Is Too Much Player Agency a Bad Thing?


Last night my players turned to me and asked, “Where do you want us to go? You’re the GM; what do you have prepared?” I blinked, smiled, and confessed, “I’m equally unprepared for both options! What would your characters do?” Crickets.

(Prep-lite is both a blessing and a curse, let me tell you.)

After getting burned by a few sessions where my own character had been outclassed and outwitted this summer, I wanted to avoid the dreaded “railroad” in my own game, opting instead for a sandbox-style world. I based the story threads on the backgrounds and motivations of the party, and crafted several leads for them to chase with multiple conflicts brewing in different parts of the world. But I might have given them too many choices–left the door open too wide such that no one story thread called to the group as a whole.

Here are a few of their options from last night alone:

  • Search out or raid stores of arcaneum (the stuff that makes magic work in this world) to provide the citizens of New Haven with a defense against the native golem population
  • Attempt to access the recently flooded pyramid via breathing apparatuses to neutralize the golem threat
  • Investigate the overgrown shrine to the north of the city in search of clues to deal with the golem problem
  • Leave the citizens with the large magitek battle-airship they stole and go and find other pockets of rebellion on the continent
  • Give the citizens the bird and commandeer one or both of the airships to go and find other pockets of rebellion on the continent

“I don’t know what the main plot is,” my boyfriend admitted.

“That’s because you guys get to choose what you want the main plot to be,” I said, confident that I was “doin’ it rite” cause I was following the advice of numerous blogs and GMing advice articles. Or something like that.

But somehow my answer didn’t seem satisfactory. The pacing was still wrong, and the party wasn’t as excited as they used to be. We have been spinning our wheels in the same town for the fifth session in a row, and I couldn’t come up with a compelling enough reason to kick them out–at least, none I was willing to entertain for fear of–*gasp*–railroading them.

As it turns out, there are a number of articles out there warning of the pitfalls of poorly done sandboxing. Gnome Stew points out that without a strong opposition, sandbox campaigns can feel boring. They even have a post “In Defense of Railroading.” Reddit and StackExchange each have their own takes on the pros and cons of sandboxing. It’s definitely a balancing act.

Perhaps I have been worrying too much. There’s a difference between nudging PCs in a certain direction and forcing them down a specific path with predetermined outputs no matter the inputs. If I can play up the threats in the world, they’ll feel more compelled to act, and soon.

Dan Clark (you might know him from the Star Wars RPG Beginner Games and his work on WFRP3) gave me some sage advice: “When you’re backed into a corner, have the villain barge in with a shotgun.”

What are your experiencing running a sandbox campaign and finding the right balance of “railroad” vs player agency?

4 Responses to “Is Too Much Player Agency a Bad Thing?”

  1. The Gneech says:

    As I understand it, generally speaking, 3-5 selections is optimal for any choice to feel meaningful. Any less than that and it doesn’t feel like you really have a choice; any more and choice paralysis sets in.

    Also, you need to have some useful information to choose from. “Door A, Door B, or Door C” isn’t a meaningful choice because you have no idea what the different choices mean. “Lizardfolk in the swamp, the crypt of the vampire lord, or sacking the giants’ lair” is better (“We have no cleric, the crypt of the vampire lord is OUT” etc.).

    That said, players giving you the “deer in the headlights” look when you ask “So what do you want to do?” is certainly a frustrating problem. As I mentioned on Twitter, we sometimes get around that by designating a group leader (the ship captain, for instance), but you can’t do that for every game.

    Sometimes you have to just hold up a neon sign that says “PLOT THAT WAY –>” and hope something happens.

    -The Gneech

  2. Marty says:

    I struggled with the came issue for a while. I presented a few different plot hooks and the party followed but as you noted, enthusiasm was down…


    I found a villain that they hated. Ironically, he wasn’t a true villain… More of an adversary. Once they had an “I hate that guy!” moment, they were a lot more motivated to make things happen.

  3. The original bon mot is from Raymond Chandler:
    “When in doubt, have a man come through a door with a gun in his hand.”
    More on that topic, here: http://tvtropes.org/pmwiki/pmwiki.php/Main/CHANDLERSLAW

    Player paralysis is often caused by managed expectations. A lot of people who play these games expect to have the micro/tactical challenge of dealing with the situation at hand, rather than the macro/strategic challenge of making the big decisions. Big decisions aren’t everyone’s idea of fun!

    Time pressure helps. A “sandbox” game often implies all the time in the world. If the players don’t know what to do, and nothing seems urgent, then nothing can be prioritized. Time pressure is often handled as a world-vs-player issue — “If you don’t stop the lich king, he will destroy the world!” But you can also render it as a world-vs-players-friends issue — “If we don’t find the source of all these monsters, they will eat everyone in our village.” Time pressure can also be done as a players-vs-rivals issue — have a party of obnoxious NPCs that the players don’t like, and have them hint they’re going to go do one or more of these things.

    Sandbox games do not really have plot, because they don’t have a timeline. Events wait for people to resolve them, in one place, forever. Your game should always have a variety of options of things to do, but you should also make the players feel that doing nothing will make other things happen, not always desirable. Reinforce that it’s a living, breathing world.

  4. Chris Herrick says:

    You can provide a plot without providing the means to reach the end (similar to having to cross a stream with no bridge laid out in front of you).

    I normally script my campaigns to have a number of events happening in the world (which the characters may or may not participate in, depending on choices and time elapsed) as well as a goal to achieve, but I don’t provide the path for the characters to follow to reach the goal. This approach normally works, but occasionally you have to guide the characters back to the plot.

Leave a Reply

Powered by WordPress | Designed by Elegant Themes