Horror Games & Die Mechanics

Welcome to Horror Week here on my blog! :D

I played a lot of GURPS back in the day, and I was particularly drawn to GMing horror games. I couldn’t tell you why that was; at the time, I hated horror movies. (This was before I watched the Alien franchise, and began understanding horror went beyond slasher flicks.) The games were successful — there was a sense of grit & dread from the system that I could tap into.

Then I played Unknown Armies, which friends & long-time readers know that I claim it’s the best RPG in existence. I loved the feel it has, even though today I feel the whiff factor is a lot less to my liking.

Of course, I’ve played a lot of Fate, and in my experiments trying to make a Fate horror game, I have formulated a hypothesis on what makes a central mechanic work in horror:

Decisions about resources to bear for a given moment are made before the dice are cast. Once the dice are cast, they cannot be manipulated by other resources.

Or, to put it simply:

No fucking rerolls.

The dice are a threshold of uncertainty & capriciousness. In a game where the stakes are set to be harsh and uncaring, like any good horror game, that feels tense. You have to decide whether this die roll (and whatever resources you’re putting to it) before rolling. This shifts the hope in this situation to a single moment in the dice, and hope is the core element of horror.

When you have a game that allows for resources to be marshaled after a roll, like rerolling or adding a bonus like in Fate‘s aspect invocations, you completely destroy that emotional grip the dice have. Immediately, your mind goes to the thought of “well, if this roll doesn’t work, I have these three aspects I can tap and some Fate points…” Some equate this with player agency or games with powerful, competent characters, but it’s more primal than that. It’s where we put our hope or faith in a moment — is it one where we have some control, or is it one where we don’t?

All games are, in some way or another, about emotional points — from the simple celebrations of being victorious to satisfaction of executing a narrative arc. Horror games are about celebrating that feeling of tension and hope of triumph in spite of the wolf in the darkness. At least, the horror I want to play. You know, Alien. Or (some episodes of) The X-Files. Placing hope outside of you (the dice) rather than within you (your ability to deal with a shit roll through resources & other effects) is key.

Games like Fate & Cortex Plus can’t do Alien-style horror, but they can do urban fantasy[1], which is the kissing cousin of horror. That gets into a different discussion, one of theme versus tropes. It takes more than a die mechanic to make a game horror[2], but the wrong die mechanic can turn it into urban fantasy very quickly. And there are die mechanics that seem like they should work for horror, but create other emotional responses that distract from the moment.

All that is what I intend to write about this week. Because, you know, it’s Horror Week! :D

– Ryan

[1] Duh.

[2] Pro tip: you can’t just tack on a sanity or madness system and say “Look at me! I’m horror!


17 Responses to Horror Games & Die Mechanics

  1. Reverance Pavane says:

    Horror as a genre should definitely feed back into the character in some way. Unfortunately when you construct a rule system to do this you are invoking the rational part of the brain, rather than the emotive, and so it tends top lose a lot of its effect.

    That being said I do like the system that was presented in the Orrosh sourcebook for Torg for dealing with the horror trope genre. Basically the monster got a large number of bonuses at the beginning of the adventure, which were slowly whittled down as the players investigated the situation. If the players forced a confrontation too early, they were likely to fail (let alone they may not realise what vulnerabilities the horror had). Even had a nice mechanism for making it sensible for the party to split up while exploring that big old house on the hill…

    While I agree that the moment of crisis is the dice roll, which should be left unaltered, I also can’t help feeling that it is the suspense of the roll that drives the situation in true horror. Particularly if you have some sort of visual/emotional symbol to represent the tension and suspense. Perhaps a Cortex-Plus-like Tension Pool is actually the answer. As the game continues (and the players succeed and fail at the minor tasks involved in the scenario) this pool grows and falls until it is rolled at the denoument. Or alternatively players get to add dice to a resolution pool. But I agree that abilities would have to limit themselves to altering the pool before it is actually determined and the state vector collapsed by rolling.

    • Ryan Macklin says:

      Unfortunately when you construct a rule system to do this you are invoking the rational part of the brain, rather than the emotive, and so it tends top lose a lot of its effect.

      There are few things I could agree with less. But then, I’ve spent years creating such things.

      – Ryan

    • Reverance Pavane says:

      <blink> My apologies. I have the feeling I have inadvertedly insulted you, which was never the intent.

      Perhaps it is better said as “the more complicated you make the mechanics the less emotive the result. That making the feedback mechanism excessively rational mitigates much of the potential emotional impact.”

      So for horror, you want to try and keep the suspense (having the result of the dice roll unknown, but not the possible magnitude of the roll), for as long as possible. And more importantly, keep it simple, or rather, with the resolution uncomplicated and in the “now” and the threat in the future and ominous.

    • Ryan Macklin says:

      Ah, yeah. There, we can totally agree. Though I would also look at the length of time rather than the complexity, even though they tend to be related. The length stretches out what would otherwise be a sharp emotional beat, and the complexity pushes more into rational processing…if the complexity is more math-y. There are some forms of complexity that can work with language language instead, which will sit in a different part of the brain.

      I got into that a bit on my post today.

      – Ryan

  2. Alan says:

    That’s an interesting thought. In general I hate RPGs that demand that I commit limited resources (hero points, fate points, whatever) in advance of the roll. If I still fail, I feel cheated. But in horror, yes, there does need to be a tension that post-roll modifications relieve.

    Now, this only applies to moments of tension and danger. A roll to, say, research and investigation aren’t usually fraught with danger, so Trail of Cthulhu’s “I rub skills on it until the clue pops out” (to grossly oversimplify) isn’t a problem.

    For Fate you could front-load the expenditure of fate points and only allow for the +1/+2 bonus, never the reroll. But I suspect this will lead to further, hard to predict ramifications.

    • Ryan Macklin says:


      I generally hate that as well. It was only by trying to make Fate work did I see how necessary it is to keep those decisions on the front-side, if they exist at all.

      The more I experiment with Gumshoe, the less it feels like a horror game to me. I think that’s another post this week, and will relate why the Fate hack you mention will lead to unsatisfying play.

      – Ryan

    • Alan says:

      Gumshoe isn’t really a horror system, it’s an investigation plot system. It just happens that some horror stories have a strong investigation element, so it makes sense to consider an investigation system for those elements. While one can debate the merits of Gumshoe, I don’t think that the investigation mechanics hurt the horror elements because failing to investigate something isn’t terribly tense. Which I suppose is my point.

    • Ryan Macklin says:

      Given that Gumshoe started as a horror investigation game (Esoterrorists, and later Fear Itself & Trail of Cthulhu), it’s hard to separate the two when looking at what Gumshoe is. Its problem, as far as horror goes, isn’t the investigation part. It’s the way general abilities work. But again, that’s a future blog post.

      – Ryan

  3. jenskot says:

    I think rerolls in horror can work if they come at a severe cost. Especially a cost that isn’t abstract.

    I try to pull Bob out of the way of the machete wielding psycho. I fail. As the machete is inches from Bob’s neck… I can try to re-roll… if I sacrifice my hand first!

    • Ryan Macklin says:


      Could be, yeah. But I would definitely mandate that that’s a cost and not just a chance (which I know is what you’re saying, I’m just being super-clear). No “I’ll risk my hand to reroll” but “I lose my hand to reroll. Here’s hoping it’s worth my sacrifice.”

      I’m gonna chew on that for a bit. Damned good stuff.

      – Ryan

  4. jenskot says:


    I ran a horror games where there was a group shared sheet of Sacrifices. All things you really don’t want to sacrifice. Having this list of awful stuff in the middle of the table was fairly creepy. I also made it so that once something was sacrificed, it was unavailable for anyone else to sacrifice. That way we don’t have everyone taking turns saying, “my hand also gets chopped off!”

    I like having choices available after the first roll and not before. Choices before a roll can create a lot of cautious methodical play. At first I thought “risk” before a roll would be fun, but it can sap the energy from play with too much risk assessment. Although it really depends on the specific game.

    • Ryan Macklin says:

      Exactly. I’m not a huge fan of choice-before-rolling in horror, as it creates that analysis paralysis that kills the game. In a fast-paced action game, that choice tends to be something that gets swept up in the moment, so it is less detrimental.

      GURPS & Apocalypse World both have that “one choice before you roll: what you’re doing” element and “no manipulation after the roll” element. Both would work well for a lot of gritty horror, and is something I’m thinking about a lot as I keep making notes on a horror game I’m developing.

      – Ryan

  5. Jess says:

    I’m not sure if you plan to cover this game, but Dread (published by Impossible Dream) is an amazing game for horror stories. It’s basically designed for it, and performs well, even maintaining the feel of tension and danger that should come with a game of this sort. It’s rather rules-light, forgoing dice in favor of a Jenga tower; conflict and risk are solved with pulls made from the tower. A collapsed tower means death, at least usually. Some areas of uncertainty and risk aren’t covered quite as well as they would be under another system, but given that it’s attempting to reinforce the sense of tension in the story through the mechanics, I’ve always been willing to give it a pass and work around the difficult points in resolution.

    @jenskot: Dread offers a solution to the reroll/second chance scenario. When a player is forced to make a pull that will almost certainly make the tower fall, they can instead choose to knock the tower over willingly. Their character still dies, but the action succeeds. It’s the iconic scene where one person stays behind to hold the door shut while the rest of the group run from the ax murderer, perishing for their heroism. In that light, I would say let the reroll succeeds automatically, as long as it’s not outrageous, but require a serious sacrifice, and not something the player chooses. Basically, trade one uncertainty for another.

    • Ryan Macklin says:

      I’ve played Dread a couple times, and freakin’ love it. Few games engage the lizard brain like it. Though since my hands twitch a bit, playing it is rough for me. So it tends to stay in the “mental exercise” space in my brain rather than “actually seeking out to play.”

      – Ryan

  6. Martin says:

    I tend to agree, but there are some exceptions or variations of this general principle. The game mechanic of ‘Don’t rest your head’ jumped into my mind: The players are able to modify the success of their roll afterwards (via hope coins), but they have no control over which color dominates and thus what negative side effects come along with their success.

    I remember when I run the game for the first time. A few hours into the game, one of the players (who allready was at 5 dice of exhaustion) was rolling against a rather high difficulty to escape from some monsterish nightmare. He succeeded and discipline even dominated… until I spend a coin of despair in favour of exhaustion and passed him the sixth black die.
    “Ha,” he said, looking on the various dice and coins and stuff on the table, “then I’ll just– oh– can’t do anything about it?! Uh-oh!” That was the moment when I grasped how brilliant the design of DRYH truly was: Giving the player after-the-roll-control about their success and at the same time making same absolutly powerless about what the success really means and at what cost it comes – awesome way to bring horror to the table.

    • Ryan Macklin says:

      DRYH is pretty awesome, though as someone who has developed for it, I’m biased. :) That said, it’s as close to being a horror game as Vampire is — the mechanics push a very different agenda, and only with conscious effort to maintain tone by everyone involved can it actually get into horror. I’ve found the horror themes tend to die by the end of the first act of that game, though of course the tropes stick around since that’s what the now-powerful characters continue to interact with.

      – Ryan

  7. Martin says:

    Hm, interessting point. I haven’t played DRYH for longer games – in the games I played, the dice mechanics worked pretty great to keep everyone on the edge and being afraid of their madness or exhaustion spinning out of control. Haven’t seen yet how it develops in longer games.