Failure in games

[Haru]

"little Suzie will be eaten by vampires."  (And for the sake of the hypothetical, rescuing little Suzie is the point of this game.)

I would even go a step further. I would say "the result of this adventure will be that you rescue Suzie". Then, you can see how you make an adventure of it. That's kind of where I was going with that quote. If you agree on something like this, you shift the focus of the game from the goal to the way to get there. There is no longer the peril of failure, there is only "pretend peril". Suzie can be abducted in the first attack of the vampires, there will be a way to get her back. Suzie has already been transported to another town when you storm the vampires headquarters, but you find an address. And so on.
You could even go and throw everything around at the end (if all players agree on that), have her taken and turned, but you find a cage full of children that you can rescue instead, and with a new vampire, you have a focus for the next game.

On the other hand, if you do not agree on something like that, every player will have a different view on those things. Most often, that will be "if we don't rescue Suzie from that first attack, she will be dead, so there can't be any failure". Like what tutori had said earlier in this thread.

I don't think that you have to have paper villains. The peril for the characters can be quite real, as long as the players are in agreement of what are acceptable outcomes, and what is off limits. Character death, for me, for example is off limits, if it just comes from bad luck on the dice. Sacrifice is absolutely possible, as is stupidity. Though I usually tell the players, that their action is stupid and will most likely end up with their character being dead. The players then need to find a way to beat a more powerful villain, find his weakness, and so forth. If you kill off the characters left and right, you'll end up with a group of characters that don't really have anything to do with what you are doing, because they are simply the 5th generation of replacement characters ("would you care to join our noble quest?").

[Vairelome]

As a point of clarification, I wasn't meaning to suggest that the villains must be paper villains; I was trying to emphasize the importance of player perception over mechanical reality here.  In most cases, adding to a villain's mechanical challenge should also add to its perceived threat, though you still need to play that up--the flipside is that a brutally tough godmode boss (on paper) won't make much of an impression if his threat level isn't sold to the players.

[Wordmaker]

There's so much I could do with that goal beyond just "Suzie dies if you fail."

I'd be more inclined to rule that the worst-case failure scenario would be that Suzie is bitten and part-turned. So the PCs do rescue to her, but now they have to deal with someone on the verge of becoming a full vampire, and figure out how to handle this, which can be an especially emotional event if Suzie is a character they've known for a while and care about.

[InFerrumVeritas]

There's so much I could do with that goal beyond just "Suzie dies if you fail."

I'd be more inclined to rule that the worst-case failure scenario would be that Suzie is bitten and part-turned. So the PCs do rescue to her, but now they have to deal with someone on the verge of becoming a full vampire, and figure out how to handle this, which can be an especially emotional event if Suzie is a character they've known for a while and care about.

For something like this I would have degrees of failure:  Suzie is okay.  Suzie is emotionally traumatized and no longer trusts you/thinks you're dangerous.  Suzie is badly injured and now you have to protect her in the hospital.  Suzie is partially turned and you have to deal with the consequences, but you rescue her.  Suzie is completely turned and you have to deal with the consequences, but you rescue her.  Suzie is completely turned and is now a powerful enemy, whom you may or may not try to save or redeem.  Suzie was used to fuel a powerful ritual against you, which you manage to survive but must deal with the fallout from as her ghost haunts you.

My players usually end up somewhere between options 2 and 4, possibly up to option 6 or 7 if their plan is especially inane. 

[Wordmaker]

That's a great idea.

This where failure can get really interesting, and why I love DFRPG's use of Consequences and the ability to concede in a conflict. You don't need to go all in for one extreme or the other. There's so much more to failure, and success, than killing someone.