How to Write a Murder Mystery Party (Plus a Sample Murder Mystery One-Shot)

I recently wrote a bit about an event I got to do with students where I teach--I was able to run 3 simultaneous adventures for almost 30 students. One of the 3 adventures was a murder mystery, a format that I have made use of in the past. I especially like the murder mystery format because it opens the door to creative thinking and roleplaying with minimal focus on combat. As I have said before, these are the height of tabletop roleplaying games, something that only these games are truly capable of. So in this guide, I will walk you through the steps required to create a murder mystery. 

I want to start by acknowledging that there are two ways I have run murder mysteries in the past. The first way is to create a series of steps that lead the players from the beginning setup into the solution of the mystery. This is the model that I used for the mystery campaign, where my players had to investigate and confirm reports in order to advance the story. 

Our beginning approach is the same for any adventure: we establish the setting and cast of characters. Here's the outline for this section that I prepared for the D&D event I mentioned earlier: 

All party members have received fancy invitations to a dinner party at a haunted house, the daunting Lennert Manor, hosted by Count Garibaldi Lennert. Minutes into the dinner, all the candles and torches in the manor go out, and when they are relit, Lennert is dead. Thinking quickly, Lennert’s right hand woman, Henrietta Dorner, locks all the exits and asks the party to question the guests and determine the murderer.

Suspects:

·         Pango Duribin, gnome, ringmaster at Duribin Brothers Circus, short-tempered and silly

·         Nia Surasian, elf, librarian at Grand Treebranch Library, solemn and empathetic

·         Prog Drenk, orc, cleric of Gruumsh, quiet and sage

·         Riell Barrel, halfling, diplomat, exciteable and kind

·         Peter Bronzebeard, dwarf, master smith at Bronzebeard Burnishing, irritable and long-winded

This simple setup and list of characters was enough for an entire session's worth of material. The players entered the manor, experienced the murder, and were tasked with solving it in the first half, and the second half saw the players beginning to question people. It's at this point that we need to make sure that one of the special murder mystery considerations is in place: knowing the direction we're headed. 

In order for the mystery to make enough sense to your players to solve, you'll need some consistent description of it that way. This is where deciding the culprit ahead of time becomes crucial. My personal recommendation, though, is this: if the culprit is one of five people in a closed room, they are going to get tripped up under the players' scrutiny. One way to get around this is to go with the approach that I adopted for this mini-campaign: make the culprit someone who isn't immediately obvious. For this game, I decided that a ghost living in the manor was the murderer and that many of the clues found will deal with recognizing supernatural happenings. So this first session is something of a red herring, which I have preached against in the past, but I would argue that this session establishes the setting and stakes while also building a tension in the players as none of their ideas come to fruition as they interview the living suspects. 

Now, originally, my plan for this first session ended this way: 

After the party makes a formal accusation to Dorner, the lights once again fall, and whoever they accused is dead when the lights return.

I decided against using this twist, as my players were nowhere near having a real suspect in mind at this point, and also because the tension was already incredibly high in the party as they struggled to come up with ideas, so there was little need to up the ante. This led us to session 2. Here are my notes:

Dorner locks all of the remaining suspects into individual rooms. She suggests that the party investigate the manor for clues. Primarily, the party needs a murder weapon, but the manor has many secret hiding places that might obscure that weapon. Dorner directs the party to the three nearest rooms to search—the study, the lounge, and the kitchen. (The final puzzle solved reveals the murder weapon.)

·         The Study – Inside the study are walls of books, each shelf arranged by the color of the books’ spines. If a book is removed from each shelf in the order of the color spectrum, one shelf recedes into the floor and reveals a staircase into a small basement chamber. In the chamber is a stone statue with hands clutching where a dagger should be.

·         The Lounge -  Inside the lounge is a series of chairs and sofas as well as a grand piano. If the piano plays every note in order from lowest to highest, the largest sofa sinks into the ground, revealing a passage into a small underground chamber. In the chamber is a small laboratory, complete with an intermediate poison-making kit.

·         The Kitchen – Inside the oversized kitchen are the signs of a feast being prepared of all manner of cuisines. If the apparent ladle in the cauldron of stew is removed, it can be discovered to actually be a deadly steel hammer.

I wanted to provide three different murder weapons to keep the mystery open. Just as in the first step, we kept from giving too much away at once, we do the same here. The discovery of the first murder weapon will excite the party, but the discoveries of the second two will throw them back into frustration. 

You may be asking yourself, "Why do we want the players to be frustrated? Isn't that a bad thing?" Yes and no. It is bad if the players experience the kind of frustration that makes them feel that they can't progress. It is good if the players experience the kind of frustration that makes them want to accomplish the goal even more. This kind of frustration is the same kind of frustration we experience when a character on TV or in a book makes a choice we don't like, but it makes us cling harder to the story to find out how it goes. This is what we're striving for. 

Here are a few telltale signs that can indicate whether it's good or bad frustration your players are dealing with:

Good Frustration: 

• Players immediately considering new plans when theirs fail and not giving up
• Laughing when a plan fails
• Making jokes about potentially bad plans

Bad Frustration:

• Players saying that they don't know what to do, especially with a negative tone
• Making suggestions or taking actions without considering the consequences
• Losing momentum for more than a moment or two

If your players are in good frustration, begin preparing to move them forward with small plot developments when they show signs of having energy that stalls out. If the players are in bad frustration, it's time to course correct. Have some force in the gameworld deliver a new advancement in the story so that the players can pick up steam again. 

I am using a combination of our two methods here: these frustration measurements and adjustments are improvised reactions to the players, not carefully-prepared notes. And that's going to happen. No game can be completely prepared for, so we need a healthy mix of the two approaches. 

Once the players discovered the weapons, they set to trying to match the weapons to suspects. This again resulted in frustration, as they still didn't have any idea of the ghost angle. When things began to slow down for them, I had a spooky moment with cold air and disappearing lights, then had the NPC leader of the story inform the party of the ghosts that haunt the manor. And that led us to day 3:

·     Perception Checks: Players can sense that the manor is colder than it should be, and they continually see fleeting ghostly images at the edges of their vision.

·   Arcana Checks: Players can sense that magical energy exists here, and it appears to be supernatural rather than purely magical.

·   Insight Checks: Players can sense that all living people in the manor are genuinely afraid for their safety.

In the attic, players discover a series of homunculi which animate and attack them. When defeated, these homunculi’s remains include a slate with “In the air of Lennert’s place / We will make living things replaced.” When said aloud, the words on the slate animate and glow with magical energy, opening a rift to a spectral dimension. Inside, the players can see ghostly doubles of the dinner guests.

There was another full paragraph I ended up cutting entirely. It proposed a system in which each dinner guest's double could be asked two questions, and would answer one with a truth and the other with a lie. I realized, though, that while this puzzle would work well with adults who had the system explained to them, it would not be as smooth with middle schoolers who had to intuit the rules of the system. So I scrapped it in favor of a simpler rule that I could illustrate to the players more easily. Anything that a living being did in the ghost realm would cause the identical action to happen back in the living realm and vice versa, but a task performed in the human realm by a human would not be repeated in the ghost realm. This allowed me to begin planting clues about how the murder weapons had to have been moved by ghosts, for instance. Such improvisations can be necessary, so be ready. 

This step also marked the first combat of the mini-campaign. In a murder mystery, the emphasis is, as I said above, on roleplaying and creative thinking. To that end, I made an encounter that required strategic thinking, puzzle solving (via arming the party with weapons that could actually harm ghosts), and advancing the story. I tried to avoid any purely combat-based challenges because that wasn't what my players signed up for. Do this when you can--remember your audience. 

This leads us into the second half of the adventure. Here are the notes for what I believed at the time would be the final session:

The party is allowed to question the remaining dinner party guests. Once they feel they are ready for an accusation, everyone waits for their claim. Once the players accuse the correct person, they transform into ghostly form, and the party must battle the ghosts. When the final ghosts have been dealt with, a massive spectral version of the manor separates from the physical object, and the ghosts leave with it. The dinner guests thank the party for their help and send them on their way with new titles and privileges.

This spare entry allowed me to bring a dramatic end to the mini-campaign. Once the players successfully recognized that the ghost haunting the manor was the murderer, it was a sprint to the end. They defeated the ghosts, and I ordinarily would have simply called that the end after granting them rewards--but as part of the event, I had to fill another day. So I extended it by adding a final component in which the players became lost in the space between the living realm and the ghost realm and which ended up transferring them back in time to before the murder--the suddenly living host granted them ownership of the building where they fought the final battle. The puzzle of setting things right in each realm became the denouement, which I felt was a nicer ending than the "you did the job, here's the pay" kind of ending that so many tabletop stories become. 

Of course, it is possible to improvise even more than this. I've explained that some parts of a murder mystery require planning for consistency--such as what the twist of the case is, like the ghostly murderer. Others require flexibility--such a pacing a session to capitalize on good frustration and avoid bad frustration. A purely written murder mystery would never be satisfying to an average group of players, whereas a purely improvised murder mystery would likely lack some of the structural consistency that planning could provide. The trick here is to walk the line between the two. 

When writing your own murder mystery, remember to consider player abilities. If a character has "True Resurrection" as a spell, the murder won't stay a murder for long. If the murder needs to happen in obscurity, make sure that it's actually blocked from all the players' view (darkvision complicates many "in the dark" moments like these). If a character has an obscenely high score for something like Investigation or Insight, it could really throw off your ability to convey story at your own pace. Be careful with these possibilities, as they can turn a campaign into a one-shot under the right conditions. (I'm reminded of the time I built a deadly combat-based rogue and single-handedly killed the end boss of the campaign and his henchmen in my introduction session--my DM had to completely rewrite the campaign after I destroyed the end boss, who doubled as our main quest giver. Things like this happen in tabletop games more often than anyone would really like.) 

A few other final considerations for developing a murder mystery: pay closer attention to table talk than you normally do. You need to know what your players are planning, and you need to know right away. Responding to normal D&D can be easy enough--providing NPCs with simple things to say is one thing. But you need to be manipulating the setting and plot in your head in real time, so not having your fullest attention means that something is likely to go wrong. Additionally, do not feel attached to your outline. Adapting to player suggestions, successes, and failures is more important in a roleplaying- and creativity-based challenge than in a combat-based challenge, so be ready to adopt the players' ideas when it works well for your story. Finally, don't forget that the thrill of playing a murder mystery is not the same as the thrill of defeating a big bad. The excitement comes from advancing through the story based on your players' wits, and that's not something we're all used to running. My advice is to make sure that as the story progresses, you try to surprise even yourself. I didn't see the ghost realm/living realm distinction coming when I was preparing, but it became the key mechanic in solving the case. Remember so long as the theories your players suggest make sense, they are valid solutions to your mystery. 

That's all for now. Next time: a guide to politics in my homebrew setting, a guide for creating 5E characters quickly and easily, and why D&D serves as a stepping stone to other tabletop game experiences. Until next time, happy gaming!

Back to the homepage (where you can find everything!)