Today I want to take a moment to examine the dice mechanic in our upcoming games: Genship Exiles and Hwaet.
Both of these games will be powered by Rowan, Rook & Decard's Resistance Toolbox (affiliate link), and as a result there will be a lot of commonalities, but I also want to talk about how they are going to differ.
Dice in the Resistance Toolbox
The Resistance Toolbox features a dice-pool that produces one to four dice for any given roll: one by default, an additional two for matching skill and domain requirements, and a fourth that comes from "mastery"; it is generally likely that characters roll 2-4 dice, with 4 dice representing a high degree of proficiency (i.e. a focus) and 3 representing a professional capability.
When you roll dice, you pick the top result and compare the result to a chart:
|Critical Failure (double stress)||Failure (suffer stress)||Success (suffer stress)||Success (no stress)||Critical Success (inflict more stress)|
This is a fairly elegant solution. If we use Anydice, we can get a pretty quick breakdown of what is likely with a given number of dice rolled:
Chart showing the chance that a particular configuration of dice rolls at least a given value.
With one die, we see the pretty obvious result that each result is equally likely based on the number of possible results; a success is going to happen 50% of the time, and a failure 50%, and the varying degrees occur based on the number of results that occur (e.g. a critical failure happens 10% of the time).
When we add more dice, we greatly increase the chances of a success; a character rolling 2 dice succeeds 75% of the time, one rolling 3 dice succeeds ~80% of the time, and a character rolling 4 dice succeeds so often that the only real question is whether they get a success that doesn't cause stress or counts as a critical success.
A counterpoint to this is that there is a difficulty mechanic in Spire, and by extension in the Resistance toolbox, which either removes a die (if a character is rolling more than one die) or downgrades the result by one step (e.g. a critical success becomes a success without stress).
This means that these really optimistic outcomes aren't what you see in play all the time, but it is worth noting that a character who is skilled in any Resistance toolbox game is really skilled.
Technically, group actions add one die to the pool, so you could help someone and gain a bonus die, but this is unlikely to push up to five dice. If it does, your chance of success becomes absurdly high (less than a 18% chance of taking stress at all, a 40% critical success chance)
Resistance-lite (e.g. Facade and Waystation Deimos)
This is a simpler version, and can be explained more simply: a result of 1-3 is a failure, 4-6 is a success. The lowest failure or success adds a complication, the highest of either adds a benefit to the player character.
This has an even more aggressive curve when applied to the rolls, and Waystation Deimos lets characters go up to five dice!
Anydice graph of the probability distribution in Waystation Deimos
One balancing factor here is that you have complications, and Waystation Deimos has a hardcore death spiral mechanic. The basic chance of success echoes the d10 system, but the chance of getting a critical success is much greater, and the penalty for stepping down for having fewer dice is mitigated by the fact that you have a "good failure" step before you get into the painful failures.
The critical success rate in the simplified Resistance toolbox is much more extreme, however; you can see in the table that with two dice you already have a 33% chance of critical success (as opposed to a ~20% chance with 2d10), and with four dice you actually have more than 50% chance of getting a critical success!
There is a technically slightly more significant chance of having the lowest two possible failures with the d6's, but with how curves work the shift is more in favor of the larger dice giving a better bonus.
The critical success rate makes the systems significantly more high-octane, and as a result it is more significant when characters roll more dice.
Hwaet is mechanically very similar to the core Resistance toolbox, and functions identically. One of the ways that it differs is being more streamlined in its design; it has Attributes, Specializations, and Icons, rather than a vague mastery description that has a fairly complicated mechanic (a good way to describe this is to compare it to the advantage system in D&D; you get an extra die to roll, but you can't stack it).
As a result, there isn't anything distinct in Hwaet that doesn't apply to the core Resistance toolbox analysis, and I'm going to keep the table identical.
One thing that is worth noting is that I was thinking of moving Hwaet over to the Resistance-lite style. I have opted not to do this in part based on what I've found in the analysis here: it's very "swingy" on the high-end. Hwaet is an epic game, for sure, but it's more about struggle than one would think.
Rather than having mechanical interactions that add more dice, Hwaet has "wyrd" abilities, which directly influence dice results. This allows a character to step up results, which provides a more meaningful benefit for most characters (characters have a better chance to succeed outside their wheelhouse, and can pull off great victories within their area of focus).
On a side-note, I need to really push on Hwaet going forward. I hope to have a "sketch" done soon, though it'll look a lot more minimalistic than what was previously shown because I've fallen out of love with the more complicated design: I don't think it accomplished what it was going for in creating a very raw, primordial feel.
Genship Exiles maintains the 1-4 die pool mechanic of the Resistance toolbox, and uses the same systems (skill, domain, mastery), but there is an added boost/entanglement mechanic. The name for the entanglement system is still in progress; it's hideously awkward but I don't like any alternative well enough to switch.
The boost mechanic means that a character rolls a d12, then takes their result, and the entanglement mechanic means that a character does the same with a d8.
This makes our mechanic really difficult to graph in practice, because characters are likely to have unboosted or unentangled dice alongside boosted and entangled dice, but there's not necessarily anything to guarantee that.
However, this much is clear: an entangled character has a significantly higher chance of a critical failure and no shot at a critical success if they don't have more than one die.
Boosted characters who have no bonus dice have a very good chance of success, even a critical success.
However, another thing about Genship Exiles is that it's focus is different than that of the traditional Resistance system; it's intended to mimic Golden Age sci-fi and TV shows like Lost in Space, where characters don't die, though there is still a strong element of central conflict.
One thing that I'm considering doing to help counter-act this is to adopt a horseshoe system for rolls:
Let's take a second look at our original probability curves:
Bear in mind that the 11-12 outcomes are off the chart and rely on boosted dice.
Characters take stress much less often than in the core Resistance toolbox, but to make up for this they highly focused stress tracks (I think they may even just have one, though I haven't given much thought to it yet).
Right now, here's the current rates: failures are much more common (60% with one die and 36% with two, as opposed to 50% with one die and 25% with two), and entanglements really push that further. Characters with four dice have a 12.6% chance of failing (versus 3.13%), and 5 dice still bear a 7% failure rate. Editor's note: I've blown past midnight and I tend to make carrying errors when well-rested, so please forgive any loopiness. If you use the Anydice scripts and switch the graph to "At Most" mode, you'll get the chance of failure with a result of at most 6 for Genship Exiles and 5 for Resistance proper.
Genship Exiles focuses on fast and light storytelling, which tends to be a focus of mine, while the core Resistance toolbox trends to grittier play.
As a result, you wind up with Failure/Complication results, which don't punish players (player punishments are limited to 1-4 results, rather than 1-7 results in the core Resistance toolbox), but do push the narrative onward (e.g. a "no, but" or "yes, but" result).
I'm not entirely sold on this, but it's a tentative solution to making a game that uses familiar systems but feels very different. The dice pool layering of the Resistance toolbox is very elegant, and Genship Exiles uses it in a more forgiving manner to encourage exploration and a more optimistic setting.
That is, if you consider the goal of the players to do what their characters want to do using their characters' abilities. From a mechanical standpoint, we're in the "rolls don't matter" territory by some peoples' standards, and I want to avoid that if possible.