Spirit of the Century had 3 different chapters on skills – one on what skills did, one on how you, as a player, could be awesome with that skill, and one on how you as a GM could make that skill awesome in play. To this day, while I regret how much bulk it added to the book, it’s a model I love because it speaks right to the *story* your skills tell about your game.
Skills are weird in RPGs. We like to think of skills as “things characters can do”, and we play around with that some. For example, I love the five corner model where a skill is:
- Do the thing
- Know about the thing
- Contact people related to the thing
- Perceive things that relate to the thing
- Perform support actions for the skill
So, the “guns” skill would allow you to Shoot, be knowledgeable about firearms, contact arms dealers, spot a sniper and repair your gun.
Now, this is a cinematic model (the roots of it are from Feng Shui, where skills cover doing, knowing and contacting) that assumes broadly capable characters, and it interacts interestingly with the overall skill list. That is, if you have a model like this, what do engineering or perception skills provide? Note, this is not the same thing as “You should not HAVE engineering or perception skills”, but rather a challenge to make sure those skill are interesting.
In contrast, Cortex (like, pre-prime cortex, specifically thinking the BSG version) did a thing I’ve seen in a few games where it had a short-to-medium list of general skills with the idea that more specific skills were specialties on that skill. For example, you could buy the “guns” skill at d4 or d6, and that covered all gun skills. But if you wanted to get any better, you bought individual skills up from there (so you might buy pistols at d8 or sniper rifles at d10). I disliked the specific implementation (because cortex granularity is such that the cap ad d6 felt punitive) but I liked it a lot conceptually. Done right, it allows for everyone to do stuff, but gives specialists opportunities to differentiate and shine.
The thing is, where it gets weird is in making differentiation and capability proceed in lockstep. In this model. There is no real difference between being good at pistols and bring good at rifles at this point of granularity. Of course, this can be solved with another layer of complexity. If you have feats or stunts or something like that. then you can use skill level as a gate on buying them. To buy the quickdraw stunt, you must have pistols D8 – easy peasy!
In practice, this always seems like the path to overcomplicated exercises in bookkeeping (form this, most stunt trees are born). That’s an annoyance, but over time I’ve come to realize that it can also be kind of fun-destructive, depending on expectations. In almost every stunt tree system I’ve found, there are hidden patterns of things you actually need in order to be cool. Their exact shape varies – they might be thematic groupings or they might be some mechanical threshold at which point things click. A very common example of this in various editions of D&D is the collection of hoops you had to go through to emulate a particular fighting style (like fencing, or two weapon fighting) because god forbid you just *start* with a rapier (newer editions are better about this, but I still have 3e flashbacks on this topic).
A further complication on this is that because you end up needing to fit the model of modular bits, you frequently invite balance problems. To continue the D&D model, two weapon fighting was a stylistic goal for some, but because the combination of rules around it were frequently broken, it also became a mechanical goal.
So adding a layer of other elements like stunts is a solution, but it’s one with some concerns. Enough concerns that it raises the question of whether we can solve these problems within our skill design.
On the surface, it seems obvious, but digging into it reveals something a little bit painful about how we use skills in most RPG design. That is, skills end up providing between one and three of the following:
- Capability (that is, the ability to do the thing)
- Differentiation (that is, a reason why this skill is different than some other skill)
- Permission (that is, an assertion that the character *can* do the thing)
it is rare for a design to openly acknowledge this, but at the same time you can see it everywhere. Whatever skill allows brain surgery has an implicit permission element (assuming the game does not permit everyone to try their hand at brain surgery). Freeform skill system constantly struggle with differentiation. But for all that, we still go about designing skills like they are only about capability, so all it needs is a rating.
Some games have addressed this by forgoing skills entirely (but that introduces its own challenges) and other games have tried different approaches to varying degrees of success. i can’t point at one thing that I think is the right answer here, but the question is rattling around in my head for the moment.