Granted, but exclusion in turn is useful for any number of things.
As I see it, there can be all kinds of similarities and all kinds of differences between a certain thing correctly categorizeable as a a story and a certain thing correctly categorizable as a game, and a few of these are relevant for a given purpose, a few of them are relevant for another purpose, and so on. (And, again as far as I can see, there is nothing to prevent some things to be correctly categorized as both games and stories at the same time.)
So it all boils down to what purpose one wants one’s categorizations (i.e. one’s exclusions and inclusions) to serve. Some exclusions, no doubt, serve very suspect purposes, indeed; but often, I’m equally sure, they are not particularly morally objectionable; and for some purposes (like medical treatment, say), surely, it would be downright wrong not to include only some things or persons and exclude others.
As for prototype theory, as far as I know, it is an attempt at an empirical theory about on what grounds people actually do categorize things: it proposes that categorization is done by means of similarity to one’s conception of a prototypical member of a given class, rather than by strict definitions in terms of necessary and sufficient conditions for class membership. Prototype theory may well be false for all or some concepts – I’m not a champion of it, particularly. But I can’t see that it follows from this theory that people can’t perfectly reasonably categorize things by means of slightly or widely different prototypes. Nor that it follows from it that people can’t conceive of new interesting ways to categorize things (i.e. seeing previously ignored similarities and differences between things) or of inventing things that fall outside, in between, or transcends any given categories.
Further, if there are no real essences inhering in the nature of things, and if reality doesn’t in some other way by itself fall into a certain number of more or less neat categories quite regardless of human or other interests, then what alternative is there to our imposing categories on reality guided by our own interests? If reality as such, in and by itself, draws no distinctions between stories, games, and plays – or between carburators, unemployment, and daffodils –, and people shouldn’t draw those distinctions unless they exist in reality, then we couldn’t say or even think anything about anything, could we?