Why do RPGs have Attributes?

OK, first of all, I know that not all RPGs do. This is only talking about the sort of RPG that tries to model characters with some combination of “attributes” and “skills”. For purposes of this discussion, “attributes” are character stats that are defined by the game as being more-or-less inherent, and either they don’t change or they don’t change easily. They’re meant to capture how we think about intelligence or healthiness in the real world—we don’t usually think of someone as becoming more intelligent or less prone to illness. “Skills” are defined by the game as learnable, or at least explicitly improveable.

So I was just thinking about what purpose do attributes serve in this model? Mechanically, (1) they often provide a baseline or fall-back for skills. If your character doesn’t have the history skill, they can make an intelligence check to see if they know who was regent in 1237; if they don’t have the athletics skill, they can make a strength check to climb a tree.

(2) Attributes often mechanically reinforce the notion that people have inherent aptitudes. Skills will be cheaper with a high attribute, or attributes are added to skills so you get to a higher effectiveness with fewer points spent on skills. There are other methods, of course, but the point is that usually skills are either coupled to attributes or stacked with them in play.

But then I was thinking about research questioning the notion of intelligence as an inherent quality, and how a person could seem to get colds easily but be physically resilient, or smart about math but poor with language, or agile but have butterfingers. Obviously, one solution to this is to have more, narrower attributes. But at some point, you get into the same problem that dis/advantage systems sometimes create, of encouraging excessive specialization that creates characters that aren’t believable.

But maybe we could capture the verisimilitude that attributes provide without needing attributes? Part 1 is completely solvable: you just need to have something to roll if your character doesn’t have a skill. You could define all skills as starting at 1 instead of 0, or have a dice mechanic that works without a skill input. Or say that without points in a relevant skill, you can’t attempt that action. You have to make some design decisions to go along with whichever one it is, but it’s a problem that has been solved multiple times in multiple ways.

But it’s particularly part 2 that got me thinking, because I’m not sure it’s been solved in without attributes (please, comment if you know of a system that has done something to reflect natural aptitudes without using attributes). Back to actual research: we’re increasingly finding that things we think of as inherent aren’t, and things that we think of as connected aren’t (and, meanwhile, other things turn out to be (maybe) connected). On top of that, some things that are typically treated as skills in RPGs may be more inherent than learned. But even then, obviously all the aptitude in the world doesn’t help if you are never exposed to something.

So, the split between learning and aptitude is probably artificial, and the grouping of “related” aptitudes is at least somewhat arbitrary. My thought experiment is: instead of trying to reflect these with attributes, lets just work directly with the skills.

The core of my idea is: the more skills you have in a given area, the cheaper it is to improve them or to learn new, related skills. If you want to make an athletic character, you don’t put a bunch of points into “strength” and “agility”, which then make it easier to get higher scores in the various athletic skills. Instead, the formula for cost of skills takes into account related skills, and costs fall as you acquire more related skills.

You’d have to define the skill groups, of course. For a rough version of this, you could start with a system that has attributes, and use the attributes that skills are based on/linked to define the groups. But the next level would be to have finer control over the skill groups—having 15 attributes gets unwieldy, but grouping skills into 15 categories shouldn’t be too much of a problem. And if you felt it improved verisimilitude, there’s no need for all skill relationships to be reciprocal: music skill and physics skill could both benefit from math skill, but music skill doesn’t benefit from physics skill, say. (n.b.: not making any claims about real-world accuracy here; just inventing an example.)

The cost to raise a skill might be something like ({new skill value}×3)−{number of related skills}. So putting a point into a skill makes every skill it is related to a point cheaper for every increment—a very similar result to putting a point into an attribute in many systems. But without having a separate attribute to track, and without having to decide what is “inherent” and what is “learned”—if it’s a thing you want characters to be able to roll dice to do, make a skill for it.

Is this actually reducing complexity over a typical attributes/skills system? I’m not sure. It would depend largely on how complex the math ended up being and how hard it was to create skills to fill in any gaps that attributes were handling. And even in the real world we often think of some abilities as being inherent and others learned, so would this be too jarring, and make verisimilitude harder? I don’t know without trying it. Let me know if you try something like this out.

Advertisements

How I Met Your Mother. Period.

Last week game night was canceled, so I had an extra night to work on schoolwork.

So of course I instead taught myself how to use iMovie† by fixing the final episode of How I Met Your Mother.

OK, here come mild spoilers for a series that ran 2005-2014. I’m going to keep this to a level that I don’t think ruins anything if you haven’t seen the series and now go watch it, but tolerance of spoilers varies, so if you’re super extra-special averse, just take my word for it that it’s excellent except the final episode, and if you’re that averse I can’t possibly explain to you how to fix it without including spoilers. But if you trust me, read on.

Continue reading

What’s a “Pre-existing condition”?

The GOP is again talking about just passed a “better” plan to replace the ACA, and thus, despite Trump’s “promises”, people are talking about the importance of not denying health insurance to people due to pre-existing conditions. You’re going to hear truly tragic stories about people surviving or not because of this, about people who would’ve been excluded from health insurance in the past and are now stable and productive members of society, about people being left to die because of lack of money or accident of birth.

I don’t have any such story. But I think it’s important to understand that we’re not just talking about multi-million-dollar expenses for comparatively rare conditions and tragic accidents. This used to be be far more widespread.

Continue reading

Well, that’s a fun new failure mode

Background:

When you add a person to (or remove from) a meeting in Outlook, but make no other changes, sometimes it automatically sends the update to all attendees, and sometimes it puts up a little dialog box asking whether you want to send the update to all attendees or just the new people.
send to all attendees
After years of experimenting, I have given up trying to predict whether or not it will give you that option.

Continue reading

On Filibustering Gorsuch

By the time you read this, it may be a done deal, since I forgot to write this last night. But I think it’s still an important statement to get out there.

The Republicans have made it very clear that they intend to confirm Neil Gorsuch to the Supreme Court, no matter what. So the Democrats have nothing to lose by opposing him, and nothing to gain by letting him through easily.

“But what about preserving the filibuster for future SCOTUS nominees?”

What about it? Do you really think that if the Democrats don’t filibuster Gorsuch, but filibuster a future nominee, the Republicans won’t just eliminate the filibuster at that point? What is going to change between now and then that suddenly the Republicans are going to be interested in consensus or compromise, after being uninterested in either since at least 2003?

Continue reading