I've cut the specifics out, but otherwise it is as I said it the first time.
I caution you against thinking about adventures in terms of story. There is a story....what happened before the PCs became involved....and there will be a story after PC involvement is done and the players are reliving the events, but I do not believe that the GM can or should know what is going to happen at each point along the way.
I would like to talk a little bit about layers and trigger events. Also about overt and covert threads.
What most people do when they start working on an adventure is the covert thread...what is really happening that the PCs must uncover in order to bring events to a satisfying conclusion. Most adventures need a layer of overt threads...things that happen out in the open, the ways that the players (and locals) first view the events and places in the adventure. If you think about an adventure as a mystery, the covert thread is what really happened. The overt threads are all of the other side issues, the alibis, the red herrings, and the daily life that conceals the covert thread from the detective until the mystery's climax.
Some rules of thumb:
- For every part of the covert thread that the characters must uncover, there should be at least six clues.
- For any part of the covert thread that it would be cool if the characters uncovered it, there should be at least three clues.
- For every location you want the PCs to go to in order to discover these clues, there should be overt reasons for them to go there. Note that NPCs saying not to go there, even if there is a hoard of gold lost on those old burial grounds, is almost certain to make any PC walk into a death trap, let alone a creepy swamp.
As an example of what I mean here, consider ADVENTURE The characters are going in to GOAL. That's an overt reason for action. They need GIZMO to get in the LOCATION. That's another overt reason for action. Along the way, they are given many clues about the covert thread (the nature of the CREATURES in this area) which should lead them to a second covert thread (maybe we shouldn't DO SOMETHING THEY WERE PROBABLY PLANNING ON DOING). The presence of various treasures and things to manipulate give the players more overt reasons to explore beyond a strict linear progression to the pool.
As the PCs examine the various clues, their understanding of the adventure changes. Some of what was covert becomes overt. This continues throughout the adventure. As a result, the players' understanding of the adventure (and adventure location) develops a layered depth created through interpreting and re-interpreting what they encounter and whatever events occur. We all experience this in film or fiction, and we all know how shallow a movie or novel is that fails to cause us to reinterpret what has gone before. It is the difference between Dark Knight and Batman Forever.
There is nothing like peeling back those layers, as a player, and suddenly seeing the whole thing clearly. It is a great feeling, a moment of sheer exhilaration. Of course, it has to be the players actually doing the work, or it is meaningless. The GM telling you Bert is Evil is nothing like putting the clues together and realizing that, very much in contrast to what you've been thinking all this time, Bert is actually the evil mastermind who is controlling the entire street.
A note on clues: Different people can be pressured to play the villain's game in different ways. One might be promised gold, and his greed makes him do vile things. Another might have a shameful secret he is afraid will be exposed. Yet another might simply be trying to prevent the villain from targeting his baby sister. Various NPCs, being made to do the villain's bidding through various means, offer more clues than do the same NPCs if they are all doing it for gold. Different motives give rise to different behaviours, which in turn give rise to different chinks in the armour of the mystery, and more ways for the players to crack the shell open. You want to provide as much context as you can, without overtly spilling the beans, because you want the beans to be spilled. And it should not matter if they are spilled early or late.
That these different motives also raise the spectre of not all the "bad guys" being bad; that "fighting them" in some cases means (or can mean) "rescuing them" is all the better....because, if nothing else, it allows the players to have moments where they must make ethical decisions. It also means that a rescued "enemy" can become an ally, and can impart information (context) to the players.
Instead of imagining a climax where the PCs figure out what is going on, try to imagine the climax where the players learn the covert thread earlier, at the time, or never, and it still works. It is better to offer clues at the end, and give the players the opportunity to either figure it out or not, than it is to spill the beans.
Never knowing is better than knowing because the GM told you.
Knowing because you figured it out yourself is best of all.
Trigger events are things that happen after a particular condition is met. I.e., after the players ask at the Rusty Fox about the creepy old lighthouse keeper, they are attacked by thugs dressed like ghouls. Trigger events, when at all possible, should follow as a direct consequence of whatever triggered them, so that the timing is a clue to the covert thread. Even the dimmest of players will eventually realize that the priest is a spy if, after every time they go to him for help, the Temple of Chaos seems to know what their plans are.
Layering requires paths to explore that are not the main thread. Each of these paths, in some way, points back toward the major issues and what is moving below the surface. Both layers and trigger events are used to create the impression of things moving below the surface, and to give the players clues to finally peer below the surface and discover just what is going on.
This relates rather directly to a recent blog post.
Anyway, I am beginning to blather here.
Best of luck with your designs.