Harry Potter and the Lost Art of Making Players Cry
Of course, it helped that she has already killed off several important characters. She was clearly willing to follow through on her threats. As a result, any time the action heated up, I thought "This could be it. This could be the end."
She had effectively stripped away the protagonists plot invulnerability. She also managed to emotionally involve me, and make me actively worry about the characters. And, I'm sure, the story was better as a result.
Joss Whedon pulled a similar stunt in Serenity. By killing off Wash, he raised the stake for the final battle with the Reavers. He sent a clear message--if Wash could die, then anyone could die. This definitely cranked the tension to 11.
So, how does this apply to RPGs? I would like to bring this kind of added tension to the table. The naive answer is easy, kill a character or two. If characters can die, and if they players know that characters might die, then conflicts will have an added level of tension.
But, wait just one second. Different players have widely different views on PC death. Some people hate having their characters die. They spend a lot of time creating their characters, they invest a lot of emotion and time into their stories. They don't want them snuffed out by a few bad die rolls. Killing their character may well ruin the game for them, or cause them to stop playing entirely.
Also, if character death is common, then the players will begin treating their characters as disposable placeholders. They will hesitate to invest a lot of time and energy into any one character, since that effort will likely be wasted.
So, to optimize the power of PC deaths, they must be possible but rare. Also, PC deaths should represent important moments of high drama--not just the result of a bad roll. Clearly, this isn't easy to pull off. So, what's the solution? I'm frankly not sure. Sometimes you can mimic PC death by killing off an important NPC. Sometimes you can plot out a character's dramatic demise with the player's permission and input. But neither of these feel like adequate solutions. They address corner cases and exceptions, not the real heart of the matter.
So, let's turn the situation on its head for a second. Many independent games use a stake-setting mechanic. During any conflict, each player explicitly states what they are willing to risk before any rolls are made. While I understand the reason behind this--for example, it helps reduce miscommunication and misunderstandings--I always find myself bristling at these rules. In part, I think they adds too much meta-talk, which distracts me from the story. Mostly, however, I feel that it leaches the excitement out of the conflict.
If I can set the stakes, then I can limit the amount I am willing to risk. I know, before I pick up the dice, that I will never face unacceptable resolution. However, real drama comes from stepping over the lines. From moving past what is safe and what is comfortable. And uncertainty always heightens tension.
I don't need to risk my character's life every time the dice fall from my fingers, but when I look across the table at the GM, and I have no idea what might happen--that's when real fear sets in. My pulse races; my palms sweat; I shake the dice nervously, afraid to let them go. That's exactly the kind of energy I want in my games.