Question Keith #8: Playing In The SandboxWritten by Keith Baker - Published on March 29, 2012
Zachary asks: Do you ever run unscripted sandbox style games? Any hints for pacing in such?
I prefer a sandbox style of play, and when I’m running a long-term campaign it’s my usual style. Even when I’m running my traveling game, I prefer a semi-linear style where the players are free to move in many different directions to a more concretely linear format.
By way of example, one of my early campaigns began with the adventurers being shipwrecked in a Bermuda Triangle-like area in another plane of existence. The players had a week of game time to find a way to get back to Eberron before the planar convergence ended. The area around their crash site included a number of shipwrecks from different eras; a crypt infested with aberrations; and a forest filled with lycanthropes. Meanwhile, they had their own shipwreck and the other survivors to deal with; some of these people were essentially commoners who wouldn’t survive without assistance, while others had skills and schemes of their own.
Within this structure, the player characters were united by a common goal (return to Eberron) and pressure (one week to accomplish it). But how they chose to pursue that goal was entirely in their hands. They could head straight for the forest. They could methodically explore each of the other shipwrecks. Or they could have spent the first three sessions interacting with the other survivors and creating a makeshift society.
On the one hand, this may seem like far more work than running a linear adventure. However, the individual pieces are relatively small compared to creating a full dungeon. Essentially, instead of mapping out a dozen rooms, I created a dozen scenes; it’s simply up to the players to determine the order those scenes occur. As the players have a common goal, there is also a certain hidden structure, because as they find clues about how to go back to Eberron they’ll find reasons that they need to visit the other locations. There are four places they have to go, it’s simply that uncovering the reasons for this is part of the adventure and even then they still decide the order – so it really feels like their story.
Depending on system, balance can be more challenging in a sandbox game. Given that you don’t know the precise order adventurers will encounter the scenes, you don’t know exactly when they’ll level or how many resources they’ll use up. In a linear dungeon you know they’ll go 1-2-3-4-boss fight. If your sandbox allows them to change that order, you need to decide how you’ll handle it. One option is to be very flexible with the values; be prepared to scale down a strong encounter if they tackle it at a lower level than expected, or to pump up a weak encounter if they come to it at a higher level. And of course even in a sandbox game you have controls. Perhaps the adventure won’t go 1-2-3-4… but you can still set things up so there’s no way to get to that ‘boss’ encounter without overcoming specific challenges. In a murder mystery, you may provide players with multiple paths of investigation, but they’ve still got to put together enough clues to identify the murderer. A sandbox places more control in the players, but it’s still up to you to have a satisfying narrative structure; stumbling onto the wizard performing the doomsday ritual by accident just because you chose door number 3 isn’t especially interesting for anyone.
As for pacing: One of the points of a sandbox game is to let the players have considerable control over the pacing. It’s an opportunity to do what they want to do. However, you’re the one who has to lend a hand if they get bored or confused. If they’re having fun poking in every corner, great; but if they just don’t know what to do next, you need to be ready. Thus, when creating a linear scenario I develop a number of static scenes (in the example above, the forest, shipwrecks, and crypt). But I also come up with a number of mobile scenes that can occur anywhere. A roc attacks. One of the shipwreck survivors has a dramatic and dangerous scheme. A storm filled with hostile elementals comes up (potentially forcing the PCs to seek shelter in a shipwreck or crypt). The party stumbles onto an exiled lycanthrope, alerting them to what’s going on in the forest. You don’t need a lot of these; just a few strong scenes that will get things moving if the players aren’t having fun and that will point them in the direction of something interesting.
One of the most important things in running a sandbox game is that you really need to be ready for the players to go off-script. The point of a sandbox is to let players choose their own paths; sooner or later, they will come up with ideas you just haven’t thought of. In my example, they could have come up with a way to get up to the dimensional portal their ship had come through (established as being in the sky)… or they could have decided that they didn’t WANT to go back to Eberron, and were in fact more interested in making an entire campaign out of exploring Lamannia. In my opinion, part of the point of running a sandbox game is that players should have this freedom. Rather than trying to force them in the direction you had in mind, it’s your job to run in the new direction. You may have to make up new encounters or challenges on the fly. But in the end, it means you get to see the story unfold in a way you’ve never expected. I have an adventure I’ve run 55 times, and it’s STILL fun for me to run it, because I never know exactly how it’s going to turn out; it’s always a thrill to see what the players come up with.