. Because this is so difficult, you are in effect excluding those players from clearing the content.
The Role of Gear
Gear serves a lot of purposes in an MMO, but gear’s most critical purpose from the point of view of a content designer in an MMO is that, as players earn more gear, it increases the percentage of the player base that can clear content. If you take those statistics listed earlier, and somehow were to record all the player inputs and replay the fights they did, but increases their gear level by a given amount, it would effectively serve to shift all of the bars to the right. If you had previously set the minimum requirements at a given quartile of players, it would now allow a lower quartile of players to clear the content.
Or, from a simpler point of view, you can imagine a graph like this:
While the shape of the curve might not be the same for every set of players, every encounter, or every game, the same basic idea applies - if content is of a fixed difficulty with respect to requirements of damage and healing, the percentage of players who can clear the content increases as a function of item level. This is, in fact, generally a very desirable thing from the point of view of an MMO designer! As long as there is an avenue available for players to continue increasing their item level, they will be able to have a future wherein they can clear any encounter in sight. It encourages them to keep playing and keep trying. And, for the highly skilled players, the prestige of being the first, or in the first 100, etc. groups to clear content is a reward in and of itself.
Finally - and this is not necessarily relevant to the specific issue of healers, item level also impacts the tuning level of future content, and XIV has systems that both target specific content to increase the clear percentage without affecting future content (Echo) and to cap the effects of gear progression on a given piece of content (Item Level Sync). This isn’t specifically relevant to this discussion, but it does elucidate the way that the developers see these systems working together to establish the overall percentage of the community that can clear a given encounter at a given point in time.
The Unique Difficulty of Healer Tuning
We’ve talked a lot about the minimum requirements for clearing content, and in general, the difficulties of tuning around the minimum level of player power required to clear a given encounter tend to be the same for all roles. If the minimum bar is too high for a given role, nobody (or too few people) will be able to clear the content. Whether that burden falls on damage dealers, tanks, or healers does affect the overall social dynamic of the group, and since all players contribute to damage, you tend to see the party’s damage output used as the primary “gate” in hard content. The healer and tank roles are already relatively less popular than damage dealer, and the fact that a healing failure, for instance, is not usually able to be “made up” by damage dealers further discourages extremely high healing checks.
And, once players are able to clear, a difference begins to be seen between damage dealers, tanks and healers. As the group’s damage output increases, players are rewarded with a shorter encounter. While the same encounter but shorter is almost always an easier experience, the fact that damage is not generally “rate-limited” by the encounter means that damage dealers are always encouraged to fully participate in dealing damage.
However, because encounters tend to do essentially fixed amounts of damage relative to player performance regardless of gear level, the amount of healing required does not scale with the players as their gear level increases. This means that as healers’ healing output increases past the minimum required for the group to clear the content, the group is not rewarded to the same degree as it is with damage dealers. There are difficult to quantify ways in which you can argue excess healing capacity is rewarded - like less anxiety on the part of the group with regards to damage done, more opportunities for damage dealers to take risks, or more opportunity to mess up mechanics and still clear - but at the end of the day none of these are seen by players as clear wins in the way that shaving a minute off of the clear time of an encounter is.
In fact, if you are a long-time MMO player you are probably familiar with encounters that did rate limit damage done, and how unsatisfying they could be. Very old wave-based encounters, where waves of enemies would spawn on a fixed timer, essentially forced a minimum encounter length as players waited for enemies to spawn. As the player’s damage increased beyond the minimum required to clear a wave before the next wave spawned, players would end up just sitting around, waiting for enemies to spawn. Not a satisfying experience, and why generally these kinds of encounters began to trigger spawning off of various mechanics that are variations on “timers or the previous wave was killed, whichever comes first” or “spawn next waves after previous waves are killed, but set an overall time limit on the encounter”.
As a healer, if you see your desired gameplay as “as many of my button presses should be dedicated to healing as possible”, it is very difficult to continue to enjoy content at the same level as your gear increases. By the very nature of your increasing gear level, and the fixed damage output of an encounter, fewer of your button presses will be required to heal the fixed damage players are taking as you play through an encounter.
The answer that XIV has employed, and that in general has worked well in my experience, is having healers being able to contribute respectable (if less than other roles) damage done to the boss. This does somewhat go against the fantasy of the healer role - gear essentially acts as a factor that requires you to do your primary role less and contribute to damage more. And there are players for whom the fantasy of being a healer is just healing. They may not find that very successful. Additionally, having to have both satisfying healing gameplay and damage gameplay on a class contributes heavily to “button bloat” - to some extent if you are going to have roughly the same number of abilities available for all classes (a good idea, especially from a UI point of view) - you are going to have to simplify the damage dealing aspect of healing classes when compared to the experience of dealing damage on a damage dealer.
There are solutions here, but they have their own issues. If you are familiar with the Mythic+ system in World of Warcraft, that was used to effectively help content keep pace with player item level and continue to reward them as their item level increased. This approach is successful but has its own pitfalls, which are beyond the scope of this article. And, FFXIV has approached it in a slightly different way with its Ultimate encounters, which do not scale with gear. It is much easier to tune a satisfying healing experience for a fixed gear level.
All in all, it is a difficult situation and one that I’d ask players of all MMOs to have patience with!
A Final Note: Feedback
Finally - I would implore players to think about these issues as they are giving feedback to developers. While, on some level, it is a developer’s job to filter through player feedback to see “what the players really want”, this process can only be helped by clear communication. For instance, as much as possible, avoid focusing on what extremely skilled players are able to do is less important than focusing on your experience and whether you found it satisfying. If the best players in the world are able to solo-heal an encounter, does that affect you directly? Or, do you feel that you don’t have an interesting enough damage dealing rotation in the times that you don’t have damage to heal? Focusing on your own experience, and how it doesn’t line up with what you want, is much more helpful than overly focusing on what the best players in the world are capable of.