Collaborative Mystery Games


Running a mystery in a tabletop RPG is pretty hard, as it turns out. An RPG traditionally has a single source of truth, the GM, and thus, all information required to solve a mystery must pour forth from the GM’s mouth. Unfortunately, players have a tendency to assume the converse, that everything the GM says must be essential, simply by the merit of having been said. Combine this with the normal problems constraining a mystery (clues being too obvious or too baroque, unclear motivations, easily sidetracked audiences, etc.) and the whole endeavor is pretty hard.

There have certainly been great leaps in portraying a mystery in an RPG. Gumshoe is a fantastic system for running mysteries, as it makes the fantastic observation that the useful part of a game is not finding clues, but rather understanding their role in the mystery at large. However, it’s still victim to a fairly fundamental psychological problem: sometimes, players just get locked into a train of thought that isn’t right, and end up frustrating themselves as they chase loose ends.

I’m experiencing this right now, in fact, as my current Dungeons and Dragons game is centered around a mystery. I can feel the frustration as my group, myself included, get stuck in our preconceived, false notions about the mystery, both angry enough at the dead ends to know we’re barking up the wrong tree, and too rooted to our current assumptions to be able to create alternative hypotheses.

This got me thinking: player agency with regards to the story of an RPG is sort of the new hotness right now. Plenty of games offer mechanics by which players can create truths about the world around them, and plenty of GMs nowadays are OK with, or even very into, the idea that players should get an amount of say with the game world.

What if players got to dictate truths about a mystery game, defining a mystery as they solved it?


So, the immediate problem with this idea is that it, well, fundamentally destroys the concept of mystery. If the players can just point at a guy and go “He did it”, and actually warp the fabric of reality itself such that he most definitely did it, there’s no mystery. They’re no longer the audience, but the writer, and a writer is not surprised by their own mystery.

With this in mind, such a system should not give the players complete control over the path of the mystery. Instead, players should have a more granular control over the clues, and be given some say as to what matters and what does not. The greater mystery as a whole is still left in the darkness, under the watch of the GM, but the players can find some fact in the world, some detail or clue, and say that, yes, this is indeed a piece of the puzzle, and here’s why. It’s then up to the GM to determine how that known piece of information bridges into the unknown mystery.

Here’s a really rough implementation idea. Let’s say that every character has some sort of knowledge domain, just like how Gumshoe does it. The rough-and-tumble street thug knows all about the underworld and crime and thievery. The posh noblewoman knows all about courtly traditions and noble bloodlines and gossip. The coroner knows all about wounds and blood spatters and poisons and bodily decomposition. Along with these domains of knowledge, the players are given some sort of metacurrency. Let’s call it Deductions.

When a character comes across a clue in their particular intellectual domain, maybe a splash of dried blood on a couch cushion, maybe a broken lockpick beneath the window, maybe the knowledge that Lady Verisimilitude or whatever left on the day of the murder to go to the the royal banquet at Bangers-and-Mash-upon-Thames, that character may burn a Deduction to make a conclusion about that clue, a conclusion that is, of course, completely pulled out of their ass.

“Real blood will dry brown if left to stain for that long. This is still red, and thus must be fake”

“This particular kind of break is most common if one attempts to pick a lock with strength instead of finesse, a common mistake for an amateur to make.”

“Bangers-and-Mash-upon-Thames cancelled its banquet this year after the local Baron fell ill, so Lady Verisimilitude must have gone elsewhere.”

The important thing is that these are facts from this point forward. That splash of blood must be fake, that thief must not be that good a lockpick, and Lady Verisimilitude must have lied about her whereabouts. The key here is that no conclusion is drawn here that cannot be pulled directly from domain knowledge. The player is not allowed to assert where Lady Verisimilitude was, merely that she was not where she said. This would have to be enforced as a rule.

In this way, I guess it would be more accurate to say that the players are not actually providing any answers to the mystery, but are instead being given the power to say what questions have answers that are important. It’s still up to the GM to say why there’s fake blood on the sofa, who in the town is a crap lockpicker, and where Lady V was, but the players can rest assured that those details have been codified as important, and thus, time and mental energy spent pursuing answers will not be wasted.

Because that’s ultimately the death knell for a good mystery, is if a lot of time and energy is spent by the players, and they feel like they gain little out of it, most likely because what they’re pursuing is simply not important to the mystery that the GM has laid out. By giving the players a sort of mechanical reassurance that what they’re doing is important (because they have enforced that it is, no less), they can be assured that they’re moving forward.


Speaking of the mystery the GM has laid out, they’ll have to do so somewhat differently now that the players have the ability to assert facts. Generally speaking, you’ll have to make sure that your mystery is solid enough that the players aren’t completely dictating the story, yet flexible enough that it can incorporate all of the things that they believe to be true. Furthermore, since Deductions are a limited resource, you need to ensure that every clue you scatter through the game has both a place where you expect it to fall in your web of mystery (in case no one uses a Deduction on that clue), but also is general enough that players can slot them in a variety of places as they see fit.

Perhaps you just meant that broken lockpick to signify that the intruder was not invited in during the Grand Ball, and had to break in, but the concept of the lockpick being broken in an amateurish fashion gives you an extra wrinkle to work with. Maybe the thief’s bookshelf has a dog-eared copy of “Lockpicking For Dummies” stashed on one of the shelves, or their garage has padlocks lying on the workbench, clearly for practice. The detail the player created hasn’t changed who the thief is, but rather acted as a piece of bait that the players bit on to, that you can use to reel them towards the revelations.

I think such a system could be incorporated into any game already capable of running mysteries. Just, at any point where the GM would describe a detail of a clue, instead turn to the player and say “What do you notice?”. I think a certain key would be to give players a finite amount of ability to do this: too much and they’ll end up convoluting the plot beyond the GM’s ability to improvise, or they’ll line up facts in such a way that it basically forces a certain conclusion to be true, ending back up at the point where they are both writing and reading the mystery.

So, where does this get you? Well, your players will be given facts that are 100% guaranteed to be both correct and useful, and since they’re dictating them, they’ll never have to worry about being stuck down a train of thought that’s wrong, because they’ll be able to demand that their focus is, at least to a degree, relevant. This eliminates the guessing game of “Is this a red herring, or is this useful?” that tends to suck the fun out of mystery games.

Here’s the really bloody secret, though: you should already be doing this. If, as a GM, you have dictated from the very beginning how each and every clue relates to your mystery, even if you’re not concealing those clues behind skill checks, you’re running the risk of players getting stuck in those logical dead-ends where they’re focusing on the wrong thing, interpreting a clue the wrong way, or trying to kludge every detail into the incorrect hypothesis they already have. Frankly, the more players are confident in their wrong answer, the less fun you’re going to have.

I’m not saying “the players should always be right”, but instead “the players should always be half right”. Don’t give them the entire mystery, but instead give them just enough that the twists and turns of the mystery come from the parts they do have set in stone, instead of coming out of left field because the players were dead wrong in the first place.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s