I've been running RPGs (roleplaying games), specifically "tabletop" RPGs, in one way or another on and off since 1981. And being a devious and twisty-minded gamemaster, I occasionally toss a mystery plot at my players. Unfortunately, this sort of plot usually ends up bring frustrating for the players and unsatisfying for me, as I end up having to practically narrate the player-characters' side of thing for the players. No matter how obvious I think the clues are, players almost inevitably run in some random direction, convinced they've got it solved when they're utterly and completely wrong.

    Worse, sometimes there is no mystery...I just expect them to hit the monster until it falls over, but they assume there's a weak spot they need to find and hatch elaborate schemes to find it.

    It was after one such "not a mystery" plot, where what was supposed to be a 30 minute diversion before the "meat" of the session (involving cleanup after the fight, dealing with government agents, etc) that I realized that not only was I going about things the wrong way, but that computer games were training players to ensure it was the wrong way.

The Problem

    Consider if you will the way a computer game has to approach a mystery or a puzzle. It's not intelligent, it can't really improvise, and it only "knows" what has been programmed into it. As a result, it's assumed that the game is right, and the players have to figure out what the game wants. There's little to no room for improvisation or clever use of abilities, so players expect to hack at the puzzle until they find one of the pre-programmed solutions (at least some games allow multiple solution paths these days). There's a bag of tricks they learn, things that have a reasonable likelihood of getting the right answer, or at least a clue. But the main advantage the player has is that they can try, try again.

    Now, contrast this to a tabletop game. Players trying to brute-force a puzzle will eat INSANE amounts of time running through the Usual Methods. And if you're running tabletop through a MUSH or chatserver (as most of my gaming is these days), multiply the wastage by a factor of ten or so. If you have a single-solution puzzle, players bringing computer-game-influenced problem-solving strategies to the table will slow things to a crawl as they search every desk for passcodes, break open every box looking for clues, etc.

    Still, even without the computer-gaming influences, solving a mystery or a puzzle in tabletop can be an exercise in frustration for all involved. And this is because the fault lies not with the players for being too computer-gamey, it's because the fault lies with the gamemaster for being too computer-gamey!

    Single-solution puzzles work fine in novels, because the writer decides when the protagonist figures it out, and the reader can try to beat the protagonist to the punch. They work quite well in computer games, which are essentially "Choose Your Own Adventure" novels with more sophisticated branching, since players who aren't as quick on the uptake can still hammer away at the puzzle until they find the solution. But as soon as you combine players with the desire for resolution within a reasonable amount of time, the approach is utterly maladaptive. Or, to translate, it sucks.

    A gamemaster trying to run a single-path or limited-path puzzle/mystery scenario is making a mistake. The setting is not conducive to brute force hammering, but players rarely like it when they become puppets of the GM in an "authorial revelation" style of solution either. They want to find the answer themselves, but not spend all night on it. Trying to get them to tell your story of their own volition rarely works well. This is especially hard in a game where characters have a lot of options, like a superhero game or a high fantasy or wild sci-fi game.

The Solution?

    So, other than never presenting players with puzzles (which gets boring, and frustrating for the ones with detective-type characters), the solution is to take advantage of the other ways in which the tabletop environment is different from the computer game environment:

    Not all gamemasters are good at this, but it's a skill that can be developed with practice and preparation. Rather than chart out a single path the players need to find, instead generate a more general awareness of the world so that you can tell if a solution the players propose is reasonable. Nail down as little as possible, and be prepared to change even the identity of the prime mover of a plot if things start trending that way. Yeah, it can make long-term planning hard, but players make long-term planning hard. It's their nature. This way you just accept that you'll need to be fluid and plan for it.

    Rather than decide how an obstacle is to be overcome, like a computer game would be, just set up the obstacles and let players propose their own ways around it. Have some idea what's possible and what's not, and assign difficulties based on that. Rather than plotting out a scenario as a series of predetermined solutions they need to find, set it up as a series of obstacles they need to overcome...however they might do so. You can even lead them in your preferred direction by rewarding their solutions with clues and pointers. By giving them more control over the short-term solutions, you build up goodwill and they may not mind as much if you're railroading them on the long-term plot! Because you've let them contribute more to the solution than just rolling dice, it's not as blatant when you do decide you need to drag them in a given direction.

    In most games, there's a way to present players with a sliding difficulty scale, either for skill use or general attempts at using their powers and abilities. If the players present a really intelligent solution to an obstacle, allow for an easy roll (or even just give it to 'em automatically). The less plausible the solution, the harder the roll. And "has no right to work" solutions may be given an impossibly high target if not outright turned down.

    Sure, sometimes a proposed solution seems wrong because of knowledge you have but the players don't. But if the player succeeds at their roll, they were right after all! And if they fail, then you already know why they failed. This is where the flexibility part comes in handy, because you may need to come up with an explanation on the spot for why their harebrained plan actually worked.

Examples

   Immediate Obstacle: Locked Door

    There's a very sturdy door between the PCs and their objective. It's too strong to just kick in, and the security system is fairly advanced. There's no other apparent way to get where they need to go.

• Computer Game Solution: In one of the desks in a work area elsewhere in the building is a notepad with a few strings of numbers scrawled on it by an employee with bad security practices. One of the numbers is the passcode.
• Dumb Solutions: A strong character might suggest hitting hit harder. A fast character might try hitting keys at random on the theory that eventually they'd get the right combination (but most keypads will lock out after a certain number of failed attempts). A ninja type might try to bypass the door by going into the ventilation system even though your setting doesn't pander to cliches like that. A teleporter might try to just teleport blind past the door. Allow 'em to try, but set the difficulty as high as the system allows.
• Reasonable Solutions: Dust the keypad for prints, reducing the possibilities to just the numbers that have been pushed, so that the "try all combos" plan becomes less stupid. The strong character might decide to have a go at the wall next to the door, on the grounds that doors are often "overkill-strong". Stuff like that is still unlikely, but should be allowed a decent chance of success.
• Sideways-thinking Solutions: Take advantage of any details the gamemaster didn't establish and declare them to be favorable (some systems have an explcit rules set for this kind of trick). "Fortunately, building codes require that all doors open up in the event of a fire alarm. I'm going to try to spoof a fire alarm signal just to the door." The statement may not be true, and even if it is true the builders might be in violation of building codes. But it's a clever idea, and should be given a pretty good chance of success. And if the roll is botched anyway? Well, then the building is in violation of building codes.

   Long-Term Obstacle: Find the Mastermind

    A mysterious voice has threatened great destruction if his demands are not met. The PCs have a week to find and neutralize his immediate plans, and then probably wants to hunt him down and stop him from trying it again.

• Computer Game Solution: The mastermind has been established before the game even starts. The players need to find the right combination of lesser thugs to beat on and follow the trail of clues to the pre-determined location of the villain's lair, then be ready to face the minions he's already got in place to defend himself.
• Dumb Solutions: Beat up every criminal you find and hope someone knows something. Go leveling accusations at the Usual Suspects (assuming the campaign's been around long enough to have recurring threats), hoping one of them is the right one and will get sloppy. Call the villain's bluff and hope it is a bluff.
• Reasonable Solutions: Use the nature of the mastermind's threat to narrow the field (i.e. if it's an atomic bomb, figure out who might have the scientific background to make one, etc). Offer yourselves as hostages, and hope the mastermind takes the bait and brings you inside his lair...then try to escape.
• Sideways-thinking Solution: Players decide who the mastermind is, and start doing the legwork to prove it, as if they'd guessed right on the computer solution. Of course, a failed roll somewhere (preferably a roll the GM made secretly) could result in them barking up the wrong tree anyway.

Conclusions

    Obviously, this is just a starting point. The very nature of the "solution" means you can't plot it all out in advance, so there's only so many examples and suggestions I can make before the law of diminishing returns starts to kick in. But the core point should be fairly clear: don't try to run a mystery like you'd write a novel or program a computer game. Take advantage of the interactive nature of tabletop gaming and let the players participate more fully in their own adventure. And if you do it right, they'll even think you meant things to go that way all along....