A taxonomy of GMs

A little while ago I was in a conversation about RPGs during which my interlocutor reminded me that classifications of different sorts of character-players in RPGs go all the way back at least to an article by Glen Blacow in an early issue of Different Worlds (“Aspects of Adventure Gaming”, Different Worlds #10, October 1980). There have been many others since, derivative or mould-breaking. Some of them have provided the foundation for off-repeated advice to GMs about how to provide for various character-players.

At least so far as I am aware, classifications of game-masters have been less numerous and less influential, though GNS Theory is based on The Threefold, which passed through a stage of being a classification of the criteria on which GMs adjudicated events in their games.

Have there been any incisive categorisations of GMs that you would like to discuss?

How can an understanding of the taxonomy of GMs be useful to character-players? For choosing whose game to join, obviously, but what else? How can classifying your GM be useful in character generation and in making adventures successful and enjoyable?

Can a taxonomy of GMs enable us GMs to know ourselves better and improve our performance?


Hmm, an interesting question. I suppose if I had to suggest a system right now, I’d suggest a two-axis model in which the X-Axis is “Visibility” and the Y-Axis is “Preparation”.

Some GMs want to completely fade into the background, to present obstacles and challenges for the PCs to overcome without imposing a fixed set of solutions or suggestions. Other GMs want to railroad the PCs through a series of events while interjecting themselves into the story (either subtly or overtly). Mathew Mercer, I would suggest (based on the two episodes of Critical Role I have ever watched) is very much the second type, for example. And then some GMs create loose structures or “hooks” for their adventures and ad lib their way through each story point, while others create very elaborate stories with each major (and many minor) points outlined the whole way. Modern module design seems to also embrace both of these elements, as you can have books of just hooks (Tales from the Yawning Portal) or fully fleshed out adventure modules (Storm King’s Thunder).

Personally, I am “Like to be inserted in the Story” with “Likes little-to-moderate preparation” in my heyday of GMing (last time I GM’d a game was… '98, so over twenty years ago), but even the “little-to-moderate preparation” is too much time for me to properly run a campaign these days.

Although I did recently pick up the Modiphius Star Trek RPG (the one the SU&SD team used for their last live RPG one-shot at SHUX, I believe)… so hope springs eternal…


Original Threefold, which largely seems to have vanished, is I think worth repeating. As I remember it:

  • Gamist: what do the rules say, or what would be game-mechanically interesting
  • Narrativist: what outcome would make for the best story
  • Simulationist: what outcome would be truest to the physical processes of the world

I don’t think many GMs are purists for one of these, and probably any taxonomy will have to acknowledge fuzzy edges.


I’m not sure that I have story points. My view of GMing has always been that the players and I co-create the story. What I create is situations, but I’m never certain what the end state is going to be for a situation; my players regularly surprise me. I’m neither playing out a story with an elaborate outline, nor ad libbing my way through a series of predetermined story points; rather, like Gromit, I’m riding the train and frantically laying down the railroad in front of it.


The core of an adventure I run is usually “this is what will happen if the PCs don’t interfere”. They’re definitely moving things rather than the static dungeon that comes to life when you bash down the door.

That said it’s not unusual for me to have at least some idea of the story that may turn out in retrospect to have been what was happening all along.

1 Like

Do you ever use “if the PCs don’t interfere nothing will change”?

I prepare adventures in terms of designing a situation to which the PCs will react when they encounter it, and which will change in response to their reactions. Often that is a dynamic situation, such as a villain or opponent executing a nefarious plot, and it implies a “what will happen if the PCs don’t interfere”. But sometimes it is a static situation or a dynamic equilibrium, into which the PCs may introduce a bit of over-due revolution.

1 Like

That’s a great image.
I would argue that you probably have some hooks… I doubt, for example, that you sit down for a session with no idea of what might or might not happen? Although I shouldn’t project: I don’t do that.

Normally I will ask the players (before we start) what kind of game they’re interested in: politics, combat, interpersonal, that kinda thing. And then before a game, I will usually jot down something like “Giant mutant bugs in valley. Scientist in hidden laboratory in mountain. Revenge? Trying to revive daughter?”, and then have that first encounter happen and see where the PCs go from there (maybe they show no interest whatsoever in the hook. They just don’t care about the giant bugs. Cool, I always have three or four hooks ready. Maybe they go totally tangential from my initial idea… they immediately accuse the Dread Pirate Franklin that they encountered five sessions ago, and I start thinking “Huh… do I want the scientist to be friends with Franklin? Or is it actually Franklin and there never was a scientist?” Maybe they follow the plan I set out more-or-less as intended… actually, I don’t think that ever happened).

But I usually don’t plan out much more than that. I will sometimes have a few jotted notes about overarching plot points (persistent enemies, the movement of armies and nations), but after that I usually try to let the PCs take the reigns as much as possible and adjust the threats/story/rewards correspondingly.

Again, keeping in mind that this was twenty years ago I last did this, but that’s kinda my style. Very open-ended, very reactionary, but with enough in the notes to make sure we don’t just spend the entire session going “Okay, so you want to go to the store. It’s a pretty boring store. And then you go for a hike… nothing happens on the hike.”

It is comedy genius. For youngsters who may be unfamiliar with the source:

1 Like

I want to come up with interesting situations, and “interesting” is largely a matter of the kind of things that might happen, so I try to come up with scenarios that have the potential for things to happen, and thus I speculate about what those might be. But I don’t decide on one specific thing that is going to happen, and I am regularly surprised, not only by what my players choose to do about a situation, but by what I come up with to have happen in response to their actions.

In other words, yes, I do have hooks, but I wasn’t suggesting otherwise. What I was saying I don’t think I have, a lot of the time, is “story points.” There are exceptions—in my modern fantasy campaign Oak and Ash and Thorn, for example, one of the PCs had a Destiny, “confront the Dark Lady of Ulster,” so I knew that had to happen by the campaign’s end—but then there was a specific game mechanic for that in GURPS. Absent such a mechanic, my posture is normally “Let’s see what happens.”


The course of events in an RPG adventure is the result of human action but not human design?

1 Like

Back in the day I used to maintain that it is the acme of skill to contrive that the three criteria agree. The position was controversial, and became heretical as the Threefold developed into a disjunctive taxonomy of games rather than of adjudications.

1 Like

Yes, well, that is my very favorite of Hayek’s essays, so it’s not surprising if it’s gotten into my praxis.

Yes, well, I’m with you on this one. I think that the game rules give a representation of what is possible in the game world. But in determining what is actually possible in that world as a simulation, the game rules are unavoidably incomplete and can be wrong (in the sense that the outcome of game mechanics is unphysical, for the physics of the game world). In the ideal case this doesn’t happen, but human judgment is needed for nonideal cases.

And on the other hand, the game rules and the physics of the game world (and its historical dynamics) often don’t prescribe a single possible outcome. So then it becomes a question of choosing an interesting outcome out of those that are possible.

But on the gripping hand, I largely resist the idea that the fact that an outcome is uninteresting, or narratively unsatisfying, is a reason in itself to negate the game rules to attain a different outcome; in particular, if doing so would go against the nature of the game world, that is a reason to stick with the questionable outcome. There can be exceptions, if the game world itself is premised to run on narrativium, like the Discworld or perhaps the Buffyverse (the RPG, at least, builds narrative causality into the game mechanics). But, in general, the story is what the nature of the game world says it must be.

I can’t imagine running a game where only one of those is always decisive.

Whereas to me, it has always seemed obvious that the three are complementary. All three are ways to enhance the central experience of role-playing, of becoming someone else, and doing interesting things as them.

  • A story can be improved by game rules that don’t get in its way, and offer ways to handle its possibilities.
  • Those rules are best if the world is physically plausible, so that our personal experience of events is applicable to the game, rather than having to think about an artificial abstraction.
  • And that, in turn, lowers the burden on the story, and lets more of its narrative energy go into enjoyment and less into operating the mechanics.

It’s a virtuous spiral, if you can get it right. That is not easy, but it’s really good when you can manage it, and I’d rather try for the heights and fail occasionally than play quotidian games all the time.


Yes, but only at what I might call an intermediate level.

So the immediate tactical situation is not going to be “this room is in stasis until the door is booted down” (unless that’s literally what’s going on). But the slightly larger situation might well be “the bandits raid outlying farms, and the farmers try ineffectively to fight them off”, and that might be stable in the sense that sometimes the bandits win, sometimes they lose, but neither side makes much progress (until a confounding factor comes along).

And then at a higher level again, fates of nations, much more usually there is a process going on and the PCs can shove at it to try to change its course and speed; in my mind that’s more common than an equilibrium.

(I recently read Suzanne Palmer’s Finder, which has a five-sided political situation in the process of becoming unstable – but one of the destabilising factors is our hero, a spaceship repo man, even before he does anything. Because even if the ship weren’t important to one of the faction leaders’ power, that person would lose face by having it taken away from him, therefore he acts prematurely to try to consolidate his power before it happens.)

But getting back to the idea of pushing to reach particular story points, I often have some idea in advance of what a climactic scene might look like, but there will be many ways of getting there, and if it doesn’t happen at all then I’ll just put those bits away and use them on some later occasion. Which isn’t to say there’s no direction coming out of the GM’s mind, because many NPCs have their own ideas about what they want to happen, and they will be using their own resources to try to make sure that it does. But they don’t have privileged access to the machinery of the world any more than the PCs do.

I think there is such a thing as a rules-bound style of play, in which it is a normal and accepted thing that if a rule definitively states what happens it’s regarded as an abuse of GM power to make something else happen instead. Those of us who read the SJGames forums will have seen this worldview in people who are trying out GURPS having played, most often, D&D/Pathfinder, games with significant organised-play and tournament scenes, where one can see that consistency of ruling in the same adventure across multiple GMs could be highly valued. I contrast GURPS which is a “rule zero” game, a term with a muddled history but which seems to mean something like “have fun and ignore any rule that gets in the way”.

I have met players who feel that the rules intrinsically represent the reality of the game world (this seems to be well-correlated with players of rules-bound games) and therefore want to apply them to all situations, and that can certainly work against both story and a sense of reality as distinct from the rules.

When I play an explicitly story-focused game, this works against what I call “immersion”, the state of mind in which I’m running a simulation of my PC and reacting to the situation as they’d react. Systems that they don’t have access to in world, but I as a player do, shake that sense and cause me to step back from “being this person” to “writing this person”. (This includes things like “now narrate how that attempt failed”.) Oddly, although Genesys does a bit of this, it doesn’t throw me in the same way. As a GM I try to run multiple processes, some of which are immersive simulations of key NPCs, some of which are more general simulations (e.g. “how the base personnel are reacting to an intruder alert”), and some of which deal with the overall progress of the adventure.

Let me say that I think both of those are pathologies. On one hand, the rules will sometimes produce absurd consequences, such as the high-level D&D fighting man who can fall from the castle’s highest tower to the rock pavement below, stand up, dust himself off, and go back into the fight, because he just has that many hit points; I think the GM not only can but should set aside such rules and mechanics because they are unphysical (and indeed silly). On the other hand, there are players who will send their character into a fight at absurd odds, confident that the GM will not actually let any of their crowd of adversaries strike a lethal or crippling blow, because they are the hero/that would not be fun/they spent so much time writing up the character; I think that leads to equally absurd narrative sequences and the GM not only can but should allow such characters to die when the dice say so. “The English hang an admiral from time to time, to encourage the others,” as Voltaire said.

Rule zero implies trust in the GM to keep things enjoyable - as I recently said somewhere else, “enjoyable” mostly means “we succeed but just barely”, not “we roll over the opposition without breaking a sweat”.

Stipulated, but I’m not sure we’re talking about the same thing. Does your notion of “just barely” include the case where the opposition is defeated, but where one of the protagonists dies in the battle?

RPGs clearly fall into the type of drama or fiction that has an ensemble cast, and at least since the Whedon era we have had a model for such drama that includes deaths: Jenny, Tara, Anya, Book, Wash, Pietro . . .

There are direct payoffs from a protagonist’s death, in the form of drama. But to my mind there are indirect payoffs that may be greater: the audience does not have a sense of “they can’t die, they’re the hero,” which heightens suspense and drama; and when the audience are also the scriptwriters, as in an RPG, they have an incentive not to have the characters take unreasonable chances—or not to have them do so without a dramatically compelling reason. (The fact that they are going into what looks like physical danger does not by itself create drama. The drama comes from their having skin in the game.)

“Just barely” encompasses a lot of space between total victory and total defeat, much of which can be satisfying to play into. I think Ken Hite pointed out that for a series hero the tension is not “will he succeed” but “what will the cost be”.

(I very rarely kill PCs in my games but the possibility certainly exists.)

Conversation with @JGD elsewhere causes me to realise that I regard all people in games as people, with their own lives that matter to them beyond the one scene in which they appear in my game. So the classic chase scene jokes like destroying a fruit-cart or smashing a big pane of glass are causing, at best, a huge inconvenience in someone’s life; as a GM I regard doing that without a really good reason as a sign of Not The Good Guys, and most people I play with seem to feel the same.

1 Like

Hear, hear. I have been known to claim that everyone in the game is a PC, it’s just that a lot of the players aren’t present for the session, so as GM, I need to look after their characters.