Why Run a Sandbox?

This post is the second in a series of posts on sandbox play in TTRPGs. The first post, on why I love playing in sandbox games, can be read here, but doesn’t need to be read to understand this one.

The past few TTRPG campaigns I’ve run have been sandboxes. I love playing in sandbox games, but I love running them even more. For me, running sandbox campaigns means low prep, collaborative storytelling, and getting to be surprised on a regular basis.

Before we continue, it would help to define what I mean when I say “sandbox”, because the term doesn’t have a universal consensus in TTRPGs. When I use the term sandbox, I am referring to a style of play that features the following elements:

  • Player-driven adventures: The GM lays out the toys in the sandbox (the world, NPCs, Factions, problems/adventure hooks, etc.), and the players decide what to build with those toys (the shape of the story).
  • High player agency: Players are free to act and solve problems in whatever way they see fit, or not at all. There is no prepared story that the GM has created for the players to follow along with. Instead, the story is created by the interaction of the players and the rest of the world (played by the GM).
  • A living world: The world must react to the players’ actions, and to the players’ inaction. Regardless of what the players do, the NPCs and factions should have their own goals that they are working towards in the background, which may or may not be interrupted by the players’ own machinations. The world should not revolve around the players, but must take their presence into account. The world must always be changing, not a static background that only moves when a light is shined on it.

At first glance, this list may seem to heavily favor players, at the expense of the GM. I have run a number of sandbox campaigns following these principles over the years, however, and I can promise that they are just as enjoyable to run as they are to play in.

Easy Prep

As a GM who chronically under-prepares, the best part of sandbox play for me is that it requires very little preparation. Before a campaign, I prep a few important things that will make my life easier later:

  • A few basic facts about the culture and religion of the people in the area where the campaign will be taking place (usually just one culture, to start)
  • 2 or 3 factions competing with each other
  • A handful of important NPCs that will introduce the players to the setting

When players are creating their characters, I make sure to try to tie them to these setting points, or even better, I work together with a player to complete this prep. This makes my life easier, and it also helps the players feel invested in the world from the start, which is an important part of making sure a sandbox campaign starts off on the right foot.

I expand on these initial prep points in between each session, on a very “as needed” basis. Rarely I’ll need to add a new faction or culture; most often my prep for each session involves the following:

  • A few important NPCs (including padding out minor NPCs created on the fly in past sessions)
  • Running a faction turn, if it’s time for that
  • Preparing any locations that I know the players are heading to

And that’s all! Sandbox preparation is mostly iterative: NPCs and factions prepped at the start of the campaign, or in between sessions, should be constantly reused. Anything that isn’t used in the session I prepped it for can always be reused, so the time is never wasted. With just these few things, and the accumulation of all that prep as a campaign continues, I have everything I need for a good 2-4 hours of play.

It should be noted that this style of play and preparation does require a decent bit of improvisation, primarily with creating new NPCs. It helps to have a system that supports quickly creating NPCs and managing factions, such as my game, The Years of Adventure, or an even lighter system like Cairn.

Free Roaming Cats

I’ve often heard of GMing jokingly referred to as “herding cats”. The implication is that the GM needs to herd the players along a prepared story, which is incredibly difficult because the players have their own (potentially conflicting) ideas for how the campaign should go. In a sandbox campaign, the GM never needs to herd. They can follow along, giving the cats (players) new toys or obstacles to play with as they go.

This is a win for both the players and the GM. From a GM perspective, I’ve found that a huge burden is lifted off of my shoulders during a session when the players are free to guide the story. As a GM, I already have too many things to keep track of. Removing the burden of corralling the players into a prepared story let’s me focus on the things I really enjoy about GMing: playing NPCs, describing the world, and empowering my players to tell their own story together.


When I GM sandbox campaigns, I am constantly being surprised by my players, and it is wonderful. Going into a session, I never know what the story is going to be. I have all the toys we need to make the story (the locations, the NPCs, the factions, etc.), but my players decide how to use those toys, and it’s almost never in the way I suspect. This point is likely best conveyed with a short example.

In a campaign I’m currently running, the PCs are attempting to overthrow a tyrant lording over a chain of islands. They recently arrived on the outskirts of a coastal city ruled by one of this tyrant’s underlings, with the objective of removing all of the tyrant’s forces from the city. After “harassing” some tax collectors in the countryside, the party decided that they needed to get into the city itself, ideally without being detected by the guards who might now recognize them after their exploits in the countryside.

I had prepped the city, and knew that there were a few possible paths the party could take. They could try talking or sneaking their way through one of the gatehouses. They could try scaling the wall surrounding the whole city, likely at night to avoid detection. The city is coastal, so they could try entering through the harbor by ship (although they would need to find a ship, as their ship was left behind to avoid a blockade). All of these actions would be made more difficult by the fact that the city was under lockdown, with a curfew implemented after sundown, a direct result of their attacks on tax collectors.

Instead, the party remembered they had a magical scroll they had collected on a previous adventure, which would allow one of them to turn into a whale. They went off to a deserted stretch of beach south of the city, one of them turned into a whale, and the rest hitched a ride on his back all the way to the harbor under the cover of night. Once there, the curfew-enforced empty streets proved beneficial, as the party only needed to worry about avoiding a few patrols as they offloaded from the whale and made their way to a safehouse that an NPC they were working with had prepared.

I could never have predicted this would be their solution to getting into the city, and I’m so glad they had the freedom to come up with a creative solution. Of course, this is not necessarily unique to sandbox play, and a situation like this could happen in a more linear/GM-plotted game as well. The difference is, in a sandbox game, these types of surprises happen all the time. Every session is full of surprises, because the players have total freedom to surprise the GM with clever solutions to the problems they have chosen to solve.


Sandboxes are my favorite way to run a campaign these days, to the point that I rarely find myself motivated to run anything else. Running games in this style takes some preparation and practice, but it is very possible to learn as you go, and so rewarding once everything is running smoothly.

If you haven’t run a sandbox campaign before, I hope this article has encouraged you to consider giving it a try. This is only the second part in a long series on sandbox play that I have planned. Next up will likely be an article on the basics of running a sandbox game, or one on how to prepare to run a sandbox game. Either way, it will be advice on how to make the style of play described here actually come to life at your table.

If that sounds interesting to you, I’d encourage you to sign up for our email list to make sure you don’t miss any of the future articles on sandbox play, and all the other exciting things we put out.