@Yasmin Haq
I LOVE this idea! Here's a few thoughts I have on this story structure in regards to how it works and how to make it work as well as possible:
My first thought is how the prophecy is structured. Given this idea, I would structure it in one of three ways: 1.) Adaptive, 2.) Branching, or 3.) Predestined.
1.) The Adaptive method is the most free-form of these. Here, the prophecy is not "finalized" until very shortly before it comes into fruition. This allows the most freedom and story impact for the players, but is the most difficult to pull off for the GM. This is because the end point of the prophecy is perpetually mutable by the players' actions until it happens and requires the GM to be prepared for any number of possible outcomes. The end is impossible to predict, essentially. Using this, the players feel like they really are making the difference in the world, because they are!
However, this method comes with problems: The players may opt to change the prophecy to the point that the story is either ruined or taken so far off the rails that you can't pull things back on track without meta-gaming. This method is also has a higher probability of the players getting lost (or being uncertain of where to go) than the others. As GM, your improvisation skills will be tested using this method, so make sure you take diligent notes on player choices so that when the endgame of the prophecy approaches, you can reference the pertinent player choices that have happened and wrap it up in a way that makes sense.
An excellent way to run an Adaptive storyline using a prophecy is to have few different options for how the final event unfolds, and have what happens in the game steer the prophecy to the most appropriate end goal. Consider the following example:
For my Adaptive prophecy, I'll have four "final" outcomes. Let's say a massive uprising of the undead is soon to happen through an unknown party's influence. Any and all corpses will reanimate and kill all living beings around them. Here's four end options I can have where is the base scenario:
- The uprising happens, and all life is destroyed.
- The uprising happens, and all life is NOT destroyed.
- The uprising is prevented, and the party responsible is dealt with.
- The uprising is prevented, and party responsible is NOT dealt with (escapes, disappears, is never found out, etc.).
Working behind the scenes as GM, start with option #1 as the end point, and start the players with options to change that end point to any of the other three outcomes. Perhaps a rumor of a necromancer stockpiling bodies in a nearby tower has a town concerned. This necromancer, in turn, could be an agent of an evil deity that wants the uprising to happen. After that, the deity may turn its attention to the players specifically and make active efforts to hinder them. This is only one of MANY ways to start a prophecy-driven story. The beauty of the prophecy is that you can transcend time and linear storytelling in a believable way.
After each choice is made, adjust the revealed informational tidbit of the prophecy to reflect how the end point has changed, or not changed. Of course, the exact details are up to you, but I find the details much easier to fabricate with a general goal in mind. Do the players kill the necromancer before getting any information out of them? Do they talk it out and let the necromancer live? Is the necromancer even involved in bringing the prophecy to reality? How does that change the end point? Doing this after every major plot-related choice guarantees the end event will make sense and be adapted to what the players have done, without telegraphing what that result was going to be.
2.) The Branching method is the easiest method to use for a game that still allows the player's choices to affect the end point with with less grey area for them to get lost in. This is because it is slightly more restrictive in what the players can do to change the outcome. For this method you must have a set number of ending scenarios at the start. I'll use the same four end points from the first example.
Again, start with option #1 as the event that will happen if left to fate, and again, give the players a quest hook to change it (the necromancer). This time, however, when the players confront the necromancer, you as the GM have two or three certain options for how the encounter will turn out. Let's say you've mapped out the storyline and the options that come up for the story to progress are either A.) The necromancer says nothing and fights to the death, or B.) The necromancer talks and reveals information when beaten. Using your story map, each of these outcomes leads to a different scenario. Option A leads to the players communing with the evil deity directly on accident, and option B leads to them finding out about a meeting of undead sympathizers happening in three days. The players never need to know that only two outcomes were possible, because you don't have to tell them!
The problem with this method is steering the story toward set number of options when things that are unexpected happen, and they WILL happen. What of the rogue sneaks in and manages to get into the necromancer's lair undetected? You don't want to completely invalidate the players efforts too often! So, instead the necromancer talking, the rogue finds a journal in his desk that reveals the same information. It's easier to improvise when you know what the outcomes are going be.
When mapping out a Branching storyline, you start with one point (the uprising and total death of all life) and end with just a few (the four options stated above, or in most systems, just two options where option is total success and option two is total failure). After each choice is made, reveal in the prophecy a small scene from the scenario that will cause the next branch in the path to happen. In this case, the players get the info out of the necromancer, and subsequently have a vision of a massive ritual where a dragon is being revived to life. When they arrive at that point, they must have choices to make, which will determine their next vision. The trick here is to reveal enough information about the next plot point to keep the players interested without eliminating their ability to make choices on how to handle it. The end points are usually pretty evident early on ion this model, so it's up to you and your map of branches to make the journey interesting!
3.) The Predestined method is the most straightforward way to tell a story. Here, there is only one end point. The first one. Each prophetic vision granted to the players is predetermined after each "choice" is made. Except it's not a choice, it's the illusion of choice. The most important thing to remember as a GM for this method to work is to NEVER make it evident to the players that the end is not changeable. Give them the quest hooks, let them accomplish tasks and change things in the world, but never change the end goal. Some unseen force or a unfortunate coincidence seems to oust the players' attempts to steer fate.
This method also has problems to be dealt with. First and foremost, keeping the balance between player accomplishment and failure. The players should never feel, much less know, that their quest is fruitless and doomed from the start. Let them thwart evil powers and seem to win, only to have the visions continue and a new danger show up greater than the last. The example I use is extreme, and having a final campaign plot point fall in the camp of "total failure" can leave the players a bit empty at the end (my players very much liked it, but you should absolutely make your end point in this method one that the players will appreciate if not enjoy. Happily ever after is a perfectly acceptable and often wonderful result).
These are only three of several ways to make this work, tweak any of them to suit you, or even use more than one at different points in your story! Maybe some plot point are totally changeable while others are absolute! In regards to providing context through prophecy: give less in the Predestined method, give more in the Adaptive method, and treat the Branching as a middle ground giving context only when needed to make the next choice apparent.
I hope this helps. If you have more questions, please ask! I'm happy to offer my opinions on anything that any time, and I'm glad to clarify on any points I made above. I can really ramble when I want to…