Recently, one of the groups I play in started a Dungeon World campaign. With a few sessions under our belts, I can say for sure that I love this game with an excitement that I have not had since my days of playing Basic D&D. After talking to the other players and the GM about it, there are a number of things that Dungeon World does right. They are things that any GM could be doing…should be doing in their games. To keep me from having too much of a fanboy tizzy, I picked just three things for today.

Origins in the Apocalypse

I was not surprised that I was going to love Dungeon World, when I first backed the Kickstarter. After all, a few years ago you all asked me to review Apocalypse World by Vincent Baker – the genetic precursor to Dungeon World. At that time, I went on about what great GMing advice the game contained, and how that advice was universally applicable and not a feature of AW.

Three Things You Can Do In Your Game

I did not want to write a review about Dungeon World. Instead I wanted to highlight a few concepts that the game gets right, which it explicitly states in its rules, and explain how you can translate those into other systems and campaigns to bring some of that DW goodness to your own table. Here are my top three:

Make Skill Checks Interesting

I have a pet peeve when it comes to how many games handle skill checks. I dislike systems where a skill check has only one interesting outcome (most often passing). In Dungeon World, a check is made by rolling 2d6 and adding an attribute modifier (and on occasion a situation modifier). If your roll is 10+, you succeed with exactly what you intended to do. If your roll is 7-9, you succeed but often at a slight cost. At 6 or less, you get an experience point, the GM decides your outcome, and the GM gets to make something happen from their list of moves. In this case of “failure”, the GM could let you succeed while using one of their other moves to effect another part of the scene.

Example: Caldor the fighter swings at the goblin and the attack roll is a 6. The GM decides that Caldor hits, but the attack is so loud that the Ogre down the hall hears the commotion and comes to investigate (DW move: Show signs of an approaching threat).

As a player, when you roll for something every part of the roll is interesting. You almost don’t care what you roll, because you know something is going to happen. Contrast that with a Spot check in a d20 game to find a clue.

Example: Caldor the fighter looks around the room. He has 10 ranks in Spot and a Wisdom of 14; +14 to the roll. The GM has set a DC of 20 to find the clue. Caldor needs a 6 or better and rolls…clunk…a 1, for a total of 15. Sorry, no clue.

Putting It In Your Game: I talk a lot more about this in a past article, but here are some quick tips for making skill checks more interesting:

  • Only require checks when it’s interesting to the story, otherwise say yes and move on.
  • Make checks with interesting outcomes for both passing and failing.
  • Re-define failure. Rather than just say no, failure can be success with a consequence.

Play to Find Out What Happens

In Dungeon World, the GM is told to never script the outcome of a scene, encounter, or adventure. The GM is told never to presume what the characters will do and to follow through with the consequences of the players actions. This lets the game unfold in an organic fashion, which is as revealing to the players as it is to the GM.

This concept traces its lineage from Dungeon World to Apocalypse World and back to Dogs In The Vineyard, so it is by no means a new concept. Though in modern game design, this concept is not always stated and encouraged, let alone placed directly into the rules of a game. You see a lack of this in many published adventures which script the story by having the outcomes of scenes mostly predetermined, and leaving how many resources (hit points, spells, ammo, etc) are consumed along the way as the variable portion.

Putting It In Your Game: The good part is that many GM’s informally adopt this style at some point in their GMing career. If you have not yet tried this, then the way to do this is to set up scenes, but never think about how the scene will end. Mentally prepare yourself to deal with that part at the table. If your ad lib skills are not quite there, then jot down a few notes for the most likely outcomes (pass, fail, avoid, etc). Then when considering the transition from scene to scene, don’t make each scene dependent on one outcome from the scene before, instead leave your options open. Consider the following example:

Predetermined: The players must defeat the ogre to reach the lost library.

Open Ended: If the players have not dealt with the ogre when they reach the library, then the ogre will find the characters as they explore the library.

In these two sentences, the predetermined design forces the players to defeat the ogre before getting to the library. The Open Ended approach gives the players freedom on dealing with the ogre (or not) and the consequences if they don’t deal with it first.

Begin and end with the fiction

Dungeon World encourages the GM to let the rules of the game take a backseat to the narrative of the game. The rules only come into play when the narrative of the game states something that falls under the rules of the game. When that happens dice are rolled, the rules are consulted, and then the outcome determined. Once the outcome is known it is described as part of the narrative and play continues.

It sounds intuitive, but few other games are as explicit about this form of play, and many games and gamers are guilty of violating this. Here are some examples that I have heard from my own games:

  • “Can I make a perception check to see if there is anything around?”
  • “I…” (roll dice) “hit…” (roll dice) “…for 15 damage.”
  • “You pass your Streetwise check, and find out the Crimson Daggers are a doomsday cult”.

In all those cases the players and GM were just stating what they were doing mechanically, and not describing how they were looking around the room, how they landed that attack, or who they were going to talk to for more information.

Putting This Into Your Game: This one is easy to install in your game. Start by taking an index card and writing on it “Begin and End With The Fiction” and put it out on your gaming table. Then as you play, when someone (including yourself) starts by talking about the mechanics of the game, ask the person to describe what happens first, then deal with the mechanics, and then conclude by describing the outcome. This is a learned skill, and if repeated enough times it will become a habit for your group.


I am sure by now you are sick of me talking about Dungeon World, but it is a game that has a great lineage and gets a lot of things right. If you have not played or run it yet, I encourage you to give it a try. What Dungeon World does better than many games is that it’s rules create a style of play that puts the focus on the role and not the roll.

Dungeon World as a game may not be for everyone, but there are a lot of gems of good GMing that can be cribbed and incorporated into your game to make it more exciting.

Will any of these techniques fit into your game? Do you currently use any of the techniques mentioned above in your gaming? How did you develop these techniques? Have you taken anything else from Dungeon World?

If you want to hear more about my group’s experiences with Dungeon World, check out Episode 56 of Misdirect Mark.