Consequence Heatsinks

So I’ve got this dilemma. I like rolling dice to see how well things go, and I like asking players to roll dice to find out how well their actions go. But I don’t like the dice stymieing a perfectly good plan or an interesting action from a player. What’s a girl to do?

There are some answers I find basically unsatisfying.

1. Roll the dice, but ignore them if you don’t get the roll you like. This can be good advice if used sparingly, but I think it bothers me enough that I’d rather avoid fudging if possible. It sort of trivializes things for me and feels dishonest in most games^[1].
2. Roll the dice, but have mechanics to alter the chance of success so that players mostly succeed. Maybe the players have a metacurrency to add points to their dice, or reroll, or they get to have 5e-style advantage / disadvantage. Now you’ve made the failure case less likely, but it’s still gonna happen. So… I’m unhappy less often, but I’m still not happy.
3. Do an interesting fail-forward thing, where the GM improvises a way the situation changes orthogonal to your actions. That sounds nice, but I can’t think on my feet consistently enough for this to be enjoyable for me. There’s a whole blog post floating around in my head about this, but it’s when a game designer says “just GM good for this part to work” they’re taking an easy out by offloading all the work onto the GM.

I’m a fan of Chris McDowall’s take on diegetic difficulty (TL;DR: different situations warrant different outcomes, sometimes the worst failure you could face isn’t that bad or the best success you can hope for is still pretty dire), but it sort of falls into the same pitfall as answer #3 from above. What if I just don’t have a very good idea for a consequence?

Well, then I probably shouldn’t ask someone to roll, at least according to some schools of thought. Maybe we should only roll if I have well-defined, interesting outcomes in mind^[2]. That’s a fine way to think and play, but rolling the dice is a form of percussion^[3]. Sometimes I feel like the rhythm calls for a random outcome, a little taste of gambling. What to do then?

I’ve come to call my preferred solution “consequence heatsinks.” They’re mechanics or techniques that let us take consequences from the present situation and store them in a nebulous, potential form to be revisited later.

The first time I ever really considered the consequence heatsink was at the end of a Blades campaign. It was my first time GM’ing, and one of the first tabletop campaigns I’d participated in. I had handed out Harm like candy, and the players felt like the game had been overly harsh. Some of that comes down to Blades’s harm & healing rules, but some of it… was because I was too eager for consequences to come in the here and now. I should have handed out more Heat! Lowered more faction status ratings! Ticked more long-term clocks! That way there’s still tension, there’s still a sense of stakes, there’s still consequence, without being punishing.

The Resistance system by RRD (used in Spire and Heart) features stress tracks that build up and create consequences. I’ve always felt like it worked somewhat oddly though, and one reason is that the stress system sits alongside traditional success and failure. There’s three main results when you roll, to my memory: success, success-and-take-stress, and fail-and-take-stress. Then stress can manifest as bad things later on (like your cover being blown in Spire or body horror shit in Heart). But if taking stress and do-you-succeed are covered by the same mechanic, then I can’t shunt failure into stress as a heatsink. That’s not necessarily a bad thing, but it makes the consequences for failing a roll even more nebulous and blurry and hard for me to adjudicate.

It may be controversial, but I actually think HP is a heatsink just like a clock in Blades. In a dungeon-crawling type game, most times when you lose HP nothing else happens. Struck by goblin? Lose 1d6 HP. Stick your hand in a dart trap? Lose 2d6 HP. The consequence is deferred until later, when you lose enough HP to die or be injured or whatever. This is a really useful function of a dungeon game, and I think when rejecting “big pool of hit points” this style of play becomes much harder to run.

Heatsinks don’t even have to be numeric. “You can bully this guy into letting you in to the conference, but roll to see if you hurt his feelings.” Or “you can break into the mayor’s office, but roll to see if they notice and announce it on the news.” These are harder to conjure up on the fly (numbers are so easy! picking off a menu as a GM is a huge decrease in cognitive load!) but they still represent a fundamental deferral. Deferring consequences is great, because it lets the current scene or situation play out smoothly while still building up a sense that actions and results matter.

Of course there’s the question of making sure your heatsinks aren’t too good at sinking heat^[4], because then you’re just back at “roll the dice and ignore the results” with extra steps. That’s the whole problem with HP, where each event (getting hit by a club, falling from height, being stabbed) is entirely weightless until the big shebang (keeling over dead). But folding events from play back into the future of the game is a subject for a different time.