GM Prep is Many Points on a Spectrum
You can't write novels or have sandboxes... just everything in between.
Prep exists on a spectrum whose poles are Player Agency and Narrative Robustness.
Truly open ended sandboxes are a Holy Grail of the RPG hobby. The sandbox becomes the ultimate expression of player agency. They can make any choice, at any point, and go off in whatever direction they choose. The game master cannot prep too far in advance, because the fractal nature of player agency means that the probability of players following any thread of prep is effectively zero.
That’s an apt metaphor, because nobody has ever found the Holy Grail. To be utterly pedantic, imagine the most open ended sandbox you possibly can. There’s still a game master, right? There’s an inherent limitation of agency when somebody else is arbitrating the results of your actions, or who you interact with, or what the world looks like where you stand. I can predict the follow-up point: “What about games without GMs, where all the players collaborate to create the world?” Well, one player’s agency is obstructed or overridden by another’s. If John says the town has a tavern that sells specialty ales, James can’t say the town is full of teetotalers who can’t abide alcohol. Only in solo games do you have complete agency.
On the other end of the spectrum is the Game Master who preps everything. They are enthusiastic and passionate and have a ton of free time, which they use to imagine and write out every scenario from the campaign’s beginning to its conclusion. The themes are omnipresent, striking, and beautiful. The characters, including the NPCs, have meaningful arcs, growth, and discovery. The world changes to reflect the ongoing story. Before session zero, they have… written a novel. We hobbyists know this pitfall well, and all modern advice points game masters towards avoiding it. We have fully eliminated player agency in exchange for narrative robustness and coherency. This story survives contact with the players, but will never see the light of day, because all of the players will leave the table far before it reaches its apex.
These are contrived examples, intentionally, to demonstrate that the end points of this spectrum are either unachievable or unacceptable. So how do we navigate this to find our personal sweet spots?
One Session at a Time
On the Runehammer podcast, Hankerin Ferinale/Ingrid Burnall/Brandish Gilhelm (pick which name you like best at this moment) advocates, generally speaking, for prepping one session at a time. This advice is well-intended and effective. It respects your time as the game master, it avoids the stress of having to create large blocks of content up front, and it circumvents the problem of “wasted” prep (a wonderful problem to have).
To place this philosophy somewhere on the spectrum, I find this advice to fall closer to the “open ended sandbox” side of the spectrum than the “narrative robustness” side. That’s the style of game that I like to play, which is convenient!
But there’s a bit of nuance here that deserves mentioning. Not all prep needs to fall on the same place in this spectrum. If I’m preparing NPCs, dungeon rooms, and treasure, this one session at a time strategy fits like a glove. These are all concrete things that won’t be useful if I prepare them too far ahead of time.
There are other things, though, that I prepare much more in advance. Things like factions and their goals and resources should look more into the future than a single session, and most faction subsystems will do that. The Worlds Without Number faction system is a seriously crunchy, involved subgame that requires the game master to have considered their sandbox well in advance of meeting with the players, and to have placed assets, wealth, armies, and strategies on the board with consequences and considerations that reach far beyond just the next session.
What Falls Where?
So if we have found one example of prep that deviates from our standard philosophy, what others can we find? It critically depends on the type of game you are running. Some games will require you to flesh out much more in advance than others.
Mysteries. If you’re running a mystery game, a la Call of Cthulhu or Gumshoe, you need to know the facts of the mystery from beginning to end. This falls very far towards narrative robustness on the spectrum, for the obvious reasons: a mystery will quickly fall flat and disappoint if you’re making it up as you go along.
Hexes. If you’re running a hex crawl campaign, all about exploration of a fantastic world, there’s no avoiding populating your hexes ahead of time. You might not need to populate every hex on the map right away (God bless you, Dolmenwood. You’re doing the One True God’s work). But you do have to have enough on your hexmap ahead of time that if players travel through some of those hexes en route to a new town or adventure, they have something to experience along the way. That’s a cornerstone of the hexcrawl, and is fairly non-negotiable.
Locations. If the scale of your campaign is limited to the small town they start in and the surrounding wilderness, you won’t need to prep the Outer Planes of your world’s cosmology. But if your group expects to escalate well beyond familiar havens and into extradimensional worlds, there’s simply no avoiding preparing the structure and concepts of those worlds ahead of time. In order to signpost the possibilities and give your players meaningful ways to decide to pursue those roads, you have to have an idea of where the road goes. They can’t decide to head to the Red Wastes if you haven’t told them that it exists and given them tantalizing reasons to risk it.
Consequences of Inaction. This is a juicy example, one that generally overlaps with the factions example from before. If you introduce a situation in your game which, if left unattended by the players, will develop into other situations, you must know what the consequences of inaction are so that you can show the players, in the world and its changes, what those consequences are. These consequences cannot always be determined in the scope of a single session. How punishing would it be to say, “Here’s the situation, and if you don’t solve it in the next four hours of gameplay, your home town will be destroyed.” It’s almost an ethical requirement to give your players a little more leeway than that, when the stakes are high.
These are just a few examples, with generous assistance from the NSR Discord community. I’m sure there are others that apply to your game, and I’m certain not all of the ones I’ve talked about will apply to your game. The act of considering them is how you determine your personal approach to game mastering and prep.
Prep is a Spectrum, So Why Pick One Point?
Going forward, I am going to carefully consider where the parts of my own prep fall on this spectrum. I imagine it will help inform the process of my prep - when I do each part, how concrete or abstract I leave them, how important they are to me, and how they affect my table. Once I see these things in practice, and observe what changes they make to my campaign retrospectively, I can adjust and reprioritize.
I hope this idea helps you manage the weight of your preparation. If you have thoughts on where your prep lies, I’d love to read about it in the comments.
There is a wonderful sequence in the Aardman Animation film ‘The Wrong Trousers’ that features a homage to classic mad-cap chases from The Keystone Cops, Harold Lloyd and other greats. In it, the dog, Gromit, is holding on for dear life to the model steam engine of a train set that is travelling preposterously fast. In order to prevent disaster, he has to keep laying the track ahead of the engine, one piece at a time, at furious speed. This is how I describe my GM prep for sessions of play. Although for some investigation games and campaigns with definite story arcs, I might have a meta-plot (in the case of my metaphor, this is ‘the chase’) I don’t permit myself to prep for anything other than the next session - and that often only in outline (continuing the metaphor, I lay one piece of the track at a time, often in a great hurry).
Of course, this means that one has to be willing to throw prep away as soon as the shape of the session becomes clear. In the most recent session for my current Cthulhu Dark mini-campaign, I prepped a bunch of locations and NPCs that could be used to advance the investigation, knowing full well that I would almost certainly never use 50% to 80% of the notes I’d made. In the end, it was closer to 80% that was junked. But that’s alright, because they were only a few lines of description and what could be found out in any given encounter. Better yet, as the characters do what they do, they almost forge the next piece of the track for me because I take their speculations and use them to inform what might happen going forward.