One of the big reasons I don’t like the word “game” to describe RPGs is that the parameters for success are so diverse. A traditional game has a well defined victory scenario (though some of the more interesting ones have several, which adds layers of complexity). In an RPG, one does not simply win, at least not normally. Since the foundation of the game is story, one can have an ending without a clear set of standards for winning.
How do you “win” a story? I think that at best you can hope for varying degrees of success, but that depends on your expectations as a player and the expectations of the gamemaster. Another variable there is how much control or authority the GM has over the story to begin with. And add to that the idea that the players can each have different ideas of what the game should be and what the parameters for success are. A game I was in several years ago ran into this problem. Another player and I played twins, and somehow we ended up with dramatically different ideas about what we wanted from the game – I wanted an extended story that was focused on character development, and he wanted a large-scale, potentially earth-shattering battle. One of us was disappointed, needless to say (it was me, and this caused a minor rift between me and the GM, who believed that he had warned me about it; if he did, I missed the hints) and only years later can I appreciate the irony.
And this story is useful in that it shows us how much easier it can be for the GM to just take over and tell one structured story. Easier might not always be better, though, because players will want to tell their own part of the story, and they might lose interest if they can’t do that. And where success can be so difficult to define in this setting, failure becomes pretty obvious pretty quickly if the players no longer want to play. In the example game, I was enjoying myself thoroughly until the bitter end, as it were. The GM was allowing me to do what I wanted (within the rules of both the game and the world) in order to build the character, so overall it was a success even from my point of view (even if it took me several years to understand it). But boy, did it feel like a failure at the time.
And the broader the standards for success, the easier it is to accomplish failure. It seems like this would work the other way, but once you give individuals the freedom to set their own success parameters, they will inevitably narrow them down, and they’ll do it differently than the player next to them. I’ve played with intentionally disruptive players before; sometimes that’s just fun for them (and it’s rarely fun for anyone else). But give the players enough rope and, well, you know.