• The Infractions Forum is available for public view. Please note that if you have been suspended you will need to open a private/incognito browser window to view it.

What "parts" are required for a mystery/investigative scenario?

Chris J

Registered User
Validated User
What "parts" are required and how best to build a good mystery/investigative scenario (I'm abbreviating that to m/i to save sanity later on)? I'm thinking along the lines of Delta Green.

For example, if you read various Call of Cthulhu, Trail of Cthulhu, and Essoterrorist scenarios (to name but three), they must all contain a particular formula that defines them as something different to an all-action leave your thinking caps at the door type scenarios.

So who, how, why, what, when, where?

Aside from the usual parts that make any type of scenario exciting, what parts are exclusive to good m/i scenarios?

A twist
A red herring (used sparingly)
A ?

More to the point is, how do you put all this together?

The "3 clue rule*" - as solid as that bit of advice is - doesn't really tell me how to structure m/i scenarios.

* https://thealexandrian.net/wordpress/1118/roleplaying-games/three-clue-rule

I've also read to start at the moment of the crime (or whatever is the focus of the m/i) and work backwards. I tried this and it didn't really help other than to get me confused when jotting stuff down on paper.

So how do you put together m/i scenarios; what parts do you use and what formula do you use?

I really enjoy GMing m/i scenarios but I cannot figure out how to construct my own.
 
Last edited:

Faethor

Registered User
Validated User
The only way I know to make mystery scenario's is to reverse-engineer them (which is kinda obvious if you think about it), you approach them in reverse order the players encounter them. So you start with the conclusion - the information - imagine major plot points, twists and set scenes you think would be cool and them look at the breadcrumbs you can lay to get to them. Produce lots of breadcrumbs or puzzle pieces and a few mahoosive ones that you can interject to create a path from point A to B should the players hit a brick wall (even the brightest players have off days they grind to a halt).

(and unless you are using gumshoe) Don't let dice rolls derail the game. If this means resorting to a degree of subterfuge and vagueness ("... everyone make a roll.. hmm...") its better then openly defeating the rules and just telling them anyway in a fit of exasperation.

The best clues don't offer answers but questions - they are intriguing, they motivate the players. Harness this and adapt your ideas / improvise to keep that excitement at the table.

I used to love the conclusions players jumped to playing coc (especially the horrible ones) - often their own imaginations actually trumped the scenarios.
 

PBWmaia

Registered User
Validated User
One thing that I know for sure to *avoid* is the clue that can only be deciphered by one, and only one, person in the group. I was in a play by post game in which the clue (kind of a choke point as in the linked article above) could only be interpreted by one member of the party (and apparently had been provided to that player in a note, so we couldn't even do the work-around of "B told us..."). Well, that player just quit posting, the DM never bothered to supply info to the rest of the group, and since the only way forward required B's input ... game went kaput.

I like to also consider mentally running it forward to the "crime" (I'll use crime, even if it's not a traditional crime that is the focus of the m/i). Then I'll have a good idea of what sort of traces might have been left (BadGuy cloak catches on a rough bit of wood, leaving a small piece behind; valuable box that has been sitting on a certain spot for ages was removed, leaving its outline), and then add in a few more hints to help the puzzle get solved.
 

committed hero

nude lamia mech
Validated User
Generally, an investigative scenario starts off with a crime scene. The three most common crimes involved are murder (probably the majority, for various reasons), disappearance, and theft. These categories blend into each other: a disappearance might be a missing corpse or a kidnap, which is essentially the theft of a person.

I tend to classify all of these as the disruption of an acceptable status quo, if only to offer the investigators' discovery of an unacceptable status quo as an alternative setup (I would argue that this is the typical entry into a Night's Black Agents campaign). A parallel in X of Cthulhu would be a cult that enacts a sacrificial ritual every month to stave off the arrival of a god - they aren't making things worse by their actions, as bad as they are. And I guess a serial killer is probably the same deal, once the PCs discover old crimes they can attribute to him to establish a pattern. But just the fact that "the mob is running the town" can be enough to hang an investigation on, without any real overt crime having happened.

The only other required scene is the final confrontation with the antagonist. In between are "encounters" (I use this term because "scene" can mean both an interaction with a witnesses and the location where that interaction takes place) that contain additional clues about the crime. If the party identifies the victim's rival as a potential culprit, for example, you can call speaking with him a scene regardless of where it occurs, or you can envision the gamut of scenes depending on where he can be found (a bar, his house, wherever). Ideally, you have a trail or web of such encounters from the crime scene to the final meeting with the villain.

I'm not a devotee of the three-clue rule (as useful as the concept behind it is) because I play GUMSHOE, and because no group looks back on an investigation and remarks on the complexity of the web - to them, the trail will have a single thread, because that's the way they went! I feel that as long as you aren't making the players jump through hoops to get clues, you won't pixel-bitch or railroad them (the latter problem should only arise if the players ditch the investigation altogether, and you didn't alert them that you expected the game to be about investigation).

Nowadays I approach investigative design (which is to say all adventure design) in terms of a video game sandbox, in which encountering someone/someplace/something opens up new nodes on your master map. In actuality, it's information that's opening up the new nodes.
 
Last edited:

TheMouse

garmonbozia
Validated User
I'm not awesome at mystery scenarios, but I've figured out a few things through repeated experience:

1. Red herrings are completely unnecessary. Players will go off on random tangents. You'll mention that the curtains in a room are yellow, and they'll flash back to that time last year that someone else had yellow curtains, and they're off. It doesn't matter if it makes any sense at all. Adding more on purpose just means that whole sessions are spent doing nothing that in any way contributes to the resolution of the mystery.

2. Make every clue obvious. Repeat it to make it even more obvious. Don't hide clues. Bash the players in the face with them over and over and over again. Mention them later. So when the players are like, "There was one of those weird insect carvings in the baker's house and in the back of the church," you can point out, "Don't forget the one from last session that you found in the woodcutter's shack." All this stuff would be incredibly more obvious and easy to remember for the characters, because they're not experiencing their lives filtered through your quick descriptions. You have to make up for that deficit.

And the mystery should be more about making sense of the clues than wandering in the dark trying to find them. Making sense of things is an interesting puzzle to play with. Trying to find any clue at all is tedious and boring.

3. Try to avoid complete dead ends. Twists should be more like, "Oh, no! It was the baker's daughter and not the baker who was part of the cult!" rather than, "Nah, that insect carving in the baker's house was something passed down in their family for a few generations that someone found in the woods. It doesn't mean anything." Because the former adds context without stopping the flow of action, while the latter is basically a brick wall to slam into.
 

committed hero

nude lamia mech
Validated User
I switched the order of these great points, sorry.

1. Red herrings are completely unnecessary. Players will go off on random tangents. You'll mention that the curtains in a room are yellow, and they'll flash back to that time last year that someone else had yellow curtains, and they're off. It doesn't matter if it makes any sense at all. Adding more on purpose just means that whole sessions are spent doing nothing that in any way contributes to the resolution of the mystery.

3. Try to avoid complete dead ends. Twists should be more like, "Oh, no! It was the baker's daughter and not the baker who was part of the cult!" rather than, "Nah, that insect carving in the baker's house was something passed down in their family for a few generations that someone found in the woods. It doesn't mean anything." Because the former adds context without stopping the flow of action, while the latter is basically a brick wall to slam into.
It's quite useful to say "based on your character's investigative skill, she is confident that she's found everything she can here" in both instances. Be willing to end a scene when there's no real reason not to.

2. Make every clue obvious. Repeat it to make it even more obvious. Don't hide clues. Bash the players in the face with them over and over and over again. Mention them later. So when the players are like, "There was one of those weird insect carvings in the baker's house and in the back of the church," you can point out, "Don't forget the one from last session that you found in the woodcutter's shack." All this stuff would be incredibly more obvious and easy to remember for the characters, because they're not experiencing their lives filtered through your quick descriptions. You have to make up for that deficit.
It's just as useful to maintain a running list of potential threads, or to mention any at the start of each session.
 

TheMouse

garmonbozia
Validated User
It's quite useful to say "based on your character's investigative skill, she is confident that she's found everything she can here" in both instances. Be willing to end a scene when there's no real reason not to.
I agree. That's a useful approach.

It's just as useful to maintain a running list of potential threads, or to mention any at the start of each session.
I hadn't considered it before, but a clue notebook might be useful. Something that the GM fills out and leaves on the table so everyone can reference it. Kind of like a video game style quest book.

Thinking on it and my obsession with using index cards for everything, it might be a good idea to give each location its own card. The people can be like, "Hey, what did we find when we searched the old cabin?" then grab the old cabin card and check.

I mention the cards because it's useful for organizing thoughts. Like if they figure out that the old cabin is the scene of the murder, they could place it into a timeline. Then if they see that the next step they've got is where the body washed up, it could imply that there's a dump site upriver. That sort of thing can be really helpful. It could also combine well with game systems that have mechanical widgets where the GM gives direct clues. ("You notice when constructing your timeline that you've got the murder site and where the body ended up, but there's an implied spot between those two where the body was dumped. Probably upriver of the place where it was found.")

Yeah, I think I like that idea.
 

ElvisSavage LordSideburn

Dilletante of all trades
RPGnet Member
Validated User
You are missing...the interloper! Few things spur players like competition, so, "Oh, you just missed your friends, the couple that were just here asking about this..." Are they trying to uncover the truth, or cover it up? Who might they be looking for? Where did they say they were going next? If your players are struggling, "Look, it's the same person we saw leaving xxxx the other day," or the other party may wish to pool clues or resources. remember, the higher the stakes, the more parties will be interested.
 

Futurella

Social Justice Witch
Validated User
The standard format of my favorite mysteries, which are TV shows like Castle and Perry Mason, is that we introduce the situation, and then work through a list of suspects starting with the most likely and eliminating them each in turn, until you get to the end, with the last suspect, at which time all suspects are suddenly back in play for the final reveal.

There isn’t a red herring per se, but about 2/3 of the way though, we discover that our defendant lied to us about a key point. This is also true for House, which is Perry Mason in hospital.

Hm. I’d never thought of it quite that way before, but that’s the Perry Mason format. Classic. Now just turn that into a game mechanic (step 2) and move on the step 3 (game for sale).


Added: I realize I went off on a weird murder mystery specific tangent here. I was just following the energy.
 

Octopus Prime

Retired User
A few miscellaneous thoughts:
-I agree with Red Herrings being a bad move, at least Red Herrings that are a dead end. That said, you can have clues and suspects that lead into the mystery, but not in the way expected. Ie; Suspect A isn't the killer, they're the killers mother, and they're trying to take the blame in his place.
-As a rule, end each session on a reveal.
-My last GM's rule of thumb was "it's always the person you know." That is, as much as possible, reveals should relate back to previous characters/events and cast the past in new light.
 
Top Bottom