Archive for February 3rd, 2010


Sandbox Dungeon Master’s Toolkit: Nudges and Libertarian Paternalism

A successful sandbox game has to balance extremes. Too much restriction and you get the Straight Line Dungeon where player choices barely matter. Too much freedom and you get the Hall of 10,000 Identical Doorways where the players have no basis for choosing one over another. When you mention that one doorway has a trail of bloody footprints leading through it, as a DM you’re using a nudge to help players navigate between these extremes.

Although I’m unable to resist the lure of discussing the theory of nudges, let me avoid burying the lede by putting this d20’s worth of cool nudges for sandbox play up front:

  1. An under-detailed dungeon map showing major threats and implied objectives. (#1 – #5 need to find their way into player’s hands to be useful!)
  2. A scavenger hunt list giving sub-goals that need to be completed, some of which specify what and where
  3. A world map on which the style used to write the place names tells them something about what’s there
  4. A city map on which “special locations” are distinctive from the rest of the mass of anonymous, of no note, buildings.
  5. A cross-section map that shows intriguing dungeon levels and their interconnections without showing how to get there from here in a top-down view
  6. Chalk on dungeon walls or blazes on trees showing where other adventurers have been
  7. A trail of small-value gems has been laid down in a regular, deliberate-looking pattern
  8. A trail of coins in an irregular pattern, as if someone was carrying a bag of loot with a hole in it
  9. Mule-pulled wheel ruts dug unusually deep, as if a wagon was laden with some very dense metal
  10. Players find part of the Rod of Law, and it points towards the other six parts
  11. Players find a wand or magic sword that detects treasure
  12. The sound of screaming or cries for help (#12 – #15 are assumed to be coming from a particular direction)
  13. A hubbub of voices and laughter
  14. The smell of baking bread
  15. The sound of a hammer striking an anvil
  16. Adventurers’ corpses, lying in a heap as if struck down unexpectedly, still carrying their gear including bulging sacks and glowing swords
  17. Adventurers’ corpses, sprawled as if running in terror from a certain direction, stripped of their gear with PC-level thoroughness (gold teeth extracted, bellies slit to check for swallowed gems), perhaps with bloody footprints showing which way the looters went
  18. Adventurers’ corpse, carrying directions to a treasure divided into steps. (Careful study of the steps that the former owner would already have completed to get to this point where his body was found shows that either he made a mistake or corrected a persistent error in the directions as written).
  19. Monsters or NPCs fleeing from a certain direction, dropping their goods in order to run faster
  20. Monsters or NPCs rushing towards a location, leading mules with empty saddlebags, mining equipment

Let me also add as a matter of practical advice that nudges like this are especially useful for one-shot sessions of adventure. It doesn’t need to be the case that sandbox-style play is poorly suited for convention games, as folks wind up suggesting in this EN World thread. What is true is that campaign-length sandboxes tend to give players more room to develop their own nudges toward adventure and information about which options will lead where they want to go. (A long-running campaign also has enough momentum to roll past a few sessions in which the players’ choice leads to boring events.) When running a sandbox for a newly-formed group, it’s a good idea to throw in some nudges – just a couple if the game is going to run for a single session, or six plus if it’s the start of a long campaign that will have time to try out and maybe reject several of them and explore the connections between each of the starting nudges.

How are these nudges different from a traditional adventure hook? For starters, an adventure hook usually combines motivation and direction. Here I assume a sandbox style of play in which everyone supplies their own motivation and  is on board with a player-driven approach, but may need some help deciding which way to drive.

More essentially, if you don’t take an adventure hook you miss out on the reward with which the hook is baited. (Players who are used to a strongly directed style of play may also suspect that not taking the hook will mean no adventure happens.) A good nudge suggests a course of action without imposing a penalty if you don’t take it, although some of the ones above meet that goal better than others. Nudges also leave open many courses of action: going through the doorway of the bloody footprints, or specifically avoiding it, or going through the next doorway over and trying to circle around are all good responses.

OK, now let’s get conceptual! When we talk about restriction and freedom, we should remember that in theory, any tabletop RPG offers unlimited freedom. Unlike a computer game, it’s possible to do anything that you can imagine, so it’s useful to think of restrictions as the costs of different choices. The cost of an absolute refusal to let your choices be limited by what other players want to do is that they’re likely to stop playing with you.

If you’re in the Straight Line Dungeon but don’t want to go A->B->C like it’s pushing you to do, you could always go back to town, hire and equip a team of dwarven engineers, and spend months tunneling to create your own path. However, the cost of doing this is high enough that most players won’t even consider it as a choice when it’s so much easier to go down the rails laid out by the dungeon designer.

The Hall of Identical Doorways gives you 10,000 choices, each with an equally low cost. But when there’s no way to tell what any of the doorways lead to, the cost of making a meaningful decision is high. By the time you’ve finished scouting out ten thousand different options, you may be wishing for the directedness of the Straight Line Dungeon!

The awesome thing about a nudge like a trail of bloody footprints is that it highlights one possible choice without raising the cost of making a different decision. It’s a lot easier to ignore the ominous clue and check out a doorway at the other end of the hall than it is to tunnel through solid rock, and it’s easier to interact with a choice-rich environment when you can make meaningful decisions about things like whether bloody footprints should be sought out or avoided.

This use of the term “nudge” comes from the book of the same name. Its authors, Thaler and Sunstein, advocate “libertarian paternalism” that focuses on understanding  choice architecture (of which dungeons are a great example!) to help people make good choices by their own criteria. I this is a great motto for reconciling the libertarian desire of a DM in a sandbox game to give players maximum freedom of choice and let their actions drive the game & the paternalistic goal of making sure everyone at the table has fun.

A dogmatically libertarian DM thinks the Hall of 10,000 Identical Doorways is the ultimate dungeon because “more choice is always better”, and so the DM has carefully avoided any details that might steer the players in one direction or another and thus reduce their freedom of choice. The problem here is that it’s hard to believe that each of those doorways lead to equally interesting places. (In fact, the sheer number of choices may make players suspect that the DM is about to pull a bit of illusionism, such that whatever one they choose will lead to the same location.)

A dogmatically paternalistic DM thinks the Straight Line Dungeon is the ultimate because “presenting sub-optimal choices is bad,” and the linear lack of alternatives saves the players from the mistake of going anywhere that doesn’t deliver the maximum fun the DM has planned. The problem here is that the DM may not know best what will be fun for the players. For many, the process of making choices on the way is more important than reaching the goal, and no amount of awesome pay-off will make up for the frustration of being forced to go there in the first place.

If you’re a libertarian paternalist DM, you acknowledge that you have much more information about what the sandbox contains than the players do, and this lets you predict some choices which are likely to lead to more fun than others.  You thus structure the choice architecture by using nudges to draw attention to the exits from the Hall of Doorways that you think lead in the most interesting direction.  But because you know that the players have the best information about what they enjoy, you avoid raising the cost of making a different decision. The other doorways aren’t filled in with rubble that needs to be tunneled through, they’re just not highlighted.

One important point that Thaler and Sunstein make is that all choice environments have to be structured, and within these structures human psychology is going to make some choices more likely than others. Reading Nudge is highly recommended for both a better discussion of choice architecture and examples of much more subtle architectural nudges than the ones I give here (some of which may not count by their definition at all). For example, if 99,999 doorways are on the sides of the Hall, and one is on the opposite wall, you’d do well to put the thing you think is most interesting beyond that one; likewise the doorway to the right of the entrance if the Hall is circular.

Note that nudges are closely related to signposts, and can be a tool for introducing the adversity James discusses as vital to a sandbox.

Past Adventures of the Mule

February 2010

RPG Bloggers Network

RPG Bloggers Network

Enter your email address to subscribe to this blog & get email notification of updates.

Join 1,054 other followers