Don’t Roll For The Horror

As part of May of the Dead blog carnival put on by the Going Last Gaming Podcast, I’m going to wax about some horror thoughts. Long-time readers know that I loves me some horror gaming and have a lot of thoughts on it. Today, I want to dive into some thoughts on a hypothetical game system[1]: what separates lowly monsters from truly horrible beings.

What notion I’ve come to is: the scariest of monsters are those that don’t miss.

Part of horror comes from a discrepancy between the protagonists’ competency and the Threat’s. Whether that Threat is Dracula, Azathoth, a Terminator, or other sight that causes nightmares in those whom encounter it, the core is that the Threat will win in a stand-up fight.

Oh, and yes, I so want to run a horror game that is about the first Terminator movie.

But many of our games don’t reflect that, at least not strictly speaking. Games often have the Threat roll to see if it hits, and there’s a good chance that it won’t. It’s reflected in our language: “the vampire attacks!”

Let’s fold, spindle and mutilate that. This means trying some experimental stuff with our games, namely (as the title says): Don’t Roll for the Horror. Start it off not with something as wishy-washy as “attacking,” but something more concrete:

The vampire jumps on you and rips your neck open with its fangs!

Now, the reason we tend to say “attack” is because we’re inviting the potential victim to respond, in no small part because the game system gives them that privilege. But by jumping right to what the Threat seeks to do with no softening, we’re doing two things:

  • We’re changing the language used to respond. Horror as a theme is partly about rebelling against that which is more powerful than you. So instead of just saying “I defend” in response,” you’re saying “No! You don’t just rip my neck open!”
  • We’re saying that if the Threat doesn’t succeed, it is entirely because of the protagonist’s action.

Those are both awesome things to me. So let’s look at how to rock that structure:

  1. The Threat does something. Something big. It doesn’t ask. It doesn’t try. It just plain does.
  2. A protagonist responds to push back, drawing the line in the sand and fighting the good fight.
  3. The protagonist rolls for that action. And just the protagonist. Not an opposed roll setup. You know how strong or dangerous this specific moment is, so set the difficulty accordingly.

Depending on the result (and the numbers involved will vary from game to game), the following happens:

  • Fail by bad enough: the Threat fully succeeds. Someone is probably dead.
  • Fail by some amount: the Threat doesn’t get what it wants, but you’re hurt in the process.
  • Succeed by a small amount: give-and-take. You’re hurt, but so is the Threat.
  • Succeed by a large amount: a moment of reversal, when the Threat is the one hurt or driven off.

Each of these is important. If we’re saying that a Threat might just straight up kill someone, the roll has to reflect that chance. Otherwise, we’re just lying in our descriptions, and everyone at the table will see through it. Tension is dropped. And the middle two reflects the notion of partial failure & success — horror thrives not on absolute moments but on small victories and setbacks. Finally, you need to give hope in the moment, which is where the last result lives.

If we’re doing away with rolling, this means we throw out the idea that the Threat might act slower than protagonists — you know, initiative. Horrific competence means Threats push first. The only time when that might be different is if the protagonists are aware of the current situation and somehow make themselves able to get the jump on the Threat, and even then that’s about chance rather than certainty.

After all, that’s how it often works in horror fiction.

Finally, since I bought up “getting hurt”…my favorite system for damage in any horror game comes from Unknown Armies. It’s easy to die, and you never know how many hit points you or anyone else has left. The GM rolls & keeps track of stuff in secret. While normally I hate secret rolls, I like it for damage in horror. It has two things going for it: one, you don’t have absolute certainty of how far you can push your character; two, and frankly far more important, it causes the table to rely on the hurt described rather than numbers. That’s very powerful mojo, because it’s language that makes a horror game really pop.

Again, this is about an idea of a new game system, but it wouldn’t take much to try some of these ideas in an existing one, as long as the game can support horror beats.


A word of note: this setup doesn’t do action-horror — at least, if it’s the sort of “action-horror” that is more action than horror. Which most are; it’s a fun subversion of classic horror construction, where competency is more at parity even if vulnerability is still vastly not.

– Ryan

[1]To be fair, it’s not entirely hypothetical to me. I have notes about using this idea for a game system that uses my Emerging Threats Unit campaign frame, but it’s far from primetime.


7 Responses to Don’t Roll For The Horror

  1. This is great stuff, Ryan, and shows me very concretely how to handle something I’ve been struggling with in my Monsterhearts game. When I play a Threat in that game, I want it to be bad-ass, but I’m not allowed to give it stats and I’m not allowed to roll for it. Because of that, I’d been working under the assumption that the most reliable way to impose upon the PCs was to passive-aggressively provoke them, have them respond in turn, and wait/hope for them to fail, so I could make a Hard Move in response – but what you’re saying in your post makes far more sense for how the game actually works. You simply state not only what their intent is, but also their execution, and if the PCs care to actively contradict, they get a chance to delay and perhaps even avoid their fate.

    I can see how some games may need to be retrofitted / house ruled to support this method, but Monsterhearts (and possibly all or most games using the Apocalypse World engine) should definitely support it as written.

    Looking forward to hearing more of your thoughts on horror play.

    • Ryan Macklin says:


      Right, totally. Sounds like what your normal MC move is “describe doing a hard more.”

      Playing AW has given me some of these thoughts, though the first engine I built for Emerging Threats Unit is much more akin to Otherkind dice. Didn’t quite work as intended, but data was gained from that.

      If I was going to do a AW-leaning hack for this specific game, which I’m not, I would rip out XP from the system. It would incentivize certain actions over the overall concept of survival itself. I suppose horror & incentivized decisions is another topic to write on.

      I have other thoughts on horror, if you click on the horror tag. Though you personally may have already those. :)

      – Ryan

  2. Albert A says:

    Some old-school stuff was on to this too. The early-90’s(?) horror game /The Whispering Vault/ never made the GM roll dice. If the heroes wanted to attach the monster, they rolled against the monster’s static defense. If the heroes wanted to be not fucked up when the monster attacked, they rolled against the monster’s static attack value.

  3. Benoit says:

    Very cool. I love the idea that the horror *just hits*. It seems counter intuitive that a guaranteed hit would create *more* tension, but it does. I’m going to try and incorporate some of these ideas into my two page delve I’m doing for May of the Dead next Friday.

    • Ryan Macklin says:

      I’d recommend trying a game like Apocalypse World, where the GM never rolls dice. That sense of “if I miss there, there is no luck factor to save me” is powerful mojo.

      – Ryan

  4. Reverance Pavane says:

    With regard to the small victories and setbacks I always did like how the Orrosh sourcebook for Torg handled the basic horror genre with Perseverance. The monster essentially started out with a whole host of advantages (essentially the ability to discommode opponents, avoid damage and the like), that the players had to slowly whittle away with minor victories before they could actually face it and hope to succeed. [This wasn’t necessarily dealing with the monsters abilities directly, but also being able to overcome the character’s own fear and horror. Also had a nice mechanism for getting people to do the horror-story thing and go down into the basement alone…

  5. Arcane Springboard says:

    @Reverance Pavane: Interestingly enough, I’m working my Torg hack for Marvel Heroic Roleplaying and specifically, Orrorsh, at the moment. You should see the result within the week.