I was listening to an episode of Worlds Beyond Number the other day, and I was surprised by how many dice rolls Brennan Lee Mulligan was calling for as they played. It was something like 5 rolls within a single “scene”, even within a single moment, and each roll lead into the other–a success felt hamstrung, and a failure felt soft. Overall they came together to tell a clear story of a moment of peril, but each one on its own felt inconsequential. Which is weird, because I usually come at rolls in games as if each one should matter deeply.

Let’s just get this out of the way, but I think BLM is a fantastic GM–certainly more versatile and expressive than I am. BUT… his approach to dice was weird to me.

My background is more Powered by the Apocalypse, Forged in the Dark, Burning Wheel/Torchbearer and other indie games, though I got started in games with 3/3.5e way back when. But in these more indie-aligned games, there’s an evergreen piece of advice coined by Vincent Baker to “say yes, or roll dice”. Which means that if the consequences of failure aren’t interesting, or don’t add to, or even worse distract from the story being told together, just say yes to whatever the other players are proposing. And when finally the consequences of failure are dangerous or interesting, then we go to the dice to find out what happens.

The corrollary to this is that every roll should matter and have stakes.

So it felt weird to me that BLM was rolling so many dice, and undercutting the stakes of each roll, in such short succession. But clearly it was working for their game–the players trusted it, and the story that came out the other end is fantastic (and ultimately that’s the most important part–they had fun and told a cool story). So this post is maybe solipsistic, but I want to explore why it felt so weird to me.

The problem I think BLM was trying to solve

One of my biggest problems with d20 games, or really any that relies on the outcome of a single die for big important questions, is their swinginess. Even if you have a target number low/high on the d20 (depending on if you’re playing a roll over/under version of the game), each roll is still truly random, and it feels bad when your character, who is supposed to be really good at something, fails.

And even worse, in highly procedural games like 5e, the time cost for each roll means that if you fail this one, you gotta wait for a long time, sitting in the dysphoria created by your heroic character failing to retrieve a teacup from the top shelf, until you maaaaybe get a chance to redeem yourself.

Obviously this chance at failure is why we play, and why we roll dice, but at the same time when you just fail, it’s basically the dice committing the cardinal improv sin of saying “no” and it feels shitty.

Genre affects this

I notice I feel this kind of thing most accutely when playing “heroic” games, like D&D from 3e onwards, where the current version of the product really pushes the “you can do amazing things” narrative, but then doesn’t deliver mechanics to support that feeling.

Games like Mothership, where I’m not playing a hero so much as a protagonist, don’t really bring this same kind of feeling up for me.

Similarly, the time cost for creating a character in 5e is so fucking big that I’m very invested in my character kicking ass by the time we start playing… so it feels doubly shit when they let me down.

Ways games try to solve this problem

Powered by the Apocalypse uses a dice curve

PbtA games all shared a core mechanic of rolling 2d6+STAT, with success on a 10+, mixed success on 7-9, and a miss on 6-. This dice curve means that with a stat of 0, you’re going to get at least a mixed success 58% of the time.

PbtA games grant authorship to players on mixed and failed rolls

The other thing PbtA does is ask players to choose what happens on certain versions of rolls.

Some games let players pick on success and mixed success, and some even let them pick on misses, depending on whether the game wants the players to be the architects of their own frustration or what. But it’s usually things like…

• pick which companion gets hit with collateral
• you have to pay a cost–what is it?
• your greatest rival shows up to cause you problems!

which all nudge the story forward. It’s nice way to make players fully engaged even when the roll didn’t go their way.

Blades in the Dark does… gives players Stress as a limited resource to rewrite the story

Rolling in Blades is a game of rolling d6es equal to your skill, and looking for at least one result of 4, 5, or 6. 4-5 are mixed successes where you get what you want but suffer some small(ish) consequences, 6 is a full success, and oduble-6es are a crit.

In Blades, each character also has stress which characters can use to…

• (before the roll) give themselves or a companion an extra die
• (before the roll) improve their effect (make their swing hit harder if it hits)
• (after the roll) say “no” to consequences the GM is going to give them on a miss

But stress is a limited resource, so players can only do these things so many times before they run out and take a Trauma. And if a character takes 4 Traumas, they must permanently retire.

I love this approach to letting players manage the consequences they’re taking while making them watching that stress meter.

Burning Wheel and Torchbearer let you use artha to add dice and reroll failed dice

The Burning Wheel and Torchbearer both give players access to Artha, which they award instead of experience points.

Tests in these games look like rolling a pool of d6es based on your skill (e.g. a Fighter skill of 4 rolls 4d6) against an obstacle (the number of successes you need). A roll of 4, 5, o5 6 is a success.

Players can spend this Artha before or after rolls to do things like:

• add extra dice to the roll
• tap their “Nature” and call on an additional dice pool in the roll, at the risk of taxing their Nature
• reroll one or all failed dice after the roll to try and get more successes
• explode 6es in their rolls (rerolling any dice that came up 6 recursively and counting up the additional successes)

I love this because the player has so many opportunities to engage directly in the roll, by pouring in their resources to try and get the success.

WFRP uses success levels…

Of the examples here, this is the one I like the least because it’s fucking clunky. WFRP is a percentile game, where if your skill is 60, you need to roll a 60 or less on a d100. And you often have opposed tests with the GM rolling percentiles against the player.

So you sometimes end up with two characters who failed their tests… but one failed “less” than the other–you look at the difference between your target number and the rolled number, and whoever got closer to their TN has the better success level.

It’s too much fucking math, but at least it lets you figure that the veteran soldier who rolled an 81 over their target of 80 (SL -0) did more damage to the peasant who rolled a 37 over their TN of 12 (SL -3).

What I think Brennan was doing

The thing you might notice from the PbtA, Blades, and Burning Wheel/Torchbearer examples is that they all use large numbers (dice pools and curves) to try and tame the swinginess of dice to some extent. They all also involve the player to some degree choosing how much to invest in the roll before and after (Blades, BW/TB), or choosing how to hurt themselves (PbtA).

So Brennan is doing the averaging thing–by calling for a buttload of die rolls, and being sure to hit character’s strength skills, he’s giving the players a chance to average out their luck over the course of a scene.

And again, it works for that group. I might hate it philosophically, but I love the result (seriously WBN is great, go listen).

Why I really like dice pools

Burning Wheel and Torchbearer might have one of my favorite dice systems. Not always what I want to play, because each roll does require seriously attention and time investment, but they’re really satisfying in terms of “table feel”.

By choosing each die in the roll by spending artha, you’re using physical tokens to represent the effort. My little monkey brain immediately translates “look I added these dice” to “damn my character is really trying their best here”.

The roll itself becomes a satisfying game, and when you’ve lost it’s either because you chose not to invest, or because chance was really against you in this case, rather than just one tyrannical d20… the story that my brain comes away with is not “fuck these dice” but “fuck… I guess fate was really against my character”.

So much more satisfying. But that’s me.

I also fucking love PbtA and FitD mechanics, for different reasons. I play them more because overall they move faster, but in isolation I think the BW die roll is the most elegant mechanical tie of narrative to numbers in an RPG mechanic I’ve ever seen.