Usage & Hazard Dice: How to Emulate Bookkeeping with Dice

In old D&D, you track the duration or quantity of items. For example, one torch lasts 6 turns; longterm spells often also last 6 turns; quivers can hold onto a certain number of arrows. When the timer runs out, your torch burns out or your spell expires; when you run out of arrows, you cannot use your bow anymore. One goal of contemporary rulesets has been to abstract these factors to avoid trivial bookkeeping during play.

One method of abstraction is through random probability. Each resource, whether the duration of a torch or the number of arrows in a quiver, is depleted at a known or expected rate. Using random factors, i.e. rolling dice, one can emulate the depletion of those resources statistically rather than manually tracking resources at constant rates of depletion.

Emulated Resource Consumption

See: cascading dice, usage dice, risk dice, delta dice.

One method of this is what has become known as the usage die. This term was popularized by The Black Hack (2016), where consumable resources are assigned a specific die size. Each time the resource is consumed, the die type indicated is rolled; on a result of 1 or 2, the die type is downgraded to the next lowest size. For example, a torch has an initial usage die of d6; if a 1 or 2 is rolled while using it, its usage die becomes a d4. If a 1 or 2 is rolled on a d4, then the resource is totally consumed. A similar method is described on Necropraxis in 2012 [1], where using any resource has a 1-in-6 chance of being consumed. However, an earlier post on Intwischa in 2011 describes using a whole set of dice in the same way that The Black Hack does; it only calls them "cascading dice" instead of "usage dice" [2], and the die is downgraded upon rolling just a 1 (not a 2). It is more likely, though, that all these methods were invented or discovered independently of each other, rather than one copying another.

As shown between Intwischa, Necropraxis, and The Black Hack, there are many ways to implement this method of abstract resource tracking. The main principle is using random statistics to emulate the duration of constant depletion. For example, if a torch has a 1-in-6 chance of being depleted, that torch will statistically last 1 / (1 / 6) = 6 rolls. Different methods of emulated depletion can be compared to one another. The blog dieheart compares the longevity of consumable items in The Black Hack versus Macchiato Monsters, which uses usage dice (called "risk dice") except that resources are depleted on a roll of 1 to 3 rather than a 1 or 2. On my blog in 2020, I compared risk dice in Macchiato Monsters to an analogous mechanic to another game by Eric Nieudan, Macchiato Micro, which uses decreasing probabilities on a d6 [4]. Hopefully these resources can help you analyze similar rules.

In situations where you are worrying about one resource, usage dice simplify resource consumption. You can just hold onto the actual physical die, and trade it out for a different one when the resource is downgraded. Then you don't even need to write anything down on paper. However, with multiple such resources, or when resources are still tracked on paper, I think usage dice become a bit extraneous. You still need to write and rewrite your resource's current usage die, and the process of figuring out if you've consumed (i.e. downgraded) your resource or not adds much more complexity than just counting down a number. Nevertheless, there is the benefit of reduced math, which is likely one of the reasons it was first incorporated in The Black Hack. You might find, then, that usage dice or some other method of emulating resource consumption will work well for your campaign's ruleset.

Emulated Procedure

See: encounter checks, hazard dice, event dice.

In 2014, the encounter die was overloaded. Brendan S. from Necropraxis wrote about how the wandering monster roll could be taken as a basis for triggering other random events [5]. He did this by indexing other results on the encounter die to different random events: finding a clue of a monster, something about the location changing, the adventurers becoming exhausted, and light sources burning out. Many of these random events were previously resources that had to be tracked, or clocks that counted down as time passed. The overloaded encounter die, later systematized as the hazard die, relies on a similar mathematical basis as emulated depletion does: you can emulate the random passage of events by finetuning dice rolls (by probability and frequency) to statistically resemble manual bookkeeping. For example, if you roll the hazard die every turn and one of the (1-in-6) outcomes is that your torch burns out, then your torch is going to last about 6 turns just like it would if you were counting down 6 turns by hand. The fundamental difference between this method and emulated consumption is that it does not rely on the moment of a resource being consumed. You don't have to remember that you're supposed to roll to consume your torch when you carry one around. Instead, it is ingrained in the basic turn order or procedure of the game.

Emulated procedure, then, allows you to emulate the passage of a wide variety of events. For D&D in particular, it encodes the entirety of the dungeon crawl with wandering monsters and resource management in a single roll, with no need to keep track of what will happen when. The roll, as it were, 'remembers' this for you. Brendan S. would expand upon the overloaded hazard die in a later post that same year, where he christens it the "hazard die" and applies the logic to other contexts: haven turns and wilderness turns [6]. Ava Islam also expands upon hazard dice, now called event dice, for her rulebook Errant, by applying an "advantage/disadvantage" system to the roll [7]. Gus L. compares the "classical" dungeon crawl procedure, where resources and other factors are tracked manually, to the "neo-classical" procedure with hazard dice which had become popular since it was invented [8].

As I have said, one great advantage of emulated procedure is that all the major factors of a game can be emulated via random probability rather than them having to be tracked individually. This means that, combined with a simple action economy ("What can I do between rolling event dice?"), hazard or event dice can encompass the entirety of your game, at least as far as how the game-world acts in response to the players. One problem is that those different events, without proper interpretation or scaffolding, can lead to unreasonable results. Even in the original post about the overloaded encounter die, Brendan S. says that results should be ignored if they are unfair (such as a torch burning out immediately after being lit, or within the first couple of turns adventuring). This is something which I've tried to resolve in my implementation, TURN, by using buffer states to delay final outcomes: for example, lanterns go dim before they burn out, or the party becomes tired before they become exhausted. Another problem is that results seem to be mutually independent of each other. When you're tracking exhaustion separately from wandering monsters, there is the chance that both events will occur at the same time. When both events are mutually exclusive outcomes on a d6 roll, that situation is not intuitively possible. Ava Islam's solution is to impose a second event die roll if the party rests, to abstract the risk of resting in a dungeon. My solution in TURN was that if the party is resting, they are considered 'busy' and will be surprised by any wandering monster on the next roll.

That all being said, the hazard dice system is a very popular method for how intuitive it can be for newcomers, with respect to reduced bookkeeping and also to summarizing a game in its entirety.

Conclusion

To speak broadly, I think emulated consumption and emulated procedure are two diametric solutions to the same problem: how can we avoid tracking numbers by replacing them with automatic, randomly-determined outcomes? We can speak of them as being structurally distinct. Emulated consumption serves as a singular operation; it does not occur rhythmically, but it is an action one does when it is triggered (by the use of a resource). It is like a step in a recipe. Emulated procedure, on the other hand, encodes resource depletion and other random events in the rhythm of the game. You do not have to remember when to erase a torch from your inventory list; the die will tell you if you have to. This makes emulated procedure something like a machine learning algorithm that randomly generates a recipe step by step. You won't know what each step in the recipe is, but at least it knows what a recipe's supposed to look like. The underlying feature, however, is that both use random probability to facilitate the game instead of manual bookkeeping. They could be used together, or one could be used without the other. I think that whether you use either one is a matter of your own preference or tolerance for bookkeeping, and your preference could totally differ between the two contexts.

Although I see the value of both, I tend to find emulated consumption too finicky and emulated procedure too random. I want to explore other methods of abstracting and simplifying rulesets which don't rely on emulating existing rulesets via random probability. These could include opt-in resource management which incentivize players to deplete resources when they call for it [10], or using less many-to-one ratios in favor of one-to-one ratios like in my post about wilderness exploration (one day costs one ration and one encounter roll; everyone gets one side action) [11]. Still, I hope this is a good overview to make informed decisions about how to make your own game run more smoothly.

[1] S., Brendan. 2012-09-16. "Abstracting missiles", Necropraxis. Abstract the tracking of ammunition by instead tracking 'ammo dice' to represent collections of ammunition rather than individual pieces. On a roll of 1 (usually on an ammo die of d6), the collection is extinguished.

[2] White, Charlie. 2011-05. "Cascading Dice: A House Rule for Tracking Ammunition", Intwischa.

[3] Brandt, Sophia. 2017-07-03. "[Let's Read] Macchiato Monsters ZERO - Core Mechanics", dieheart.

[4] B., Marcia. 2020-12-10. "Risk Dice vs Deltas", Traverse Fantasy.

[5] S., Brendan. 2014-02-03. “Overloading the Encounter Die”, Necropraxis.

[6] S., Brendan. 2014-12-23. “Hazard System v0.2”, Necropraxis.

[7] Islam, Ava. 2021-02-16. "Errant Design Deep Dive #2: Core Procedures", Permanent Cranial Damage.

[8] L., Gus. 2021-10-06. "A Structure for Classic Exploration Procedure", All Dead Generations.

[9] B., Marcia. 2022-04-26. "TURN: Post Mortem", Traverse Fantasy.

[10] B., Marcia. 2022-04-05. "Cyclical Resource Management", Traverse Fantasy.

[11] B., Marcia. 2022-06-30. "Fast Travel & Watch-Keeping Procedure", Traverse Fantasy.