Perhaps I should clarify what I mean by “Challenge-Based Gaming.” After all, nearly anybody who plays an RPG can legitimately claim to be after some sort of challenge, whether it’s to cooperate with the other narrators to construct a satisfying story, to most accurately portray or experience the inner thoughts of their character, or just to carve through enough identical bags of hit-points with different skins to qualify for a level-up. That’s not what I mean, though. A challenge has to be something that can be failed. Many, perhaps most, styles of RPG don’t and shouldn’t allow failure to accomplish the goal of the game as a real possibility; e.g. if you’re playing to construct a satisfying story then it’s a real problem if the game system will allow months of play to wind up in an unsatisfying story, and if the system is constructed to challenge you to complete a story despite it something is very wrong.
But Challenge-Based Games do exactly that: they are constructed to challenge you to accomplish your goal despite the system working against you. This isn’t particularly uncommon for board games; Reiner Knizia’s Lord of the Rings game is a nice example, pitting the players cooperatively against the rules of the game, which will give Sauron the victory if they don’t destroy the ring before he corrupts the Ring-Bearer.
So when I talk about Challenge-Based Gaming, I’m talking about games where the Challenge of winning against the game is the whole (or at least biggest) point. I’m talking about a style of gaming that pits the players against the game environment, with the GM acting as referee. (As a historical aside, my crowd actually called the DM or GM “The Referee” for at least the first few years we were playing.) The characters are tools that the players use to try to beat the scenario (almost always a dungeon, at least initially); the environment is a tool that the scenario-designer uses to try to beat the players; the GM is the scrupulously neutral referee between them. There can be a bit of confusion here, because many GMs are their own scenario designers, but there really are two hats here: Referee and Designer. When running the game the Referee is not supposed to be altering the environment or rules on the fly to thwart the players. Even if the players would put up with it, it’s just too easy. On the other hand, the Designer is absolutely supposed to be creating an environment that’s going to be hard, but possible, for the players to defeat. Again, even if the players would put up with it, making it impossible is too easy. In a game system like D&D that has “save or die” as one of the central mechanics, TPK is never more than a few badly designed rooms away.
In fact, even perfectly fair fights (where fair is defined as 50-50 whether the PCs or the NPCs win) will inevitably crush the players as they’re repeated; that’s just the law of large numbers. So the Designer’s real problem is to make PC victory both possible and the result of good play rather than chance. It’s trivial to make challenges that the PCs could survive that are equivalent to spinning a roulette wheel; you can even control precisely how likely they are to “win.”
On the other hand, Challenge-Based games require the possibility that the players will lose. In old school D&D it’s often said that there are no winners and losers, and that’s true in a sense. Players can never lose the entire game–they can always roll up another character and have another go at it–but they can fail at a particular attempt and lose a character. It’s a crucial part of Challenge-Based games that this is possible. You could add a house rule to RK’s Lord of the Rings so that if Sauron ever overtakes the Ring Bearer on the corruption track, the players win, but would anyone ever want to play that?
There has been a fairly steady move away from Challenge-Based play in RPGs, for a lot of good reasons and some not-so-good ones, almost from the get-go. I’m not sure that it’s even possible to reach the full potential of the RP part of an RPG in a Challenge-Based game; the challenge is perforce directed at the player, rather than the character…rolling to see if your 17 INT wizard solves the riddle just isn’t the same kind of experience as trying to solve the riddle yourself. And a challenge-based game can’t reasonably demand that players firewall their knowledge from their character’s knowledge; it would be insane to expect players to get any enjoyment out of having character after character fall victim to the same trap just because the character can’t have any way of knowing it’s there.
But I do know that just because it’s rare and not the current fashion doesn’t make it an illegitimate approach to gaming. The “problem” that players of RK’s Lord of the Rings have that Sauron sometimes wins if they don’t play well can’t be solved by having a discussion around the table so they can drop the Watcher in the Water and replace it with something “fun.”