on a golden rule for writing rpg scenarios
Writing for RPGs as a GM is time-consuming and exhausting, and I was thinking of a "golden rule" for this which would let me not waste effort on pointless things.
The obvious one is:
Don't waste time writing things that the players may never encounter.
This stands to reason right? If you spend ages on an encounter or scene that the players may never see, that's wasting your time. Try to avoid that.
The idea is not to turn you into a scenario-making machine, but to make sure that you concentrate on the bits you will need in play when preparing, rather than spending weeks writing background then having to wing it in the actual game.
That would make a bit of a short post, so here are some more details.
You should know what the start and end points are, and what the PCs are doing there at all
Call a "scenario" an overall period of play that has a defined beginning and end - a start point where players find out what the goal is, an end point where they've achieved the goal - and a reason why they should be goal-focussed. It helps to have one.
"This starts when the PCs are arrested for (some crime) and find out that they will be freed if they can find the Lost Necklace Of Gorg. It ends when they return with the LNOG and are pardoned. They need to do this because otherwise their prison anklets will summon demons to eat their legs."
The start and the end points should be known to the players, even vaguely. You need to tell players what the goal is somehow, or they'll be confused and annoyed, not knowing what they're doing there. It may need to be abstracted a bit - for example, if the end point is that they return with "an appropriately valuable artefact", the LNOG being the most immediately available one may be part of the scenario, but they should at least know that they need to come back with a valuable thing.
(Players can be good at being goal-focused, even when realistically their characters might see other options, because they meta-game and realise that you have a scenario here and it's not going to be fun for them if they just ignore it and do something else. This is handy but don't exploit it or rely on it.)
You should know what is required within the scenario to get to the end
You can work backwards from the end conditions, or forwards, or however you like, but you should know what things the PCs will definitely need to do to get between the start and the end.
The Lost Necklace Of Gorg is in the shrine room of a dungeon (predictably). The PCs need to enter the dungeon and get to the shrine room, then return to the citadel and make their case to the magistrate.
Between the start and the end is where events take place, but you can break that down to individual points which are required, and where you can put effort in. They're definitely going to need to enter the shrine room and deal with the guardians in there.
(They need to be able to know about all of these required places somehow, or, again, they'll be confused and annoyed. It doesn't have to be obvious or immediate but there needs to be some way for them to find out.)
Apart from those required areas, the rest of the dungeon is optional, no matter how big it is is.
Anything which isn't strictly necessary is optional and you should not spend (too much) time on it
It may be very tempting to spend ages on the amusing personality of the dwarf shopkeeper in the village outside of the dungeon, but don't (unless it's fun, and in this case be prepared to re-use your effort somewhere else). Even if going to his shop could provide the PCs with some great advantage, it falls foul of the golden rule - if PCs are not required to go there they probably won't. Just have an idea of what's there and few adjectives ready so you know what the situation will be like and can improvise if it comes down to it.
Rooms in the dungeon which PCs may or may not go down are similarly optional. Give them a few adjectives. Don't spend time on them. Try not to make this obvious in play, and consider just removing them if they're simply filler.
If there's a part with lots of options, all of which might fulfil a condition to get to the end and none of which are necessary, don't spend time on those either. If the PCs need to have a particular token to enter the shrine room which can be found on multiple enemies, don't spend much time creating those enemies - either that or rethink the premise, because the players likely won't appreciate the increased number of options anyway.
Be careful when making random lists
Random lists are great for when you need to improvise or you have no idea what's going to happen, but not required for scenes that are pre-prepared. They also fail the golden rule unless you know that the number of times you'll roll on them is the number of times you will need to roll.
Don't make a maze which the PCs might need to navigate, then spend time creating a 1d20 table of possible things they might meet inside it, if they'll only ever meet one or two of them. Just decide beforehand what they meet.
Don't write a long detailed list of things that might be found while searching if that's only going to happen a couple of times, if at all. A short list of vague descriptions is fine.
- If you have an existing list that someone else has put effort into, why not use that, sure. It's your effort that counts here. There are lots of them in supplements or on the internet.
- A list that you can re-use elsewhere. A random junk list that works in multiple places can be really useful, for instance - I often wish I had one when players say "so what's on the shelves in this room then?" Shelves seem to be quite empty in my games. Similarly, random magic item lists let you just say "if they look under the bed there's a magic item in a box" and then get on with the rest of it.
So the dumb scenario I gave examples for might look like:
The blue nodes are the ones you can spend time on. The green ones, you can skimp on. Yes, even the magic sword chasm, because it's just something the PCs might encounter. If you think the magic sword chasm is a really cool idea, make it required.
There are no guarantees
Of course you can never guarantee that anything will definitely happen. Players can always find a way to bypass anything if they have any real control over what they do, and if they don't, you have a scenario on rails and they might as well not be there.
One might conclude that one shouldn't put too much detail into any situation, and that has a note of truth to it for me, but probably something for a different post.