Sometimes Player Characters die. This usually happens for one of three reasons. One of the way that this occurs is when the GM feels like their role is to play the adversary for the players, and/or to give them the biggest challenge possible. For some play groups, this is the default mode, and how they like to play the game. The second way that this happens is when the GM gets so frustrated with their players that they just kill them off in a fit of godlike rage (rocks fall, you die).
The third way that you can have PCs eat it, the one that has happened to me, is when your players make a crucial mistake or a series of botched rolls and end up dead. While this is usually not a fun occurrence for the players, I actually think that it’s a good thing when this happens. Or rather, I believe that the real possibility that it can happen is critical to a fun experience.
Sometimes players do really dumb things, or at least decide to refrain from doing smart things. As a GM, it’s easy to feel an inclination to say “Are you sure that you want to punch the Voodoo Death Priest instead of talk to him first?” because the actions that the players have chosen would totally knock the adventure, or even the whole campaign, off the rails. What makes role-playing fun, though, what separates it from other mediums, is that things can go off the rails. In my experience, the best moments come not from the carefully arranged set pieces, but from the times when something crazy and unexpected happens and everyone is forced to react. If you, as a GM, discourage your players from making decisions that can have negative consequences, you’re also robbing them of those opportunities. Effectively, you’re railroading them and robbing them of their role in the collaborative storytelling.
As I’ve mentioned before, I ran a campaign set in the Old West with the addition of magic and the standard fantasy races. The Big Bad at the end of the campaign was the old Indian healer woman who lived outside of town. One of the PCs was left with her to be healed overnight, during which time she put him under a spell where she could control his mind. I kind of expected this to happen at some point, so I had a safety valve – a tribe of shamanistic Ute Indians that would take one look at the guy, see the spell on him, and remove it. This was supposed to be how they learned who the villain was. My players, on the other hand, never took the hints to go visit the Ute tribe. So when they got to the big battle at the end, instead of it being 3 PCs against 1 ancient sorceress, it was 2 PCs against an ancient sorceress and the other PC. This did not end well. You may not have noticed this in the history books, but the world was overrun with twisted hellbeasts that killed everybody in the year 1867.
As a GM, there were a few ways that I could have handled this. I could’ve dropped some heavy hints that they needed to go visit the Ute Indians before they went up to the old mine. Or I could have powered down the sorceress such that the 2 PCs had a better chance to succeed on their own. Or I could have given the mind-controlled PC some sort of dramatic opportunity to break the spell at the last minute. Any of these things might have given the players victory in the final battle. But it would have been a hollow victory, one that was given to them rather than earned. In the short term, that hollow victory might have been more fun. In the long run, however, anything that removes authorship from the players is damaging. And if they had somehow prevailed, their victory would have been all the sweeter. When the players know that the GM won’t bail them out when they make a bad decision, they also know that they are earning their victories. Your players aren’t 8-year-olds playing chess – they can tell when you’re letting them win, and it diminishes their fun.
Dice Don’t Lie
Sometimes your PCs eat it because they missed that one critical roll – or a series of rolls. The thief fails to detect the trap, then fails his dodge roll to avoid falling in the pit, then fails the dex roll to grab the ledge. Bad day for the thief. The GM, with their godlike powers, can save the thief’s life. They can have them miraculously land on a ledge. Or somehow miss the spikes at the bottom, taking significant but non-lethal damage. The GM can save the poor, innocent thief, who is doomed to die by no fault of their own. But they shouldn’t.
Once again, it’s a question of short-term vs long-term fun. In the short term, the player playing the thief would have more fun if the GM magicked up a way for them to live. In the long term, however, it breaks suspension of disbelief. Your players, perhaps unconsciously, become aware that you will save them when things go badly. As soon as they feel like they can’t lose, the game isn’t really fun anymore. It’s like playing a video game with the cheat codes on. For a little while, it’s awesome. You’re super-powerful and you can’t die and you can just run around causing mayhem with no consequences. After a while, the appeal wears off and the game loses its fun. When you save the players from bad die rolls, you turn on the cheat codes. Now that’s not to say that you shouldn’t give them some extra chances. I’m a big fan of the “well you fell off the cliff, but you still get a dex roll to grab on to a branch as you fall” approach. If it’s plausible, go with it.
The other lesson that I’ve taken from this is to avoid rolling the dice if you only want one outcome. If your player absolutely has to be able to leap across the chasm to save the princess, don’t make him roll for it. If it’s going to Rock More for the PCs to fall into the evil villain’s trap, don’t give them a chance to avoid it. As long as you’re doing it to tell a cooler story (and don’t do it too often), your players won’t mind. Similarly, if you have a player who is extremely skilled at something, to the extent that failing at it is an unreasonable outcome, don’t make them roll.
For example, I have had a couple campaigns with characters that were extraordinarily adept climbers, and they would routinely scale buildings to access the rooftops. If a normal character were going to attempt that feat, I would definitely make them roll for it. When the super-climber wants to do it, however, I would just let it happen. Why? For the same reason that you don’t make players roll dice to turn doorknobs, or walk through a room without stubbing their toe. 99% of the time, they’re going to make the roll anyway, so it’s just a waste of time. And in the 1% case where they miss the roll, it’s anticlimactic, jarring, and unfun.
There Are Always Exceptions
So you should always just let your players die? Well… as a general rule, yes. I wouldn’t say always. If you, as a GM, can find a way to creatively shove more Rockmost in to your campaign by letting a PC live, then go for it. Comic books do this constantly. In fact, they do it too often, and they’ve reached the “cheat mode” point where no reader ever expects a character to stay dead. There is one example, however, where this was used to terrific effect. The X-Men character Angel tries to kill himself after losing his wings. Apocalypse, a major X-Men antagonist and mad scientist type of the biological variety, saves his life. He enables Angel to grow new wings, made of razor-sharp metal, and brainwashes him to serve as one the Four Horsemen. He christens him Death, but when he inevitably returns to the side of the good guys, he called himself Archangel.
This was a terrific way to avoid character death – the character had something happen to them that was cool and interesting, and actually became significantly more interesting as a result. Perhaps even more importantly, it was plausible in the setting, and it was unexpected. Those should be your guidelines as a GM: if you’re thinking of intervening to save a PC from dying, it needs to be plausible, unexpected, and Rock More. Otherwise, just let ’em die. It’s more fun that way.