I try not to do investigations in games because I think they’re generally a poor fit for roleplaying games given current design in general. That is, I’m sure we can develop some better tools, but mostly games are pretty bad at it, yet it remains something people try to build a lot of campaigns on.
In other media
Let’s start by considering movies, books, tv shows, etc. that work on investigation. A lot of things happen here which we don’t see transferred well to roleplaying games.
Source rule #1: You always learn something
First off, what makes investigations entertaining is a combination of tight editing to keep the pacing rolling, and good characterization. Either you’re seeing all the fun parts where someone learns something neat about the plot, or it reveals something interesting about the characters involved.
Source rule #2: Information THROWS itself at the protagonists
If you do end up reading/watching a lot of investigation stories, you find that there’s a lot of contrived and implausible things that happen. The protagonists just happen to stumble upon things by accident, overhear the conversation that makes you suspicious of the right people, accidentally find the hidden note, etc. Even half the time they come across something completely unrelated that reminds them of an important thing they already knew that puts the whole situation into context.
The one RPG I’ve seen do this well is Dogs in the Vineyard – the problems are all about people, and the rule in the game is that everyone wants to tell the PCs something, and that something is usually only part of the truth best biased in their favor. Or, if they out and out lie, the GM should tell the players, “They’re lying! You can tell they’re lying…”
Source rule #3: Protagonists are ALSO a source of information
Good investigation stories build on how a protagonist thinks and contextualizes the world. How they get a read on a character, and how it fits in to the setting as a whole.
“This was a clean murder. A cold one. Mr. Govey’s shouting and shoving was a pretty good indication this wasn’t the kind of guy to do the murder. Well, to murder like that, anyway. The problem when you’ve got people who know some dirt, is they always assume you’re looking for their dirt, and it makes it a giant hassle to learn anything. And everyone knew the docks had a long history of dirt.
‘Govey! Do I look like I’m here from the feds? My Uncle Carlos was friends with Jim, and he’s just trying to make sure it wasn’t his gambling that got to him. Carlos owes money to the same people and he has to know if its time to leave town.’ Sound like you, too, are trying to cover your ass, then watch them spill the dirt. Nothing like comrades in crime.”
The “Not-Really-An-Investigation” story
There’s a lot of action stories which use the idea of investigation, or clue trails, as props to tie together fight sequences. These aren’t real investigation stories, and so, they’re even more contrived and shaky as far as clue logic goes.
The unfortunate part is, a lot of RPGs tend to borrow from this kind of fiction and this is where a lot of investigations in RPGs fall down, completely. If this is the kind of story you want to run, there really shouldn’t be any dice rolls around the clues at all, or any question of finding them – they’re basically signposts to the next fight. Don’t waste your group’s time, or your own, by making them into challenges that can only really block the point of your game.
Bringing it to your game
Assume Competence, Give information
I always like to assume the characters are competent as some things, and they have angles, leads or ideas to start with, based on what they observe. Giving those to players is a good way to feed context and often “Do I know about…?” kinds of questions can have you suggest some leads.
“The body was hit hard enough that it slid several feet on the floor, you can tell by the blood smear. Whatever attacked him was very strong. Nothing else in the house has been damaged or disturbed, which says this thing was intelligent. So that rules out the werewolf theory… you can poke around the house to look for more clues, or maybe check with the Old Man and see if he’s got any suggestions. Or you can try to convince Lacey to use her psychometry to read the room, but you still owe her from the last job…”
So it’s about feeding some information based on things the characters are already good at, and pointing at obvious paths to go from there. If the players have other suggestions, try to go with those, too, and unless it’s completely weird or unfitting, try to make sure some useful information can be found that way. If it is totally off track, then try to simply summarize in a short sentence, “You spend an hour looking around, but you realize this is a dead trail.” and move on.
Stakes and Dice Rolls
The other half of the problem is what do we roll for? If you roll to see if you know something, or if you learn something, the obvious answers are “You do” or “You don’t”. If you do, then how much of the mystery is solved? If you don’t, where do you go from there? These are both kind of unsatisfactory outcomes, which makes setting up conflicts and tests based on this, a terrible thing. Instead, consider these options for when and how to make investigation rolls:
When the character is trying to learn something outside their personal expertise, the roll is more to see if they know who to go to, and where to look for more information. Success means they know “just the person!” to seek out, whereas failure means they know who to go to, but dealing with that person is a pain or they owe them something.
Leaving Clues in Return
Are there people who would look unkindly upon this investigation? Failed rolls mean you are spotted, you ask questions in a tactless, obvious way, or otherwise, leave clues that you’re involved. Folks may take action to scare you away, attack your reputation, or otherwise stop you. Notice that this doesn’t have to be the villains – it can be other authorities who don’t like or trust you.
People who will give you info, or access to something to learn what you need to learn… for a price. It’s rarely money or goods – often it’s favors in return. These favors may be risky, or implicate your into other kinds of trouble.
Do friends use friends?
“Look, I’ve done you too many favors already. I can’t do this, this time. I’ve got too much else going on.” “Please, this is really important.” “You can’t ask this of me.” “C’mon man.” “You know what? If I do this, everything from before? You can forget it. This is the last time.” “…ok.”
Failure means losing the respect of a friend, family, mentor, or contact.
Clues put together… too late
You roll the dice ONLY at the moment of truth. “You failed the die roll? You just realized the vampire would only recognize the Eastern Orthodox Church as holy ground – this church doesn’t count. None of you are safe here.”
This is a lot of fun, but there’s two things you have to do when you do this trick. First, you have to give bonuses if the players have taken good precautions and made intelligent research beforehand. Second, if the roll is failed, there needs to be a way to get to success nearby and that needs to be set up as the goal once this roll is failed.
If it’s a monster of the week, the key to defeating it’s weakness might be across the street – if you can get there. If it’s finding out who the murderer is, it’s trying to catch them in a chase scene, etc.
One of the best games for investigations is Inspectres. Inspectres is one of the first games to really focus on narration trading as a key mechanic, and it works well for investigations. Instead of the GM having a set of clues hidden for the players to discover, successful rolls mean the players make up, on the spot, what the clues are and what they point to.
Naturally, this means as a GM you can’t have a hidden solution, or a full backstory worked up, since the players might take it in a completely different direction – on the other hand, it also means you don’t have to prep much. Just take the clues the players give you and run with it.
A lot more investigation games can work like this, though it changes the experience from “uncovering” a puzzle, to one of creating a story as you go. It means any success will always produce something one of the players found interesting and it also means that there’s not a lot of time wasted poking at dead ends. The thing to be careful about is having everyone in the group on the same page about what fits for genre expectations as to what kinds of motivations/types of mysteries make sense.
If you find my blog entertaining and valuable, consider supporting me on Patreon.