As a Dungeon Master, I never make extensive use of randomization during my Dungeons & Dragons games. I usually prefer to have the story direction and the development of a game session determined by choice rather than luck. However, I recently changed my approach, and I’m really glad I did.
Critical Role Campaign 4’s new DM Brennan Lee Mulligan asks for a lot of “luck rolls” from the players, a custom mechanic that he’s been using regularly. He’ll select a particular type of die and then assign outcomes based on the result of the roll. It’s no different from rolling on any random table the game provides, really, but they are cooked up on the spur of the moment, when a player will take an action without a clear predetermined outcome.
I decided to try these at my table, not because I worship Critical Role or BLM, but because it looked interesting and, more importantly, different from my usual practice. The outcome was great, and it made me reflect on the often-debated balance between pre-determination (or preparation) and randomization (or improvisation) at the D&D table.
In a recent session, my players had just come out of a big battle that involved an entire city, including a host of NPCs they befriended, against an army of oozes. When a cleric player asked if two specific characters, a brother and sister duo, survived, I let the dice decide. I asked the player to roll a d20: on a 1-4, both NPCs died in the battle; on a 5-9, only one died; on a 10+, they both survived. Obviously, the player rolled a 4.
This led to an emotional sequence where the players found the bodies of their two friends, disfigured after the battle, still holding hands, their melted flesh joining them forever in death as they had been in life. The party decided to perform the last rites, and it was meaningful because the cleric had previously started to convert the sister to worship her god, Lathander, during the time they traveled together. She had slowly begun opening up to the faith — but then died defending her home. As the funeral rites ended, I decided to reward the players: the bodies of the NPCs were restored, their hands opened, and inside there was a Prayer Bead, a magical object containing a spell. I randomized the type of Bead, and the spell that turned out was exactly the one the party needed (and lacked the means to obtain) to release another important NPC from petrification. You can’t plan this type of stuff.
It makes me wonder if randomization and improvisation are truly the hearts of this game, and every DM should rely on them a lot more. I’m not an obsessive planner, but I like to have a strong outline before each session. I recently realized, however, that my improvisation skills were rusting, and that’s something that you always need, even if you’re the most prep-heavy DM in the world. D&D players find sadistic glee in derailing the best laid-out plans, so a good DM has to be able to think on their feet, pivot quickly and effectively, and make up stuff on the spot.
The use of luck rolls, or any other way to randomly determine the outcome of a situation on the spot, is a great way to train these skills without stepping too much out of your comfort zone. The trick is to use them for small situations that don’t have a massive impact on the direction of a session or campaign. I wouldn’t use this method to decide if the King’s advisor is a traitor working for the Demon Lord, for example, but I could use it to determine if the player characters step into the King’s chambers the moment before or after he’s stabbed to death. Will they get a chance to intervene, or will they be blamed for the assassination?
Luck rolls are also great to keep the players on their toes and make them feel like the game is alive, shaping and progressing according to their actions in real-time, rather than feeling like characters in someone else’s story. It enhances the strengths of collaborative storytelling.
This aspect has actually always been baked into the core of the game, at least in its early stages. The D&D of old was enamored with random tables, which made sense when the game was based mostly on exploration, loot, and combat. (Many games of the Old School Renaissance wave went back to this approach, with great success.) With modern D&D being very skewed towards role-playing, many DMs feel like they have to do extensive prep, lest they find themselves frozen when their players decide to skip the carefully laid-out town or ask for Random Farmer Number 2’s backstory. But that may not necessarily be true. This video from the great Bob World Builder explains all the perks of running an RPG session without excessive prep, so I suggest watching it:
There is nothing wrong with prepping, as Batman knows well. But there’s also nothing wrong with letting dice and luck decide a few things, instead of you. As I wrote here, control is a big thing for DMs. We need it to run the game, and we struggle to let go of it, even when it’s a good call to do so. My advice is: Don’t be afraid of losing control for a while, because the outcome may be worth it.
