Last week, we discussed the advantages of the “railroad tracks” approach to designing adventures. This week, we’ll unpack another approach – the Sandbox method.
Where the railroad gains its momentum from the plot, the sandbox is about exploring an open world (though just how open it truly is might depend on the story), and letting the players natural curiosity – or the various storylines in play – pull them in different directions.
This isn’t to say that there can’t be a strong plot in a sandbox campaign, but sandbox stories do tend to be structured a bit differently. For example, in a sandbox, I generally consider that certain events will happen at certain times, depending on whether or not the players interfere. But the difference between the sandbox and the railroad, I believe, come down to the choices.
In a railroad plot, you know the dramatic through-line of the story – there is usually an antagonist, or perhaps a significant event, that must be overcome. With the sandbox, that story is more fluid. As reference, I like to look at my two favorite pieces of actual-play RPG entertainment: Critical Role, and the Star Wars arc of the Campaign Podcast.
On Critical Role, there is a clear and established world, developed over years of play on- and off-camera. The characters have many choices before them at any given time, and even when offered a mission, the party can easily say no, or agree but then reconsider – the plot does not stop if the party decides to follow one objective over another.
The Campaign Podcast, especially when they played in the “Star Wars: Edge of the Empire” system (I'm behind on their "Skyjacks" episodes so I can't comment as much on that structure), was also structured by arcs. These arcs were clearly defined by the planets the group visited. Usually the group had a specific objective, such as “buy armor” or “sell contraband,” as well as general objectives like “avoid the Empire” and “find new Rebellion contacts.” However, due to the nature of their interactions (and basic story structure), their situations usually became very complicated very quickly. There were always a few episodes of general shenanigans, but Game Master Kat Kuhl always managed to bring the arcs to a satisfying finale, while also paying tribute to the silliness at the beginning of the arc. The Mandalore storyline did an especially good job of tying up loose ends that had been building for over a dozen episodes of nonsense and hijinks.
The biggest concern with a sandbox adventure is preparation; since your players have pretty much free run to go wherever they’d like, it’s easy to assume that it takes a lot more preparation to write a sandbox adventure. To a degree that’s true, especially if you want your adventure to be a truly open sandbox.
However, bear in mind: your players can only open the doors they see. If you give them five different storylines to explore, then yes, they will have a ton of options in front of them, and you’ll need to do a lot of prep. However, if you only give them two possible storylines, that’s still enough to offer your group a choice. If you give the party two very different adventure hooks – for example, “Help us clean up a town full of goblins to the south,” or “Help capture an arsonist in a village to the north,” your party has a decision to make, and they know their decision will have consequences.
What defines a sandbox adventure, at least in my mind, is agency. The party is in control of what happens in the story, because they aren’t following a pre-planned adventure. Now, to be fair, an arc can have railroad tracks – the opening arc of Critical Role was a pretty standard dungeon crawl, just on a grand scale, sprawling throughout the Underdark and dealing with various factions and monsters that roamed the caves. As far as stories, it was fairly linear – the further they went, the closer they got to the Big Bad. But there were plenty of opportunities for side quests and additional rewards, especially the closer they got to the final boss. Later arcs opened the world a bit more, and my favorite arcs tend to feature several objectives spanning multiple nations or continents, giving the players a lot of agency.
It’s also key that player decisions have consequences. If there are two towns – one beset by goblins, and one beset by an arsonist – then whichever issue the party ignores has to have some sort of repercussions. Perhaps, while the party was off dealing with goblins, the arsonist is captured by the city watch… but his actions leave many guards injured or dead, which leaves the town defenseless. Then someone else is able to invade, and either take over the town or leave it in ruin – all while the players are off following other plot threads.
It’s important to make it clear that these decisions aren’t meant to punish the party. They had to make a choice because you made them make a choice. After all, you could always put both towns within a few miles of each other, making it easier for the party to try to accomplish both tasks. The fact that you split them up means you knew the party would have to choose one – if every decision the party makes has a negative consequence, the party will lose interest in the choices you give them. Then, the next time they have to choose between two towns in danger, the party might ignore both problems, assuming you will just screw over one of the towns anyway.
This leads me to another danger of a sandbox – it’s very easy for the plot to get lost. In my Princes of the Apocalypse campaign, my party would chase every side quest and barely interfaced with the main plot. Did they not know what options were available to them? Or did they feel paralyzed by choice? That’s something I have to address whenever I run a sandbox adventure; I need to ensure that the party knows where they can go next, so they don’t feel I’ve dropped them in the middle of nowhere without a compass.
As for the world itself, both Critical Role and Campaign play in big, fleshed-out worlds (especially when Campaign was set in the Star Wars universe), but remember that all of that world-building happened slowly. In the first Star Wars movie, we saw only two planets, and neither of them were even named in that film. (That’s right, Tatooine isn’t even referred to by name in A New Hope. That came later.)
When you’re building a sandbox, it’s the same as with any adventure: start small. Let the world fill in around the players. Sure, you can have a rough list of ideas for the world – maybe a city name, a few landmarks – but you don’t need to worry about any of that stuff yet. Start with your initial story, the environment your players will definitely be dealing with, and the environments your characters might deal with, depending what they do.
For my Valkyries campaign, it took a long time for the group to realize just just how much of a sandbox story they were in. For so much of the game, everything they did followed a pretty linear progression, but there were actually several potential side adventures they missed, because they didn’t explore certain places I expected. And that’s fine, because those adventures weren’t fully written, just story hooks. If they went to the blacksmith, I knew what adventure he would give them; same goes if the party went to the general store. And those adventures were still waiting in the wings, in case the party went back to that town once their current adventure was over.
The trick to managing a sandbox is to start with small nuggets of ideas, with a sense of where you will take the story if the adventurers pursue it. Don’t make more work for yourself than you need to – that’s where you give yourself gray hairs by trying to write up every possible plot the party might pursue.