Designing and Running a Murder Mystery
'Eliminate the impossible, and whatever remains, however improbable, must be the truth.'
The words of Sherlock Holmes perfectly capture the essence of the classic detective story, the process of elimination through logical deduction and reasoning, leading to the ultimate and inevitable revelation of the culprit. Early murder mysteries were essentially a game played between author and reader, a complex puzzle in which the reader was presented with the clues and could attempt to match the great detective in uncovering the dastardly plot.
It is this puzzle element which makes the classic detective story the perfect model for designing and running murder mystery adventures in RPGs. By following the basic framework used by the likes of Agatha Christie and Arthur Conan Doyle, gamesmasters can construct intriguing and challenging mysteries with which to perplex players. A good mystery can enliven any campaign, but it can be a difficult scenario to design. Hopefully this article will go some way towards simplifying that process, by identifying several key elements which underly the majority of popular murder mysteries.
A good mystery needs at its heart a murder. Other crimes simply don't have the same impact. The crime is your hook, and it must be sure to grab the attention of your players. A murder is a significant event (especially for the victim). And besides, it's traditional.
But a scenario requires more than a simple murder at it's heart. Your crime must be startling, perhaps even bizarre, in order to grip the players and sustain their interest through the course of their investigations. Upon arriving at the murder scene the detectives should immediately be presented with a number of questions beyond the obvious how, why and who: What could have caused those scratch marks? What is the significance of the arcane symbol daubed in blood on the mirror? Why was the victim wearing sunglasses and a sombrero?
Having decided on the strange circumstances of the murder itself, you need to devise some logical if unlikely story to explain it. Work backwards from the murder scene, filling in details and adding new ones as you go. It will help you to visualise the scene, and to explain how it came about, if you write down exactly how the murder took place, even though you will never present that information to the players directly. Make a note of where each of the major NPCs was at the crucial moment, what they were doing there, and why. Trace their movements immediately prior to and following the murder. Only by gaining a complete knowledge of each suspect will you be able to decide how and when to present the players with the clues they need to solve the mystery.
You have your murder, what you now need to do is decide how and when to introduce it to the players. If the PCs are known investigators or detectives, they can become involved in an official capacity. More likely, you will have to come up with a reason for them to start investigating a suspicious murder. The victim could perhaps be a friend or contact (a great way to remove an NPC who the players tend to rely on a bit too much); they could be approached by such a person who trusts their abilities; they could be witnesses or even suspects, needing to uncover the truth to get themselves out of a fix. Provided your players are interested in puzzles and mysteries, you shouldn't have too much trouble getting them involved.
There remains the question of when the murder should occur. The traditional method, and the simplest, is to have it happen immediately before the start of the adventure, with the PCs called in to examine the scene. However, if you wish them to get to know the victim, and the suspects, as well as passing on a few discrete clues that will only later be revealed as significant, there are a number of ways you can lead in to the murder:
There is no need to limit yourself to just one murder. A second killing can be used to present additional clues, or spread further doubts. It can also spice things up a bit if the plot is starting to drag; or possibly reveal a pattern, suggesting a serial killer. In addition to working out who the killer is, the players also need to discover who will be next on the list.
Alternatively, there could be more than one initial victim. The possibilities include, but are by no means limited to, the love triangle death scene, and the classic ghost ship - space or ocean going - littered with unexplained corpses.
Locked rooms are another staple of the early murder mysteries, typically with the body found in a room locked or bolted from inside, raising the difficult question of how the murderer could possibly have entered, and subsequently escaped.
An important principle in the world of mysteries is that of fair play. When designing a mystery adventure you should always be aware of the need to fairly set before the players all the information they need to solve the puzzle. One way in which you can maintain that essential fairness, and at the same time help yourself, is by limiting the number of suspects to a defined, closed circle. The detectives know at an early stage that the murderer is one of a group of six to eight individuals, each of whom appears to have a very good reason for wanting the victim dead; and the gamesmaster can focus on those suspects, making them more detailed and interesting characters. This may not be realistic, but it makes the investigation a great deal more manageable for players and gamesmaster.
Some form of physical limitation is the easiest way of closing the circle. Isolate the suspects, along with the victim and the detectives, in a remote country mansion, a private island retreat, on a luxury train, or a riverboat. For a more ambitious and expansive mystery, you could delimit suspects according to particular knowledge. For example, the murderer clearly had an intimate knowledge of poisons, or was familiar with firearms, or knew about human pressure points. Perhaps only a handful of people knew where the victim would be on the night of the murder, or that she was afraid of heights. In this type of mystery the closed circle is less obvious, and the freedom of movement means that the investigation may involve tracking down suspects. It is the more realistic method, but also the most difficult to run, and should be attempted only by experienced gamesmasters.
Now we come to the question of evidence. You have your murder, and you have your closed circle of suspects. What you now need to do is determine which of them actually did it. For the detectives, this means establishing that one of the suspects had access to the means of committing the crime, a credible motive for doing so, and sufficient opportunity to put their plan into practice.
The gamesmaster, having decided earlier precisely what means were used by the murderer, now needs to determine who could reasonably have had access to those means. For example, if the victim was killed by a gunshot, which of the suspects had access to the gun, and the ability to use it?
Next comes the question as to which of them had a genuine reason for wanting the victim's death. Typically, most if not all suspects will have some form of motive, with varying degrees of credibility. This ensures that the detectives can't dismiss some suspects at an early stage. Some suspension of disbelief is necessary here; after all, it is unlikely that half a dozen people would all want the same man dead, and even less likely that they will all gather in the same place, but I'm sure your players will bear with you.
Here is a selection of possible motives for murder: money, self-protection, fear, preservation of status, protection of another, revenge, acquisition of power, love, jealousy. Add to that the possibility of accidental death, mistaken assassination (wrong victim), suicide, and insanity, and you begin to see the wealth of potential for confusing the players.
Lastly comes opportunity. The detectives must prove that not only did the suspect have the means with which to commit the murder, and a genuine motive, but also that they were in the vicinity at the time of death. This brings into the equation the issue of alibis.
Most suspects will have an alibi, though in many cases it will be false. To determine alibis, go back to your description of events surrounding the murder, and account for the whereabouts of each suspect. Who could possibly corroborate their story, confirm that they were nowhere near the bedroom when the gunshot was heard? Work out who has a genuine alibi, who provides it, and how the players can come by this information.
More difficult are false alibis. Clearly the murderer will want to establish that they were elsewhere, but who will be prepared to lie for them? A friend, a lover, or a relative? Perhaps they have bribed someone to support them, or resorted to blackmail. Or maybe the person providing the alibi intends to use their knowledge to blackmail the murderer, making them a potential next victim.
You might cause further confusion by having an innocent person also concoct a false alibi, perhaps to hide an illicit liaison, or a dark secret, unconnected with the murder itself.
Again, by working backwards from what actually happened you can determine what type of false alibi will succeed. If no-one knows where the murderer actually was, where could they have reasonably been? Alibis can provide a great deal of confusion, as conflicting accounts leave the detectives wondering just who they can trust. As always, for the sake of fairness you should provide some means by which the players can identify who is telling the truth, but that doesn't mean it should be easy.
The players can only solve the mystery by making use of clues put fairly before them by the gamesmaster. Clues are the signposts with which you guide the players through the mystery maze, and you need to ensure that you have left enough signposts, and in the right places, in order to create a satisfactory scenario. Your riddle must be challenging, but it can't be so devious as to be virtually impossible for anyone with an IQ under 200. Similarly it should not be so obvious that a child or a PE teacher could work it out. The line between the two is indeed a fine one.
There are two fundamental questions concerning the distribution of clues in your adventure: what type of clues to give, and when to give them.
Clues can be divided into three categories:
Deception is the name of the game when introducing clues into the plot. Some clues will be obvious, to get the players started, and provide them with encouragement at strategic points during the scenario, but most of the clues should take some finding. Ideally, you want to give a clue in such a way that the players don't realise it actually is a clue, at least at the time. There are a number of techniques you can use to disguise clues, all of which satisfy the requirements of fairness:
Of course, it is easy to try too hard to disguise clues. Often it can be just as effective to state them boldly, and thereby conceal their importance in sheer obviousness. To quote an example from Poe, often the best place to hide a letter is in a letter rack. This is a most satisfactory method, as it provides the players with a moment of revelation when they eventually realise that the clue was under their noses all the time. If all your clues are well hidden, players will feel that they are being cheated.
Now we come to the question of timing. The placement of clues is vital to controlling the pace of the investigation. Initially, there will be an abundance of clues, as the PCs investigate the murder scene, and identify the suspects. After that, you need to ensure a steady flow of fresh clues, as the finger of suspicion points towards each suspect in turn. Gradually, some of them will be eliminated, but you should delay one final, major clue until the adventure is nearing its climax.
Allow the players tentative speculation throughout, as theories come and go, but make sure that the arrival of the final missing piece comes as something of a shock. Perhaps it confirms the players' ideas, but it should also include some novel information, some twist in the tail. It may have been a genuine accident, or perhaps more than one killer was involved, or perhaps they correctly identified the murderer, but misunderstood the motive.
Finally, once the murderer is revealed, they generally accept that they have been out-thought. Etiquette requires that they congratulate the detectives, and that they then provide a full explanation of what actually happened and why, possibly filling in a few remaining blanks which managed to elude the players. You can always have them make a break for it afterwards, but not until they have explained everything to the detectives.
Designing a murder mystery is not just about creating a grisly murder with an obvious group of suspects. You must envisage precisely how you are going to structure the investigation, and get the timing of your clues just right. Be prepared to let things go slowly for a while, before leading up to the big finish.
A murder mystery does not need to have a contemporary or 1920s setting. With thought, an intriguing murder can be integrated into virtually any campaign. A fantasy town or city, particularly one with a fairly sophisticated legal and political structure, could allow you to explore the use of magic for murderous ends, or the motivations of demi-humans. However, if your players are known for their liberal use of extreme physical violence as the solution to life's problems, having them suddenly turn detective may be stretching the imagination a little too far.
Distant planets, alien civilisations, and advanced technologies can provide added complications for detectives. I suggest you read Asimov's Elijah Baley novels - The Caves of Steel, The Naked Sun, and The Robots of Dawn - for excellent examples of what can be done with the sci-fi detective story.
Other genres lend themselves with similar ease to this type of adventure: horror, supernatural, cyberpunk, superhero, secret agent. You might even be able to come up with a Toon whodunit? if you have a mind that is sufficiently warped. A well-done murder mystery can provide an added dimension to any roleplaying game, and I hope this article has given you plenty of ideas for creating your own.
Of most use to the roleplayer are stories by the likes of Christie and Conan Doyle, and Dorothy L Sayers, where the element of a puzzle features strongly. Also recommended are the seminal Dupin stories of Edgar Allen Poe - The Murders in the Rue Morgue, The Mystery Of Marie Roget, and The Purloined Letter.
Film and television adaptations of classic detective stories are too numerous to mention, but they can be of great help and require less work than reading the book. Of particular note are the Cadfael series, based on the Ellis Peters tales, and the Adventures of Sherlock Holmes starring Jeremy Brett.
On the techniques of constructing murder mystery stories, try Writing Crime Fiction by HRF Keating, which provided the basis for much of this article, and Hillary Waugh's Guide to Mysteries and Mystery Writing.