Tuesday, March 10, 2009

So where did these roles come from, anyway: An Intro

I've been thinking a lot about how two of my favorite games (D&D and WoW) and, by extension, two of my favorite pastimes (tabletop RPGs and MMORPGs) wound up with character roles: abstract "templates" that describe a character's essential function in the action part of the game (fighting, usually). WoW (like most MMOs) has the basic triangle of tank, DPS, and healer. D&D has always had similar roles, and in 4th edition we defined them as defender, striker, controller, and leader.

(If you aren't familiar with both genres, suffice it to say that tank = defender, healer = leader, and strikers and controllers are two flavors of DPS.)

But the weird thing? There's nothing inherent to either game in the abstract that leads you inexorably to the specific roles we wound up with. So how did we get here? 

I'm going to take a shot at answering that question in the next few posts. So call this a little blog series: "So where did these roles come from, anyway?"

But I'm going to start in a circuitous way. I think you can learn something about the emergence of those essential roles by looking not only at what a player picks as the role for the character he's actually playing, but also by looking at what a player picks for his next character--or (for MMOs) the character he's playing "on the side" in his spare time.

So let's talk about alts: alternate characters.

First: Warcraft Alts

Some of my best friends are WoW altaholics (people with multiple active characters). Once they get a character to the level cap, they start one or two others, and pretty soon they have a stable of decently-geared, max-level characters. I've always been sort of envious. Maybe it's because my main character is a level 80 feral druid (one of the game's better hybrids, and just plain fun to play), but I've never been much of an alt guy. I always have plenty of things on my main character's to-do list.

I have alts, of course.

• There's the undead rogue who was my first level-capped character. I stopped playing him only because I wanted to challenge myself with a PvP server (and thus my druid was born). I pretty much never play him anymore; he's stalled at level 61.
• I have a level 65 mage that I soloed up mostly to learn the Inscription profession and because I was taking a sabbatical from guild leadership and wanted to make myself scarce temporarily. He stopped at level 65 (the minimum to max out Inscription) and appears in Azeroth only to grab herbs out of the mailbox, turn them into ink, and (hopefully) make Noble Darkmoon cards.
• I've got a few low-level "experiments," pretty much one of every class. They're all between level 20 and 30.
• I recently started a Draenei paladin on a different server, mostly so I could enjoy the leveling-up process with a different set of quests than I'm used to. He's level 38 now, and I play him more than anyone else. 
• And like pretty much everyone, I've got a few characters that exist solely to run between the mailbox and the auction house. 

That sounds like a lot. But when I compare my arsenal of alts to the folks in my guild (which for obvious reasons tends to match my overall play style and available time to play), I'm just a piker. At least half of my guildies have at least two level 80 characters. For example, Broteas (resto shaman) has a mage alt. Shieral (bounces between boomkin and resto druid) has an enhancement shaman and a warrior that used to be our main tank. Divona (shadow priest) has a DPS death knight, a hunter, a holy paladin...and those are just her Horde characters. 

And here's the interesting thing: Almost none of them have two max-level characters in the same role (except for a few delightful but clinically insane friends who have five or six level 80 characters). On one level, that isn't surprising: people are looking for a change of pace when they select their alts. Vive la difference, right?

But on another level, it's interesting that those altaholics are throwing away all those man-hours spent learning the ins and outs of a specific role. As a guy who plays a hybrid (in my case, bouncing between DPS and tanking), I feel like I'm doing new content when I'm tanking for the first time--even if it's a dungeon I've run several times as DPS. There's no question that someone who picks a different role for their alt is volunteering to learn the game all over again. That brings with it some inefficiency (but to be clear, some fun, too).

And there's supposed to be differences in player psychology that would make you gravitate to a specific role. More on that in a second.   

Second: D&D "Alts"

When you look at alternate characters for D&D players, it's important to realize an essential difference between the tabletop and the online experience: The moment of character creation is far more constrained at the tabletop. Of course you can sit there with a sharp pencil and a stack of blank character sheets and make up as many characters as you want (and unlike WoW, you can make 'em whatever level you want!). But to actually play the new character, you've got to "retire" (often through death) your existing character, or you've got to start a whole new campaign. You can't level up your ranger at 2 a.m. when you're alone in your dorm room--not really, anyway. The tabletop game exists as a consensual experience with your friends.

So look at the two times you get to make up a new D&D character: character "retirement" (oh heck, let's just call it "death") and campaign starts. 

When your character dies, there's a strong incentive to roll up a new character in the same role as the old one. But where does that incentive come from? The other dudes at your table. It's an externality. Given your druthers, you might or might not choose the same role. But at the moment of character generation, you've got a table of buddies who are suddenly short a defender (or whatever). If you roll a leader instead, you hurt not only your new character's viability, but that of the whole table. The tendency toward "role persistence" is a function of the group's needs at the moment, not necessarily the individual's natural inclination.

In my long-running Thursday night game (spanning eight years and multiple campaigns, with long breaks when I was a new dad), the best example of this "role persistence" was my buddy Cameron, who was famous for a long line of "brave but lightly armored" characters. He played a lot of rogues and rangers, and, well, he died a lot. Cam plays aggressively and is a natural "instigator" in a good way; his characters actively seek out stuff to do, rather than waiting for NPCs to do stuff to him. Every DM wants a guy like that at the table. But the consequence is that Cameron's characters tended to be shishkabobbed, drowned, and/or stung to death by unusually venomous bees. 

When Cam showed up with his next brave-but-lightly-armored character, it was a natural thing to think that he liked that striker role. But really, Cam was looking around the table, saying: "Leader? Check. Defender? Check..." and so on. Put Cam in the other character-creation circumstance--the start of a new campaign--and bingo! Role persistence evaporates. When the whole party died in late 2007 (due in no small part to Cameron's rogue pulling a lever that should have been left alone), everyone made up new characters and we fast-forwarded a thousand years. Unfettered by the needs of the table, Cam made a dwarf fighter that was the very essence of a defender.

I spent ten years at Wizards of the Coast, so I always had plenty of campaign-start opportunities, and I was lucky enough to play in multiple campaigns simultaneously. As I think back, almost everyone in those games--professional game designer or not--bounced around from role to role when a new campaign started and their role choice was unfettered. Rich plays a paladin in one campaign, then a sorcerer with delusions of godhood in the next. Toby replaces his warforged fighter (rest in peace, Hammer) with a wizard. And come to think of it, my barbarian gets replaced with a wizard, and in the campaign after that I'm a melee cleric, and in the 50 Fathoms campaign after that I'm functionally a defender. (Miss ya, Roo!)

Let's Talk Psychographics
Based on my personal experience (the best data I have), there isn't much role persistence among either WoW or D&D players, once you subtract out the "we need a _______" factor that happens a lot at D&D tables and sometimes in WoW guilds. 

Yeah, my evidence is anectodal, but it's the best I have. I suspect that maybe WotC could tease better data out of the RPGA database (although organized, RPGA-style gaming is still something of a different beast than traditional "basement" gaming). And there's no question that Blizzard could extract more definitive data about how likely it is that a player prefers a specific role.

But here's the thing: Wizards R&D certainly believes that there are differences in player psychology that manifest themselves in role choice. In other words, some players are naturally drawn to play a leader, others a defender, and so on. I'm pretty sure that the Blizzard devs believe the same thing. And certainly some players are convinced that "you know, healing is really what I'm good at."

But given the proclivity of players in both games to happily change roles, I don't think the psych differences are that meaningful. And those psych differences are tenuous enough that I don't think you can say that roles emerged out of the psych differences in tabletop or online RPG players.

So the specific roles we wound up with--where did they emerge from? Maaaaybe from the source material: the broader fantasy genre. More on that tomorrow; I've got some offline writing to tackle first.

Out of Context: If it snows a lot, can we go saucing, Daddy?
Music: Red Hot Chili Peppers, Mother's Milk


  1. Interesting thoughts.

    I've been musing on ways to divorce a characters' Role from their Class. After all, a Class represents 'what they do', whereas the Role speaks more about the characters' attitude and personality. For example, a Wizard is (by definition) a Controller, but does it have to be that way? A Wizard who is a Striker would have a very different outlook on life - perhaps less tactical, more brutal and direct. Or a Rogue who's more of a Defender such as a bodyguard or Guild Enforcer who acts to protect rather than kill.

    One of the strengths of 4e D&D is that it's easy enough to achieve this with the multi-class feats. But I guess you already know that, right? :D

  2. "Externality"? Sounds more like a constraint to me.
    I'm waiting eagerly to hear where you go next to root out the origins of these archetypes. I've got my own theories...

  3. Aw, man, busted by Mr. Definitions. Oh, I mean _Doctor_ Definitions. Good point, though.

    And @greywulf: Yeah, expect me to delve into multiclassing in what I imagine will be part 4...if I can keep my hands off it that long.

  4. I'm a wee bit confused:

    The traditional fantasy roles emerged from d&d (which borrowed from Tolkien, right?)

    Ah but the PLAY STYLES, (ala bartle player types) are more rooted in the psychology of the player - perhaps the answer to your question is where "Tolkien meets the psychology of the player"?