July 29, 2013

Rules vs Setting

Jesse Burcar asked me this over Twitter:

Naturally, I said I couldn’t fit the response into 140 characters, but the general thrust is Mu. They’re not distinct.”

I’m now going to write a hell of a lot of words that probably don’t answer the question. Sorry. Shit happens.

Crunch vs. Fluff

or, The False Dichotomy

This isn’t a direct response to Jesse’s question. Instead, it’s a whole bunch of related thoughts. Recently I’ve seen a resurgence in arguments along the lines of:

Crunch (rules, systems) is hard to write. You only write fluff (setting), so you’re not as good a designer. You’re also probably a failed novelist. Because I’ve written some terrible houserules for Truncheons & Flagons,1 I’m automatically a better game designer than you are even though I’m actually a dickhead with self-esteem issues desperately trying to achieve some form of validity for my life by denigrating professional game designers.2

Yeah, it’s going to be one of those posts.

A game is a blob of information. Some of that information pertains to describing the type of story the game is supposed to create, even if it’s a strict world-sim type thing. Some of it describes the systems used to reflect that story. Both parts are just as necessary, and both parts are just as hard to write.

Elements of a Roleplaying Game

The real dichotomy in role-playing games is between setting and fiction.

Fiction can describe a setting. For the vast majority of roleplaying games, the only thing needed is fiction describing a setting and characters. They’re carried out on blogs and livejournals and forums. Most people who play tabletop roleplaying games don’t interact with these games, which is a pity because they’re in a serious majority.

Most tabletop RPGs have three distinct elements:

  • Fiction, which I’m going to break apart into what happens in the game” and descriptive text showing how the game should run”. Many gamers assume everything that isn’t rules revolving around dice[^3] is fiction. They’re wrong, mind, but that’s a popular opinion. The latter is used as an example of setting. The former needs both setting and rules.

  • Setting involves the presentation of the milieu in which the fiction happens. The fiction needs the setting. Setting involves a combination of broadly-written areas that the writer designs—just as much as they might design rules—in order to a) evoke and b) provide hooks for creating the fiction through play. Setting may include scraps of the second meaning of fiction as part of the evocation.

  • Rules, which includes specific roll these dice” and lose 10 hitpoints”, but also what White Wolf games call theme” and mood”—which should really be replaced with leitmotif” but that’s a different argument. The former is easy for anyone with a grasp of probability, decision theory, and game theory. The latter is usually conflated as part of the setting.

Which is Harder?

None of them.

Writing a setting is hard. You need to know what genre you’re wanting to emulate, but you can’t just sit down and write fiction and hope that it’ll evoke the setting enough that you can toss in a couple of story hooks and be done. Sometimes it works. Usually, it sucks.[^4]

Writing fiction to support a game and provide examples of what the game should look like in play is hard. Let’s face it, if it were easy you’d not be writing fucking RPGs. You could be making enough as a novelist that you wouldn’t need a day-job.[^5]

Writing rules is hard, especially if you don’t know probability, decision theory, or game theory.[^6] Even if you do, the rules have to by their very processes create the outcome that you desire—the fiction that you’re aiming to emulate.

You’re No Help

Probably not. But I have answered the question.

How to Make a Game

Now, on to a related question: How does one make a game?

If you’ve got some ideas for systems, play with them. What’re the odds? The expected values? On strict analysis, is there a single optimal strategy that a rational person would take? How do the strategies change when other people get involved?

Once you’re happy with your results, write them up somewhere—along with the results of your analysis—and stick it in a drawer somewhere. You are not ready to write a game.

A game starts with an idea for either a setting or the fiction you want to emulate. Once you’ve got that, flesh it out a bit. If you know what kind of fiction you want to emulate, create a setting that emphasises that kind of fiction. If you’ve got a setting, write about it—not about ancient history or powerful NPCs, but about bits that you envisage the characters in your game interacting with.


One of the worst things you can do is write a lot of information about your setting that won’t ever come up in-game. The most egregious example I can think of is the Iron Kingdoms RPG—the modern one, not the D20 one. The book opens with an introduction then page after page of ancient history that doesn’t help me as a player one bit.

I came into the game cold, wanting to play in the setting. I don’t want to read twenty pages on dead gods and who stabbed who fifteen thousand years ago. I want a precis of what’s where, some recent history that can throw up story hooks, and current interactions between individuals and nations that can throw up story hooks.

Iron Kingdoms feels like a game that was meant to be read, not played. It’s a supplement for the wargame showing off a bunch of worldbuilding[^7] to people who already know the setting and want more. It’s a terrible way to write an RPG for people who aren’t already invested in your setting.


Anyway. You need to develop your idea—the fiction and setting in tandem—until you’re able to present both in a way that gets everything about your game across to readers without you being present. This takes a lot of work. Setting development takes time, and as mentioned is a hell of a lot harder than just writing stories. Fiction development is hard because literary analysis is hard, and if you can’t be fucked doing literary analysis you’re never going to figure out what makes your chosen fiction tick. You need to create hooks and elements that communicate what your game is about.

This doesn’t look like a lot of hard work, but it is. It looks easy because you don’t have many specific questions that you can ask and answer. Knowing literary analysis and criticism helps, but it’s like writing an essay on the evolution of dada and surrealism into situationism and its impact upon the modern world. One question (or rather, one title), but a shitload of work to answer.

Once you’ve got your idea, then attach your system. Now, you’ve got a choice. Do you want a system that emulates the fiction, or one where the players help create the fiction?

Some mouthbreathing idiots want to claim that one type or the other isn’t a roleplaying game”. Ignore them, because they’re so obviously wrong that engaging them will make you stupider, and you don’t design a game by getting into fights on the internet.

If it’s the former, think about a few things: How focused is the game? If it’s got a tight focus, think about how to represent that focus in the rules—mission creation in BLACK SEVEN is a good example of a hard system that reflects the game’s tight focus. Marvel Heroic Roleplaying uses Action Scenes and Transition Scenes as a soft system to create a similar focus.

Does an existing system do what you want to do? If so, how should you change it so it better represents the fiction you want to create? This can be just as hard as building a new system, because you need to analyse how your changes impact the original system in case of emergent weirdness.

How often do you want characters to succeed? Do you need a randomiser, or would a blind-bid feel more appropriate? Do you need degrees of success? Is a flat percentage chance good (one die), or would a bell curve be more appropriate (multiple dice)? How about a binomial distribution (dice pool)? Do you want two axes of results (cards)? Do you want to model characters as a whole, or only as they engage with the story—the latter is good for tight-focused games. How does the system play if you don’t approach it already knowing the kind of fiction you want to create?[^8]

That’s a lot of questions. You need to answer them. This is known as doing the core maths”.[^9]

Examples of this kind of design among my games: Æternal Legends, BLACK SEVEN, the upcoming Rescue Squad.

If you want a system for creating the fiction, think about how you want that to work. Again, do you need randomisers (and all the other questions above)? How do they drive the fiction? Do you want players to each have a character, or do you create a pool of characters and players pick and choose every scene? What do you need to know about that character? Do you represent that mechanically? How? Do you need a GM? What rules does the GM operate under? What restrictions apply to the GM? And to the players?

Again, that’s a lot of questions and you need to answer them.

Examples of this kind of design among my games: Beyond, Tales of the Space Marines

This looks like a lot of hard work, and it is. But it looks harder because it’s made up of concrete questions and pieces that you can test via the tools of probability, decision theory, and game theory. It’s no harder than setting creation, it’s just more obvious in its difficulty.

Game design involves hard work in all areas. If you don’t want to do the hard work, you might end up with a game, but it’ll suck. Plenty of games suck after doing the hard work. Learn from them and move on.

[^3]Or cards. Or blind-bid point-distribution. Or the expenditure of resources. Whatever. [^4]: Oh, pick your own bloody example. Cthulhutech is low-hanging fruit even if it wasn’t written by idiots who can’t help but jam sexual violence into every goddamn supplement. Pretty much any game decried as hasn’t learned the lessons of early White Wolf” fits the bill. [^5]: Hi, Chuck. [^6]: Though this level of ignorance would get you a spot on the D&D Next design team. [^7]: The difference between worldbuilding and masturbation is the number of hands involved. And the amount of shame you feel at exhibiting the results to the public. [^8]: Not answering that is why D&D3 and its inheritors in the dungeon fantasy” genre are all awful. [^9]: Plural. Only dead people say mathematic”.

  1. More likely the currently-popular republication of an earlier version of that game, which isn’t so much ripe for abuse” as pre-broken on a fundamental level.”

  2. I may be paraphrasing.

gaming prescriptivist design of subnarrative metalevels self-publishing for stoats

Previous post
Deluxe W20 Changing Breeds Kickstarter The Deluxe W20 Changing Breeds Kickstarter is GO! This is the first Werewolf book I’ve developed, and I’m psyched to see it out on Kickstarter!
Next post
Pay What You Want and Me So I’ve been rocking Pay What You Want1 on DriveThruRPG since June, when the option became available. What have I learned in that time?2