What's the best way to write an adventure?

[ghost whistler]

Do i need a massive codex covering all possibilities, with stats for everyone and everything, plus maps of everyone and everything?

I've yet to see an rpg that explains how to do this - and do it right.

Or is there a right way at all?

[deaddmwalking]

The best way to write an adventure is to have an outline of significant events, possibly with a timeline. It is important to have some 'if-then' reactions laid out. Some significant encounters will need to be developed in advance with appropriate stats (whether pulled from online sources or generated).

You would have a sense of what will happen if the PCs do nothing (likely the nefarious plot will succeed). As the PCs interact with the plot elements, the opposition reacts accordingly to try to restore their plan and/or eliminate the PCs.

As a GM you're limited in the amount of time you have to prepare. There are shortcuts you can take to ensure the players continue to have fun. You may have a few 'side treks' on hand that are not necessarily related to the main plot(s) that can be used as necessary - for example, if the planned adventure finishes earlier than you thought. These can also be used to foreshadow future events.

You may want to consider always having 3 adventure hooks on hand - finding out which one the players will follow before fleshing them out fully. One adventure hook may be the one the PCs involve themselves with; the second may be handled by another group of NPCs or otherwise fail and the third may be successful.

If at level 1 you have 3 things and the PCs leave one undone, the 'advanced state' of that plot may be one of the available hooks for the next adventure (along with two new hooks).

As the players develop a reputation, you will have more opportunities to introduce hooks to them. As the players gain abilities, they may have the ability to resolve multiple hooks simultaneously.

If you do this right, you'll minimize the amount of extra work you have to do, but you'll always have material on hand to keep the game going, and the PCs will have the sense that they exist in a living world that truly responds to their actions. Incorporating their successes and failures into future plots helps tie them to the world - try to avoid 'stand alone' adventures that don't interact with anything else. If you use published adventures, think about how you can tie them together.

For example - let's say you have a published adventure with a goblin tribe being led by a werebear in a cave complex. You have an idea for a future adventure that involves duergar in the same region. It would be very appropriate to insert duergar into the goblin adventure, perhaps delivering an order of forged weapons to the goblins. Not only will the presence of duergar not be a surprise when they become the subject of an adventure, you'll have already established that they are involved in 'evil activities' in the area. While it is not critical to have the duergar plot fully developed at this stage, but having that in mind in advance you can create more verisimilitude. Having established elements of a plot in advance also gives you anchors for building more content.

[mlangsdorf]

What kind of adventure? A murder mystery is different than raiding a tomb is different than hexcrawl exploration.

You need to cover all possibilities than you don't think you can cover by improvising, but you only need to cover the possibilities that the PCs are likely to encounter in the next session or so. So if you're running a game based on Lord of the Rings, and the Fellowship is trying to get over the mountains, you need some detail on the mountains and some detail on the entrance to Moria, but you probably don't need details on the balrog and you certainly don't need details on Shelob's lair. You'll need those details later, perhaps, but not now.

The more you feel comfortable with improvising, the less detail you need. I'm starting a campaign and I've got the opening setpiece more or less detailed, but I need to fill in some details about the macguffin because I know my players will ask and the answers are possibly going to be important to the campaign. But I'm not sure what they'll do next, so I'll have to improvise, and thus I'm also writing a lot of general backstory and locations so I have a framework for that improvisation. But I'm not even writing up opponent stats, because I know I can just pull those from the relevant books on an as-needed basis and a codex of them would just waste my time.

[Blade]

Is your adventure for you and your group, or to publish somewhere?

[MisterDee]

I suggest this site.

More seriously: what you need depends on so many variables that it's impossible to give you a straight answer. I run a campaign where we play about once a month, and I can invest maybe 2 hours of prep per hour of playtime. So if I want torun a detailed map crawl, well, that means I have to run a simple protect-the-caravan scenario prior to that.

The metaplot is also fairly slow moving, so it's easy to have one-off adventures. If you're running a complex, tightly plotted narrative, you need to invest more on that than if you have to run "Random Heist of the Week #3"

Still, I try to have one big moment for the evening (a metaplot-advancing social scene, a fight with dynamic elements or special funky stuff, a cool puzzle) and then build around that with the remaining time I have.

[Whipstitch]

Having established elements of a plot in advance also gives you anchors for building more content.

Yeah, it's nice to at least have some loose mercantile and geographic setting details ready to go. That way if the players want to wander off somewhere to acquire or ask about something you're not stuck there saying "Uh, the locals of Nameless Village suspect those orcs were the infamous Orc Boys of Orc Town."

[tussock]

You need scenes, and locations for those scenes, not always strictly tied together.

You need ways for the PCs to discover the location of those scenes, and get to them, and interact with them, and understand them (in no particular order). Large dungeons make those steps rather obvious and easy.

You then need to add resolution of failure, how else can the PCs find and get to and interact with and understand various scenes if they fail the first way. And then how can they do all that if they fail the first two ways.

Which is like the three clue rule for letting people pick their own way through your scenes, but also three paths to each scene and multiple interaction options (including no interaction, thank you very much) and different things that can be understood in case they just don't get the others at all.

Then your scenes can respond to other scene outcomes and PC approaches in attitude or arrival, move to new locations, combine together or split apart, drop additional clues for other things PCs might have missed (or tie together the things they already found, as the case may be).

And then you cover what if the PCs switch sides or bind the demon or just teleport to the end and shortcut most of the actual adventure, how do you continue to offer paths into it to make some use of your work and still validate the PC's choices and abilities. Even if it is just to say that it's good they did it in order otherwise "bad things would've happened".

And finally some notes about how things change if the PCs start wasting time on whatever isn't this adventure, what happens if they are unexpectedly absent or dead or just something you can drop in an email if the group calls it off.

[Mord]

I recommend this series of essays from The Alexandrian.

[PhoneLobster]

Try this...
http://www.tgdmb.com/viewtopic.php?t=51662
... you seem to need the introductory guide there.

[Dogbert]

Depends, for you or for publishing?

If it's for you, just write your adventure's Key Points in the most bare-bones way possible and be ready to adapt as your players' action change things.

Regarding canned adventures, I'll always have more respect for a canned product that it's only a dungeon, monsters, treasure, traps, and no plot altogether than for self-agrandizing idiot plots (which is pretty much the case of ALL canned stories, all can be boiled down to an Idiot Plot in the best case, and a Second Order Idiot Plot in the worst) with artistic pretentions *cough*RiseOfTheRunelords*cough*... but then I'm not the target demo for canned adventures, so I'm not the person you want to ask.