Sunday, May 18, 2008

Missing the Clue

I've recently picked up Robin D. Laws's The Esoterrorists. I'll write a complete review after I finish reading the entire game, but I did want to discuss part if it right now.

The Esoterrorists's mechanics are largely based around a single idea--it's hard to run mystery plots in typical RPG games. Typically, the characters must make rolls to discover clues. If they fail their rolls, then they don't uncover the clue, and the plot grinds to a halt.

The Esoterrorists (and other games using the Gumshoe system) solve this problem by guaranteeing that characters can find important clues. Your investigative skills give you a pool of points. You can spend these points to automatically discover clues (and sometimes additional details). This guarantees that the plot will continue to move forward.

Now, that's not a bad mechanic, as far as things go. It looks like a great way to run mystery games. But, I'm not sure I agree with Robin's analysis of mystery games.

True, mystery games are often incredibly frustrating. But, I don't think the dice are to blame.

Sure, it sucks when characters miss an important clue. And maybe it matters, if you're running a simple whodunnit. If the bad guy has already killed all the people he intends to kill, then one missed clue and he might get away. However, that's not the type of mystery I want to run. I prefer mysteries where the characters are peeling away layers of a larger conspiracy. Sure, they may not be able to solve a single murder--but, just wait. There will be more.

If the characters botch a single roll, don't sweat it. Just quickly move time forward to the next incident.

The bigger problem is pacing. The GM is the only one who knows the full story. Ideally, he should dribble out clues at just the right rate. They should come fast enough to keep the player's engaged in the story, without giving away the ending too early. This is incredibly hard to do. And I'm not sure The Esoterrorist's mechanics can really help here.

Here are some suggestions:

1) Give the characters clues that are hard to understand, but that open up possibilities for further research. The classic example would be some sort of occult mark on one of the victims (or bad guys). The characters have no idea what the mark means, no matter how good they roll (at best, it seems somehow familiar, but they can't put their finger on it). The clue remains unsolved, but opens up opportunities for future research or investigation.

The main goal here is to have the players proactively direct the investigation. This helps with pacing, since the players will pursue those things that most interest them. Proactive players help define how quickly the story unfolds. And, once players start acting proactively, they will usually find ways to ferret out all the clues they need.

So give them clues they can follow: A phone number ripped from a phone book, The picture of a famous celebrity, taken from a cell phone camera, An informant turning up dead. The opportunities are endless.

2) Reward the players for being proactive by providing them with additional clues. Always remember, ever scene should move the story forward. In the case of a mystery, each scene should provide one (or preferably more than one) paths that the characters can pursue further.

Also, you don't have to force the characters to roll for every clue. You can reveal clues as part of the description (or clues to clues). Let's say there's a few bloody fingerprints smeared on the window. You could mention that there is blood on the window when you describe the room. Or just mention the "dim light filtering through the smeared window", and wait for the characters to ask about it.

When someone actively investigates the window, you may or may not require a roll to identify the smears as blood (or to notice the fingerprints). Obviously (at least obviously to me) if someone asks, "are there fingerprints on the window?", just say "Yes" and move on.

Don't let yourself get bogged down by the rolls.

3) Be stingy with the clues. RPG players are typically very smart. Don't make it too easy on them. It's better to be too stingy than too generous, especially since you can always spice things up (see hint 4).

4) Don't let the plot slow down. If things start to drag, have something exciting happen. Maybe the conspiracy thinks the players are getting too close, and tries to off them. Maybe they take the next major step in their evil plan. It doesn't matter. Pick something that gets the blood pumping again and keep things moving.


Saturday, May 03, 2008

Predestined Consistency

It's often said, if you mention a gun in scene one, someone has to shoot it in scene three. We expect this sort of consistency from most forms of fiction. A good story fulfills the promises made at the beginning.

However, this is often hard to pull this off in RPGs. In old-school games you could sometimes fake it by railroading the characters, but this was rarely satisfactory. In a narrative-heavy game, the GM has even less control over the story, and that can make fulfilling these promises even harder.

I've said it before, and I'll say it again. Consistency is hard. And, as gamers, we don't have the freedom to go back and edit things. We've got to get them right the first time.

So, instead of just hoping that everything will work out, let's create a framework to set and resolve these expectations.

Let's start with Rich's first law of gaming, every conflict should move the story forward. Generally speaking, the winner of a conflict gains some narrative control over the scene. They could use this to gain some clue, earn an ally, or otherwise throw a wrench into the opposition's plans. If they fail, things generally take a turn for the worst. But, most of the time, these are immediate effects.

But, do they need to be immediate effects? What happens if we let the winner establish an expectation related to the conflict. We record this expectation, perhaps on the character sheet or in the GM's notes, with the explicit promise that it will come to pass sometime in the future.

For example, our hero decides to seduce an opposing femme fatal NPC. If he succeeds, he sets the expectation that she will switch sides at some undeclared point in the future. We don't know when this will occur, but the expectation becomes part of the story. The GM may change her behavior to foreshadow this event. Then, at a dramatically appropriate moment, she betrays the other bad guys. Of course, she may just stab our hero in the back five minutes later, but at least we've fulfilled our initial expectation.

Of course, player's shouldn't get all the fun. For example, the same hero loses a dual against a minor opponent. The gm could set the expectation that one of the characters will kill the other, turning a throw-away villain into a major nemesis.

Or, lets turn this all on its head. I always like the idea of letting losers define the cost of their failure. Maybe the player decides he must kill or be killed by his opponent. Or the femme fatal decides that she's fated to change sides. Having the players self-select these expectation will help them buy into the story.

Long time readers might notice how well this fits into my Evolve or Resolve mechanic.

Remember, these are player expectations, not character expectations. The character may not know anything about them. He may believe the exact opposite, and may struggle against the expected fate. Rather, these are tools for the players. They are the load-carrying beams that connect past and future. They help give the story structure, while also giving the players more control over its shape.

I haven't thought this out completely. And it could probably use some play testing to flush out the details. What sort of expectation can we set? What are the limits? Which are better, specific expectations, or open-ended expectations? Can other players trigger an expectation's fulfillment? Should players be rewarded for playing into these expectations? Under what circumstances can we break these expectations?

I'd be interested in your thoughts here.