Anyway, that's by the way.
The game on Saturday wasn't great. Our characters were agents of the Inquisition, seeking out signs of heresy on a far-flung outpost of the Empire. So, we arrived, made our introductions, and started investigating the strange goings-on.
The investigations didn't go well. We spent several hours absolutely convinced we were missing something desperately important, because there seemed to be absolutely no connection between events. There was a sense that we urgently needed to put the mystery together... and absolutely no clues as to how we should do that.
At length, we found the heretic and dealt with him, but we did this via the cunning expedient of just waiting around until things got out of control, and then just mopping up.
It turned out that we hadn't carefully avoided all the clues that would have helped us solve the mystery sooner. It turned out that there was no pattern to events, and that it was just a bunch of stuff that happened. And, to be honest, we caught the bad guy more because we were just itching to cause some trouble, rather than because he actually hame himself away.
The key problem here is in the way the adventure was written (it was a pre-gen adventure, so not the GM's fault). And the crux of the problem lies in one of the key differences between fiction and a game.
When Sherlock Holmes investigates as crime, he walks into the room, finds all the clues that the author wants him to find, comes to the conclusions the author wants him to come to, goes to the next location, and repeat... Because the author controls everything, he knows both who committed the murder, why and how, but he also controls all of Holmes' thought processes, the steps of his investigation, and so on. Moreover, he can give Holmes as many or as few clues as needed, include red herrings or irrelevancies, or whatever else, safe in the knowledge that Holmes can just jump to whatever conclusion is needed, no matter how flimsy the evidence was.
In an RPG, this just isn't the case. There's no guarantee that the PCs will even look for the clues, much less that they'll look for them in the right place. Even if they do think to look where clues are to be found, there's no guarantee that the dice will be kind enough for them to actually find the clues.
Even once the PCs have found clues, it is by no means certain that they will have all the information they need to work things out. And there's no guarantee that the players will come to the conclusions that were expected, especially given the penchant of players for latching on to irrelevant details, and especially if there is a red herring in their somewhere.
Unfortunately, while most adventure writers realise these crucial differences between fiction and games, they tend not to be very good at adapting for them. Too often, adventures can become stuck because a single vital clue is missed, or because the author expects the players to follow a particular chain just as Holmes would, or because the author decided to be clever and put a false trail in the adventure.
I don't have a lot more to add here, not least because most people who read this blog have little interest in writing mystery adventures for RPGs (but also because most of what I know I've lifted from the work of other people on the subject). Still, here are some handy rules of thunb:
- Make clues easy to find. Generally, if the PCs look in the right place, they should find the clues.
- Have a number of 'floating clues' - if the PCs just generally investigate, or "ask around" or whatever, let the roll, and on a success give them a clue.
- Give at least three clues for each conclusion you want to PCs to reach. This assumes they'll miss one, not realise the significance of a second, but finally 'get it' on the third.
- Don't build your mystery as a chain of events for characters to follow. But then, that should really be true of every adventure!
- Don't include red herrings. The PCs will find their own irrelevant details to obsess over, will come to all sorts of wrong conclusions, and will generally do this bit for you. There's no need to muddy the waters further.
- Provide some sort of 'escape hatch' for the GM to use for those cases where things just go horribly wrong. Otherwise, the players are likely to get really bored once they find they simply can't solve the puzzle.
And that's pretty much that.