For the past couple of weeks I have been noodling around a how-to process for creating a murder mystery adventure for D&D. I'm not going to bother with any great intro here: you want to create a murder mystery adventure, so here's a step by step guide:
Step One: Review PC Abilities
The very first thing to do is to review the PC abilities, or potential abilities. Ideally, you're going to want these to be useful but not immediately conclusive - if the PCs cast detect thoughts then they need to get something from it, but you don't really want it to immediately shine a "He Did It!" light on the villain.
This step is obviously easier when dealing with a specific set of PCs than a general set, but the process is much the same: go read up on the various spells and powers that might be accessed at the appropriate level, with particular focus on divination and enchantment spells and powers. This should be done first, as it will inform the various decisions made later.
Step Two: Start at the End
Now that you've grounded yourself in what the PCs are likely to do, you can build the mystery proper. And like with any good riddle, the thing to do is decide on the solution first and work backwards. So, who was your victim, who was your villain, and what happened when the latter killed the former?
There are three basic parts to this: means, motive, and opportunity:
The means are the way the villain killed his victim: how did he do it?
The motive is the reason: why did he do it? (Poirot is always sceptical of motive, and it is true that there's always something. However, when designing the adventure you probably need to know why.)
The opportunity is about timing: what allowed the villain to do it now? Or, when did he do it?
Once you've got those three, you can now work backwards through your investigation to get to the point where the PCs come in.
Step Three: Consider the "Dungeon"
Imagine for a moment a dungeon adventure: the PCs come into the entrance room, they then proceed from that room to the next, until they reach the BBEG and the adventure's end. A good dungeon will have multiple paths through, with options to shift from one path to another. A better dungeon will include various loops and twists. Ultimately, once you straighten out the map, you're likely to end up with something like this:
Now, instead of a dungeon, imagine that each room is a segment of the investigation (the PCs examne the body, the PCs question the doctor, etc), and each link between rooms is a clue pointing to another room.
And so you now have a "dungeon" where the PCs come in and they have various options for the investigation. Each of these then points to other avenues of investigation, and so on and so on, until the trail leads to the BBEG. And, ideally, you have multiple paths through the investigation, loops and twists, and so on.
The big difference here is that the PCs are considerably less constrained than in a dungeon, and may easily jump from one "room" to a completely unconnected one. But that's fine - the map here is used to verify that they can solve the thing, not to force a particular way of doing it.
Step Four: Build the Links
Now that you have a map, the next step is to determine which clues go where, and thus build the all-important links between the rooms. What is the clinching piece of information that identifes the ultimate killer, what points to where, and so on?
Here I advise bearing in mind the "three clue rule": for each conclusion you want the PCs to reach you'll need three clue - they'll miss one, misinterpret the second, and "get it" on the third. That's not an absolute rule, but extremely useful as a guideline.
Step Five: Populate the Challenges
The other thing to consider is that in most adventures the rooms have something in them to impede progress: monsters to fight, a trap to evade, and so on. The mystery should be no different - each time there should be something to do so that the whole thing remains engaging. There will be a mix here, including social challenges, sub-mysteries, and indeed combat, but they definitely need to be there!
And that, essentially, is that.
There is of course, more to consider when building the adventure:
- Don't make it too difficult! While Poirot and Holmes are able to put the whole thing together based on a couple of tiny details, they are only able to do so because the writer has control over exactly what information they have and how they interpret it. Your PCs don't have this advantage, so they'll need more. Pitching the difficulty is tricky, especially at first, but the players will enjoy a too-easy mystery a lot more than a too-difficult one.
- Include sub-mysteries. A classic blunder in D&D mysteries is to have a cast of helpful good-aligned characters and then one evil-aligned villain. This makes it really easy to find the bad guy: even if detect evil doesn't do it, all the players have to do is look for the one guy with something to hide. But in a good mystery everyone has something to hide, and many of these sub-mysteries will come out as a result.
- Don't forget the action. It is easy to build a mystery with a lovely intricate set of clues for the PCs to follow, let them go, and watch them puzzle it all out... and to forget to include any combat and so miss out on arguably the most important pillar of the game. Be sure to include some exciting action scenes: combat, chases, and the like, for the PCs to enjoy!
- Remember that you're trying to build a solvable crime, not a perfect one! When looking at how the murderer did the deed, consider also what went wrong: what did he forget to consider? What clues did he leave? What did the maid overhear? And how does the villain respond when some of this starts to come out? (This latter is especially useful, as it may well generate more clues for the PCs to follow up on - the hasty improvisations are never are carefully calculated!)
Good luck!
