Someone recently opined within my hearing that a particular RPG was "an Atlas of RPGs", meaning that the GM carried the entire game. Good game? Because GM. Bad game? Because GM. And I though, "I can see it that way, it's lots of work running that game". After a few days I've formed a rough proto-theory.

I though at first of a range, from games that were lots of work to GM to those that were little work to GM. But that's a very simplistic model that leaves out something important. The idea has shaken out to be a description of how much effort a GM has to put in to the mechanics of running a game. Don't care about the setting here, nor the session prep like making maps or picking out premade NPCs to use. Not talking about how fast or slow or complicated the game rules are, it can be related but it may well be more correlation than causation. This is just about how much work and effort the mechanics of the game make the GM do in order to have a functioning system at the table during the session. The scope of the game, what it tries or is supposed to cover, does matter. But reasonable people can reasonable disagree on exact boundaries of a game's scope while agreeing that, for example, a wild west/cowboy themed game should probably be able to cover cattle stampedes and horseback chase scenes without the GM having to completely make them up from scratch.

The idea is that the amount of effort it takes a GM to mechanically run the game can be expressed as a triangle. Three corners, in one corner is Atlas, another is Zeus, and in the third is Hephaestus. The Atlas corner represents RPGs where the GM has to do a ton of work during game to keep things running smoothly. The Zeus corner represents RPGs where the GM doesn't have to do anything to make the rules work smoothly. Hephaestus though, represents an RPG where there's a ton of up front work picking out which bits to use and hammering them into shape, and then the game runs smoothly and easily. Any particular game, or serious portion of a game within the game's scope, will fall somewhere in the triangle representing how much effort the GM will need to put into manipulating the game mechanics to produce a decently fun game. This isn't about the effort to plod through a rule itself, long division to get the dice result or a table lookup aren't part of this they're the rule in action and being applied. This is about the amount of effort put on the GM, by the game mechanics, before and after actually getting to roll a die and get a result.

With more precision and examples:

For a Hephaestus leaning RPG the GM does a bunch of prep work to make the game mechanics play smoothly. But once all the work is done the game is easy to run and the GM doesn't need to adjust the rules, look obscure optional rules, or suddenly change the way something works because it's too strong/weak. I did this once for a D&D 3.5 game; defined what would and wouldn't be used for class-spell-item-etc., prepped a bunch of notes & NPCs, laid out house rules, wrote down specific rulings for some rules questions. Then when it was time to play... everything worked. I only once had to puzzle over what some spell or feat should do, never worried about whether or not to call for a skill check, didn't get caught by any nasty conundrums of if something was or wasn't allowable. Ran that campaign for over a year from 3rd to at 14th level. The system mechanics were never any work for me after all the prep and I could focus on the adventure. Gurps and many supers systems work this way a lot, when done properly. The GM makes a bunch of decisions and adjustments up front about what books & rules & options to use. Then once you're in session there's very little thinking about the mechanics of the rules, you just roll the dice, get the result, and go to the next thing.

For a Zeus type game the mechanics & rules are either sufficiently comprehensive or simple and robust enough that there's no "how does this rule?" type of work to do. Lightweight games like Risus lean this way. For others the rules might not be simple, but they're comprehensive enough within the scope of what the game does that the GM never has to worry about which rule to apply or how to use it. There's a rule, general or specific or in between, you make the rolls (or spend tokens or compare values for diceless games or whatever), and you get a nice result that doesn't make people go "wait wait that can't be right" and suddenly have an overwhelming urge to house-rule it mid-session. To be sure I don't think there's going to be any really rule heavy games that are perfectly 100% out in this corner of the triangle, but I could be wrong. Does anyone actually read my posts. Personally, I find the Paranoia Anniversary edition to lean pretty far out this way, for me at least. I just make up some lunatic mission to send the PCs on, think up some R&D tech they get to play with and die from, decide on any secret society orders to add extra backstabbing, random gen up some NPCs & rooms & corridors, and its a go. I never have to check that the mission length/type or R&D toys break anything, don't worry about probabilities or improbabilities or "legal" NPCs. I just build the setting stuff, have people roll dice, and I get to focus on keeping the game funny and moving quickly along.

For an Atlas type RPG the GM has to engage in adjusting, changing, skipping, and making critical decisions about the mechanics all the time in order to keep the game working. Any time players get up to something the GM has to think through a whole list of things; "do I want to use the game mechanics for this?" "which rule or optional rules should I use this time?" "how do I use the rule this time?" "am I rolling too often or not enough?" "do I need to change the rule or use a different optional rule to fit the circumstances?" "will this rule give me the right result or an unacceptable outcome?" "player rolled X and the rule answer is Y but how do I use that?" "should I house-rule or change something about this rule?". That's all work that the GM is doing just to keep the game mechanics functioning smoothly. Not saying that the game automatically breaks if the GM makes a mistake, or that this is "bad" or anything. It's just about the GM having to keep making rules decisions all the time that are going beyond "what do I tell the player to roll?". For my group D&D 4e was like this. Not combat, that was firmly in Zeus territory (just slow and eventually annoying). But pretty much everything else outside of combat had the GM hemming and hawing over whether to use the skill challenges, how to deal with rituals, the "oh god not again" for mounted combat, if PC powers could be used for stuff, and how to get us more horses because the last batch exploded into chunky salsa from turn one AoE splash damage again*.

To be sure, there's no "right answer" here. This is just an idea for describing the amount and type of work a GM has to do to keep a game system working during a session. Some people may prefer one way or another, people who don't GM don't have to worry at all. Many games are somewhere probably closer to the middle than anywhere else, and parts of games may be in one corner or another. This isn't a problem or a description of a problem, it's an observation.

* OK, that last one isn't really a rules problem because we kept finding it funnier and funnier. Plus it really was the rules working as intended, damage scaled and horses didn't. But some people consider it an issue and the DM was going nuts for a while because there was a mission timer and we had to cross lots of plains/prairie.