Last time I wrote a bit about how my frustration trying to build characters inspired the design of Bardworthy. It seems only fair to continue by considering another character it could be hard to represent: me.
If My Life Were a Story, I Wouldn’t Believe It
I’ve been struck by lightning, sucked out to sea, hit by a truck, and even stabbed a few times. I’ve fallen out of trees and off buildings and been thrown down multiple hills. I survived it all mostly unscathed: I broke my leg once and have a couple of entirely unimpressive scars, plus some lingering nervousness about oncoming traffic and the ocean.
Meanwhile, I spent a large chunk of my first year of life in the hospital, spent most of my childhood sick, and for a decade had to be careful what sort of water I drank, because if it wasn’t sufficiently filtered it would make me ill. There are breads I think are spicy. This past New Year’s Eve, I accidentally poisoned myself with almonds.
If I were a character in Dungeons and Dragons, would my Constitution be high or low?
I seem to be an absurd blend of sturdy and fragile, and that’s not even the only contradiction. I’m strong but was never athletic. When I was younger, I was nimble but never graceful. (I’m still not graceful, to be clear, but now I’m also not nimble.) None of this feels unusual to me, but it’s difficult to express in a system that expects those traits to rise and fall together.
The Complexity of Modeling Complexity
Dungeons and Dragons began as a war game. War games model the movement, positioning, and conflict of groups: usually squads of soldiers. When the designers tried to represent individual characters, they did so using a small number of variables suited to the game’s original purpose. The result was a framework optimized to describe what individuals do in war-game contexts: fight, endure damage, notice threats, and lead others.
As the game grew into a role-playing game, subsequent development built on that framework rather than replacing it. For a long time, this worked well, because the focus of play stayed close to the kinds of stories war games already handled effectively: dungeon crawls, tactical combat, epic battles, and strongholds. Outside of those contexts, the framework was always going to be an awkward fit.
Any attempt to model something complex with a limited number of variables involves compromise, and every compromise creates limits. The limits you accept depend on what you’re trying to represent. Dungeons and Dragons compromised on the variety of individuals it could represent because it was primarily concerned with representing a narrow set of actions and outcomes.
Its ability structure reflects that. Even in later editions, more complex character ideas tend to grind against the sharp edges of those original compromises. Interesting flaws are glossed over, distinctions between different kinds of aptitude are smoothed down, and traits that don’t need to be related are forced together.
Think of it like rounding in math. Values that are meaningfully different, 2.6 and 3.4 for example, become the same if you round them to three. If the numbers represent something outside your focus, not at the core of what you’re trying to accomplish, three is probably close enough. Dungeons and Dragons’ abilities were premised on war games, and for that they were close enough.
But what if the differences between 2.6 and 3.4 are exactly what you care about? Well, Bardworthy starts from a different premise.
How Would You Describe Yourself?
Imagine that you’re trying to describe yourself to someone who has neither seen nor met you. There are all sorts of adjectives you might use, depending on your characteristics, preferences, and self-esteem. You might call yourself strong, fast, or tough; knowledgeable, quick-witted, or insightful; charming, kind, or loyal. (There are, of course, just as many negative adjectives, but let’s assume you’re excited about being you and trying to describe your best features.)
The Bardworthy project began by asking how a game might represent the many kinds of people in the world in a way that could account for as much of that variety as possible.
The first thing we noticed is that people tend to describe themselves along three broad lines: bodies, minds, and relationships. If we wanted our game mechanics to reflect a similar range of human variation, we would need at least three kinds of ratings. We called these categories Arenas, because they describe spheres in which a person acts: the Physical, the Mental, and the Social.
That only got us part of the way there. Within each Arena, there are still many different ways a person might excel or struggle. Take the physical Arena as an example, because it’s often the most intuitive. How would you describe yourself physically? Are you strong, quick, graceful, resilient? It’s easy to imagine someone being one of those things and not another. In fact, many classic archetypes are defined by those contrasts: the brute is strong but slow, the sidekick is fast but clumsy, the acrobat is graceful but fragile, and the survivor is resilient rather than powerful.
For those contrasts to matter, the traits involved had to be distinct. You can’t contrast a thing with itself. As we brainstormed the adjectives people use to describe themselves, we noticed that within each Arena they tended to answer the same four kinds of questions—questions about how a person’s aptitude expresses itself.
If you imagine each Arena as representing a place where a person can exert power, those questions become: how much raw power does the person have; how quickly can they change how they’re using it; how precisely can they control it; and how long can they sustain it? Each answer describes a different mode of aptitude. We called these modes Aspects.
Returning to the earlier archetypes: the brute has a great deal of physical power but can’t adjust it quickly; the sidekick can react rapidly but lacks precision; the acrobat can carefully apply even limited strength; and the survivor’s power may not be overwhelming, but it endures.
The same distinctions appear in every Arena. A person might be intelligent but slow-thinking, quick-witted but unoriginal, insightful but uneducated, or knowledgeable without being especially clever. They might be charming but easily flustered, confident but inconsiderate, polite but unwelcome, or popular despite being dull.
We named the Aspects Capacity, Adaptability, Integration, and Longevity. Combined with the three Arenas, they form the Attribute Matrix.
The Attribute Matrix
The Attribute Matrix has some immediate and obvious benefits. Most simply, the four Aspects and three Arenas combine to produce twelve Attributes—twice as many data points as traditional six-stat systems. Qualities that are bundled together in games like Dungeons and Dragons are unpacked, allowing players to build characters that are more nuanced, more distinctive, and more internally coherent.
More importantly, the Matrix gives the game three different paths to affect each Attribute. Any single Attribute can be adjusted directly, through its Aspect, or through its Arena.
A character might train to develop Dignity (Social Longevity). Their species might be particularly stalwart, granting a bonus to all Longevity Attributes. Then a public humiliation might penalize all Social Attributes. All three influences converge on the character’s Dignity—but in different ways. The character may still be more dignified than most members of their species, even after the embarrassment, yet the emotional shock ripples outward: Charisma is tainted, Confidence shaken, Prudence made suspect. A single event doesn’t just change a number; it subtly reshapes the entire social profile.
In this way, the Matrix models characters not as collections of isolated traits, but as interconnected systems. Changes propagate naturally, producing consequences that feel organic rather than arbitrary.
More subtly still, the Matrix provides a framework for designing countless other mechanics that benefit from a more detailed model of a person. Combat is a good example. Traditional systems collapse defense into a single “armor” value, bundling very different kinds of protection together. But blocking an attack with a shield is fundamentally different from stepping aside. Both are different from simply enduring the blow, which is itself different from recovering quickly afterward.
Those four defensive modes map cleanly onto the Aspects: Capacity, Adaptability, Integration, and Longevity. By expressing these distinctions mechanically, the system allows character traits to meaningfully shape how danger is faced, not just how much damage is absorbed.
The same structure extends naturally beyond physical conflict. Because the Matrix includes Mental and Social Arenas, it supports equally rich forms of confrontation in those domains. A battle of wills between rival mages, the slow erosion of resolve during an interrogation, or the rhetorical parry and thrust of public debate all become mechanically expressive encounters. Every character is already built to participate, and the rules can consistently represent a much broader range of dramatic situations.
Once characters are allowed to be complex, the rest of the game can afford to be honest about complexity too, which means it can tell richer and more interesting stories.
Next time we’ll look at modular classes, another way Bardworthy allows characters to be as interesting as the people all around us.
Until then,
Daniel
To be clear again, Dungeons and Dragons is a trademark of Wizards of the Coast, and I have tremendous respect for it.