I'm used to, in games like D&D, character power being well-defined by level, with the assumption that characters of the same level have equivalent strength (at least in combat). I talked about the "more strength, bigger rock" issues of this in a previous post. But another interesting aspect of this is that a lot of stories don't have this property: often there are characters of quite different strengths, or at least combat strengths, working together. And the D&D default is that the foes you face aren't so strong that you can't defeat them in a reasonably fair fight, whereas fantastic stories in fiction often have heroes face foes whose raw strength is much greater. (Obviously there are many other kinds of stories, I'm just focusing on these types of situations for this post.)
One of the focuses in Worm is that the protagonist has what might seem like a weaksauce power, the ability to control insects, and comes up with all sorts of crazy and creative uses for it (and more generally, coming up with creative uses for a variety of powers). But that's not what first got me thinking along these lines. The superhero bureaucracy in Worm classifies "capes" by type and power level, in a way that's a lot like class and level in a tabletop RPG. But, unlike in a D&D-style tabletop, the protagonist's group isn't a bunch of capes that all happen to be the same power level; some are stronger, and some are weaker. Moreover, this doesn't translate directly into effectiveness, even combat effectiveness. The characters also vary a lot when it comes to creativity, fast thinking, and dealing with unexpected situations, which often determine the winner of a fight more so than raw power level. So, that got me thinking about what I might do to build a tabletop designed for varied power levels and focused more on player creativity than mechanical conflict.
An obvious basis for comparison is the Dresden Files RPG, based on Fate. The Dresden Files books have ordinary humans that go up against wizards and werewolves and vampires and who knows what else, so it's definitely playing in a similar space. It's approach to balancing this is rather simple: you've got refresh points that you can spend on supernatural powers or mortal stunts. Supernatural powers are stronger, but if you stay pure moral, you get +2 refresh. It's simple, which is good, and it probably works well in practice (none of the PCs in my campaign were pure mortal). Refresh is assumed to be even among PCs; this doesn't do anything to support supernatural characters of different strengths or mortal characters of different levels of badassitude. Which is fine for that game, but it's a bit different than where my thoughts were going.
The perspective I find the most intriguing is, surprise surprise, one from Chuubo's. (Noticed a pattern?) It has (I believe originating in Nobilis) a principle that actions aren't usually directly opposed. If a world-defying horror is immolating the city in scarlet flame, and I'm trying to protect myself and my friends from being burned to death, it doesn't necessarily matter if the world-defying horror is using a miracle with a much higher rating. Its miracle succeeds: it immolates the city. My miracle succeeds in that context: I save my friends in an immolated city. It might be different if it's specific goal was to burn my friends to death, but in general weaker actions work to the extent that they can given the context of the stronger actions, rather than just failing. This seems like it leaves lots of room for creativity and uneven parties or conflicts. I think, again, this is helped by the lack of a turn-by-turn combat mini-game, so you don't need a pool of standard action types that are directly comparable.
(It's also interestingly different from Apocalypse World's lack of opposed actions: in Apocalypse World, there are no opposed actions because the GM actions are fundamentally different from the PC actions; the NPC context may set the possibilities or consequences of success or failure, but the difficulty of the roll is not based on the strength of the opposition.)
The flip side of uneven parties is making the group dynamic work well. A tabletop is not a novel: players probably mostly want to get similar levels of screen time and feel like they're accomplishing things and driving the plot. Even when characters are theoretically the same strength, either in-practice power variations or other things like plot balance or player personalities can lead to some characters dominating play more than others. (Similarly, in a LARP, where it's easy to (intentionally or not) have characters of different effective power levels or plot relevances, you need to be careful to make sure that you don't end up with characters feeling too marginal or insignificant.) This is mainly a GM and group skill issue, though some systems, like Chuubo's's having characters fade after taking their XP actions, provide structure to support balance here.
I'm probably not actually going to flip out and write a Worm RPG, and I do need to actually play Chuubo's sometime, but the idea of mixed parties does seem interesting for the right setting or game, and maybe I'll explore it more intentionally at some point in the future.