Why MMOs Should Move Past the Trinity


Note: This article originally appeared on MMOGames.com.

I remember well when I stopped supporting the trinity. It took some convincing, particularly as this wasn’t long after the lackluster approach to its removal by what was, at the time, a still freshly launched Guild Wars 2. Up until then, I’d wrapped myself up in the arguments still employed by the trinity’s contemporary advocates, foremost among them the still ubiquitous misconception that the trinity somehow possesses a monopoly on the potential for defined party roles in MMORPGs. Perhaps I was conditioned by what I now see as the incredibly restricted variation on encounter design found in trinity games, or perhaps it was simply a lack of imagination. Regardless of the source of my conviction, once I learned to abandon it, my eyes were opened to a whole new world of possibility waiting to be realized within the genre.

What is the trinity?

Crowfall Knight Banner
A key part of the mythos perpetuating the trinity revolves around the fundamental misunderstanding that without it, MMORPGs would be left with no defined roles. This is factually incorrect; in fact, MMORPGs could even abandon the trinity while still retaining tanks, healers, and damage dealers as their primary role archetypes, if they were so inclined. To explain how this is possible, we have to delve into what exactly the trinity is, and the mathematical relationship between tanks and healers in encounter design.

Anyone who has played an RPG seriously is probably aware of the concept of a character’s Effective Health Pool, or EHP. For a simplistic look at the concept, imagine that tanks have two defensive stats, vitality and armor (in practice, this often more complicated with other forms of resistances coming into the mix). Vitality increases the tank’s health pool, while armor provides a percent reduction to all damage received by the tank. A tank with 500 health points from vitality and 50% incoming damage reduction from armor will have an EHP of 500 + 500 * 50%, or 750 health points, which is the true amount of incoming damage the tank is able to take before dying.

While increased EHP for the tank does provide some utility in enabling them to survive spike damage, its true power is in the multiplicative effect it has on the efficacy of the group healer’s efforts. If the tank has 50% more mitigation than the other members of the group, all healing received by the tank is effectively 50% more powerful than if it were received by any other member of the group. This mathematical synergy between the tank’s mitigation and the healer’s regenerating effects allows trinity-based group compositions to survive challenges with a far steeper numerical challenge than groups using any other format.

Thus, mitigation is the true purpose of the tank in a trinity set up. Maintaining aggro, getting hit, and everything else you actually see the tank do is just a means to that mathematical objective of stretching the healer’s contribution further and further. Theoretically, a tank could perform its role just as well with skills allowing it to absorb a portion of its allies’ received damage, or by applying damage reducing debuffs to its enemies, which is an approach we’ve actually seen taken for PvP tanks by games like Warhammer: Age of Reckoning and Star Wars: The Old Republic.

Damage dealers, the third member of the trinity and ostensibly an equally important part of the strategy, are in reality largely ancillary to the synergy between tanks and healers. Effectiveness in this role only becomes important with the introduction of arbitrarily restrictive mechanics like enrage timers, which are necessary to prevent groups from “cheesing” content by overloading a group with defensive archetypes to further exploit the trinity’s innate synergy. While specialized damage dealers are an essential part of progressing through the actual encounters, it is the tanks and healers who are the true stars of the trinity system.

Put simply, the trinity is the specialization of labor in the talent and item economies that make up the two sides of the formula for a group’s defensive capabilities, and it is very good at what it does.

Why should MMOs move past the trinity?

League of Legends Faceoff BannerThe effectiveness of the group composition is also the trinity’s downfall, because as a result, every encounter in a trinity game is limited to the same basic premise; the tank runs in, grabs aggro, and the boss stupidly attempts to kill the single individual in the group that it will not be able to kill. Due to the combined strength of a trinity tank and healer working in tandem, any damage scaled to challenge them will be too much for any other group member to survive, and if the damage were scaled down, it wouldn’t provide a challenge for a trinity group at all. This means that with the exception of specific scripted attacks individually scaled to be taken by the rest of the group, the boss must focus almost all of its attention on the tank in any given encounter.

As they say, this is why we can’t have nice things. Having dynamic fights or actual AI instead of scripted encounters are ideas wholly incompatible with the rigid formula basing a game around the trinity mandates. The trinity must have simplistic aggro tables allowing the tank total control, or else it ceases to function. This is why the system falls apart as soon as players are put into a dynamic environment like PvP.

None of this is to say that you can’t have good, fun, or challenging encounters with the trinity, as developers have been providing all of that for decades. However, almost all of the unique and interesting mechanics we’ve seen in the raid encounters of the past could easily be adapted to function without a traditional tank and healer set-up. In 2016, are we really not ready to move past the scripted boss fights of the 1990s to dynamic encounters with actual AI?

What should the trinity be replaced with?

Conveniently, we already have an obvious example of multiplayer games with dynamic fights and clearly defined group roles that don’t rely on the trinity: MOBAs. These PvP only titles include pre-made ability kits conforming to a variety of combat roles, including those typically found within the trinity. Tanks, while still pretty beefy, perform their role with a heavy emphasis on close-range control and disruption, while support classes provide a wide variety of buffs and debuffs, frequently including the ability to heal. It’s important for MMORPGs to include distinct roles, and there are proven avenues for doing so that don’t create a trinity-like situation.

Many MMORPGs are already taking cues from MOBAs for class design, but so far this has largely been seen in PvP focused titles. Adapting this class and combat design philosophy to the large-scale MMORPG boss encounters we all know and love would not be without its challenges (and believe me, I am not proposing that we make just make PvE play like PvP and call it a day), but I believe that with the right team, these challenges would not be insurmountable. That said, it is absolutely crucial that attempts to leave the trinity behind include a substantial rework to NPC AI and boss encounter design.

Where have existing non-trinity games gone wrong?

Guild Wars 2 Battle Lines Art BannerIt’s hard to talk about moving past the trinity without addressing the specter of previous attempts to do so, and while Guild Wars 2 remains a very popular and highly regarded title, it’s always the first to be brought up as an example of why abandoning the trinity just doesn’t work if you want to have good group PvE. So why didn’t it? As stated in the previous section, the trinity must be replaced with two distinct parts: a new philosophy on group roles, and improved AI.

In the case of Guild Wars 2 specifically, the game actually has an interesting and unique perspective on group roles that doesn’t get the credit it deserves in the popular narrative. Many encounters calling for group members to bring specific utilities running a gamut from buff stacking, projectile defense, blocks, condition clears, CC immunity, and more, all of which engage players by requiring them to pay attention to the fights and use the appropriate skills at the right time. However, while I see the virtues of the game’s philosophy on group roles, the complaints that “everyone is DPS” are too widespread to be merely waved away as people missing the point.

Class homogenization and role impermanence, both intentional design choices stemming from the game’s designers really wanting players to be able to form capable groups of any class combination, gave most classes access to a wide array of utilities and allowed players to switch trait and utility choices freely whenever they’re out of combat. While this succeeds in making almost all group compositions viable, it has lead players to feel no attachment to their role. No one thinks “I’m a reflector!” or “I’m a condition clearer!” in the way a player in a traditional MMORPG can identify as a tank, healer, or damage dealer. Additionally, a character’s stat choices play no part in the effectiveness of almost any of their utilities, making it even easier to swap roles and leading to the game’s glass cannon “zerker” meta, in which every member of a group uses gear with itemization only optimized for damage dealing. When combined with the game’s surprising lack of effort towards explaining the way different utilities contribute to group composition, it’s no wonder so many players have fallen back on trinity MMO terminology to see every player in their group as simply DPS.

When it comes to implementing dynamic AI and using a fresh new take on encounter design, Guild Wars 2 does fail completely. Its dungeons, which have now been abandoned, disincentivized, and all but removed from the game entirely are a poorly designed, buggy mess that easily rank among the industry’s worst PvE experiences. There was almost no innovation on AI or mechanics when these were designed; they are simply the standard, scripted encounters of the past without the order provided by the trinity. As a result, they were a serious downgrade from their forebears, and in this state, their inclusion probably caused the game more harm than anything else.

The importance of replacing the trinity with something better, rather than simply removing it, cannot be overstated, and while Guild Wars 2 made a respectable, if flawed, attempt at redefining group roles its lack of innovation on enemy AI and encounter design may be the larger issue that held it back. It will require both before the genre is able to move on to something better.

Should all MMOs abandon the trinity, then?

Guild Wars 2 Healing BannerOf course not. The best argument in favor of the trinity is simply that some people prefer controlled, scripted encounters to more dynamic content, and those people aren’t going anywhere. Most people consider real time action to be an improvement to turn based games, but that hasn’t diminished the popularity of games like Final Fantasy, XCOM, or Civilization. There are more gamers now than at any time before in history; we don’t have to be scared of catering to people with differing tastes any more.

Personally, I see a natural synergy between action MMORPGs with aiming, dodging, and limited action sets with the implementation of dynamic encounter design and the non-trinity group roles they require. As the more frenetic combat found in those games seems like a far more natural companion to ever-changing, less predictable fights than the more controlled, detailed rotation oriented combat found in traditional, tab-targeted MMOs.

The question of MMORPGs moving past the trinity is not a question of if they should, but a question of when they will. Eventually, our genre of choice will succumb to the allure of more modern, dynamic AI of the type already found in other genres, and when that happens, the trinity will cease to function. I, for one, embrace our new AI overlords.

4 thoughts on “Why MMOs Should Move Past the Trinity

  1. Hey Isarii, I enjoyed reading the article. After reading, however, I think my initial reaction is one of disagreement. My feelings mostly boil down to an argument you make yourself justifying the trinity as a design principle. “Put simply, the trinity is the specialization of labor in the talent and item economies”. I believe this specialization is inevitable for groups of players who have a prior intention to succeed. Specialization will occur regardless of whether or not content design strictly requires it.

    You reference MOBAs as a style in which the trinity is not enforced in terms of design, but I think it’s not really a valid example. Even under the MOBA archetype there’s a default winning formula that specifies running a healthy of roles: tank + support, offensive magic caster, self-sustaining bruiser, and an assassin. Early game you’re 1:1 or 2:2 at which point there’s no trinity because there’s fewer than 3 people involved in the encounter. Late game, however, when you get into 5:5 team fights you’re going to be screwed if you don’t have a healthy specialization of focus on the team. A 5-tank or 5-assassin team is going to struggle big time late game in most games.

    Anyways, to get back to my main point: the only encounter design strategies for which specialization within the group is not ideal are really boring ones. There’s also a big concern with what demographic of players you target with your encounter design. Consider two extremes:

    Case 1: Design such that any combination of player roles can be successful if the players work well together. In this case where specialization is not required groups who DO specialize effectively can completely overpower the content, thus alienating the top-end competitive demographic.

    Case 2: Design such that only a specialized group can succeed, requiring groups to have a mastery of defensive, offensive, and support skills. In this case non-specialist groups cannot effectively complete the content creating a wealth/power gap between the can’s and can not’s.

    Ultimately, when it comes to moving on from the trinity, I’m not convinced that there is anything to move on TO. Perhaps if you define trinity as “you must have a class who specifically heals and does no damage”, then its easier to have this argument. Character specialization and identity is part of what makes MMOs so great, and I think the only way to dilute the incentives to specialize when confronting group content is to remove the ability to specialize. That’s not a design route I want to see games following.


    • Actually, MOBAs having defined roles and preferred group compositions is part of the reason I chose to highlight them as an example. I do think players should have roles they identify with, and while MOBAs come with a certain degree of flexibility that I find appealing, I think that in a PvE game, there needs to be a ‘meta’ that developers can rely on players following when designing how an encounter will play out. Given your examples of Case 1 and 2, I’m firmly planted in the Case 2 camp; just not with the trinity specifically.

      My issue with the trinity really boils down to the way its use constrains encounter design to scripted, controlled encounters that play out the same way every time. We don’t have AI or dynamism in MMO PvE right now, and the way mechanics like aggro function seems anachronistic when you compare our genre of game to almost any other.

      Pro roles. Pro specialized groups. Anti stale fight design. It’s time for something new.


  2. Enjoyed reading. Maybe the best write-up I’ve seen on the topic. I’d never given much thought to role identification. I wonder if there is also room for compensation through the encounter and the AI. I.e. the halfling who is good at sneaking, running, and pulling the levers, etc. Love your “about” blurb too – Stockholm syndrome victim. I’ve always thought of it as sadomasochism.

    Liked by 1 person

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s