Finding the fun: reward cycles

Since the Forge has lost some currency in popular RPG discourse, we don’t much talk about its theoretical fruits. In the following blog post, I’ll recover and clarify the Forgie notion “reward cycle”. Once it’s developed, we can in future work untangle confusions around “roll vs role play”, fun, incentive, and the overall role of rules in a game. We can talk about how specific rules help play groups achieve their creative ends.

Group RPGs are social games. When I play well, I get a collective affective response. When things are going well, creatively, I can literally see it on the face of everyone playing. We’re all leaning in. (Note that “things going well, creatively” is not the same as “things going well for our characters”, as I will discuss.) Likewise I can see it when we’re not creatively engaged, because we’re leaning back, our eyes are neither widened nor narrowed.

In some games, “good play” means the same as “tactically effective play”, stepping up to the bat, meeting a challenge. In such a game, I lean in when I see someone making a bold, well-considered maneuver, or when a new challenge is introduced. I’m likewise hooked when I get to see the results of those maneuvers, whatever those results are.

Example of a good maneuver: The main characters have ambushed some monsters and the survivors are running back to their base. They’re right at the door. I have my character charge past them and block the way. He’s alone, and he might die, but he’ll slow the monsters down enough that the rest of the group can kill them before they alert the base.

By contrast, I check right out around sloppiness, silliness, laziness, and foolishness. Dickering too. Lengthy white room strategizing too.

Going forward, I’m going to call a maneuver “good” or “brilliant” or whatever when it gets the play group’s approval, and “bad” when it doesn’t. These evaluations are always relative to a single play group.

Example of evaluation relative to a play group: I once played a session of a game where players could buy a day’s rations, or a pack of 12 eggs, for the same price. One player suggested that they buy eggs, and everybody, including our horse, could simply eat one egg per day. This was almost unanimously hailed as a brilliant maneuver. I disagreed, and I did not play with that group again; its standards for good play were too alien to me.

Example, ditto: The main characters enter a dangerous area and encounter their first token opposition. They win a fight, though they are slightly wounded. One player suggests that they immediately retreat to regroup and strategize. Everyone nods slightly in approval. Good sober judgement. In another game, months later, with an entirely different play group, we find ourselves in the same situation. One player suggests the characters should retreat to regroup and strategize. Another player objects: “I know you guys play a bit differently – I don’t know if – I hear about simulationism – but I’m here to have fun”. The rest of the play group agrees: retreating at this juncture would be bizarre.

A game’s reward cycle is the process in which it reinforces or intensifies the play group’s collective affective response to fictional events. I’m paid off by a reward cycle through further fictional events, some kind of formal game-mechanical consequences, or both. If I’m in a reward cycle, my contribution matters. I can only identify a reward cycle after the fact, when I see that the play group is in fact more pumped up, or more devastated, or whatever, by the game’s next move, than they were just before it. Through experience, one can get a feel for game structures that tend to produce reward cycles, but this sort of thing always has to be established experimentally, never a priori.

When I come up with a brilliant maneuver, and it goes off without a hitch, so that I hose my enemy, I’m in a reward cycle. When my enemy comes up with a brilliant maneuver, and it goes off without a hitch, so that I am hosed, I’m in a reward cycle. When I play poorly, and I am hosed, I’m in a reward cycle. When I sacrifice an important resource, and it really is gone, and I feel its loss, I’m in a reward cycle.

Example of a reward cycle: After much fictional struggle and tactical brilliance, the main characters have snuck their way into their enemy’s headquarters, where they catch the enemy leader unawares. They kill him. Shouts of triumph from the real-life players. When they next meet their enemies, they are disorganized, and have made poor plans. More shouts from the players.

Example, ditto: A monster is chucking rocks at the main characters, who are behind heavy cover. My character has a bow and arrows. I suggest that it’s worth leaning out of cover to shoot the monster, and I do so. Following a resolution procedure, my character is hit with a rock and dies. I literally scream.

Example, ditto: My character has gone through many trials and tribulations, and achieved her goal. I have reached the then-peak of my strategic ability. Unfortunately, she is almost dead. She runs into a potential enemy. I do not think I am strong enough to fight him. He will let her pass, but he demands a fearsome toll. (I won’t even write it out, it’s magical realm bullshit, blame Adam.) She pays the toll, and is marked by it for the remainder of the game. Everybody else laughs in disgust. I feel my stomach curdle.

The reward cycle might be absent, or at least, not immediately present in any given action in functional play. Most of our actions are not themselves entire reward cycles.

Examples of “mundane” actions, not immediately part of a reward cycle: (For the following, imagine that there’s no particular affective response in the play group.) “I attack.” [Procedure.] “He dies.” | “I search the room.” [Procedure.] “10 minutes pass. You find [whatever].” (On the other hand: “That description sounded a little hinky… I want to search that room!” [Procedure.] “10 minutes pass. You find a secret door on the north wall!”. Everybody’s pumped. Or: “10 minutes pass. You don’t find anything.” “Damn!” What seemed like a good lead was just a waste of time. – Can you see the difference?These latter exchanges are expressions of player skill, part of reward cycles.)

We might miss a reward cycle where we expect to see one.

Example of a missing reward cycle: “You said he’s wounded badly, particularly his sword arm. Ok, I press my advantage, driving hard from his right!” Everyone is impressed with the player’s mastery of the fiction. The play group goes through their normal resolution procedure, which does not take this specific fiction into account. In some different game, this would have been part of a reward cycle. (Not saying those are better or worse games, just that this game had an opportunity for a reward cycle, and for whatever reason doesn’t have one here.)

We might find that we inadvertently reward play which we don’t actually think is good.

Example of a dysfunctional reward cycle: I notice that we spend a lot of time having low impact conversations with side characters. I don’t care about the side characters, and I’m bored. “Ok, I punch the dude in the face.” What! Everybody reacts. Straight to combat. Now we’re cooking – from my point of view, anyway. Nobody else is happy. If I can only make an impact on the game by acting in a bizarre, disruptive fashion, our game isn’t functional.

A game doesn’t need a reward cycle for something to be worth doing in that game.

Example: A player describes his main character, an immense, immensely hairy woman, climbing out of the ocean. She shakes her hair. I ask: “Like a L’Oreal model, or like a dog?”. Laughter. We don’t need to answer the question. Obviously she shakes like a dog. This was a fine question, which contributed to our understanding of the fiction, but which wasn’t part of any reward cycle, didn’t hook up with any part of the game later.

As a rule of thumb, a game based around tactical skill and strategy should make losses hurt, fictionally. Crucially, this punishment should be meted out by the game itself; that’s what makes the punishment part of the game.

I need to be clear about a few things, because I worry that the reader might think I want GMs to grade plans and give good outcomes to good plans and bad outcomes to bad plans. On the contrary: I want the GM to do whatever they normally do, and for players to notice that they often get good outcomes from good plans and bad outcomes from bad plans. In the next section I’ll say what I want a GM to “normally” do, and in the one after that, what I mean by “game”.

What does a GM normally do? – There are plenty of possible sets of responsibility. Most of my readers are in the OSR, and this essay is mostly about stepping up to challenges, so I’ll look at a neutral referee in a typical OSR game. They should:

• Prepare a scenario for play, using whatever tools are standard
• Frame scenes, including describing locations and marking the passage of time
• Tell other players whatever their characters would know
• Play their characters, on and off screen
• Adjudicate task resolution
• Make models
• And that’s it, literally nothing else.

You should play everything at its usual scale. So you should probably never say, for instance, “Yeah, that’s a good plan, roll 1d6, on a 5 or less the keep is yours”. (Conceivably this is part of an avant-garde game, but it has little to do with anything I’ve ever played.)

Example of a good maneuver, a neutral referee, and a reward cycle: The main characters are in a dangerous place. The ground ahead is dotted with man-sized holes, some of which probably contain monsters. We want to draw them out and kill them one by one. I have my character advance alone, sans heavy armor, and toss rocks into holes, waiting a little while between each throw. (At least I think that was a pretty good plan!) A monster emerges from a hole. My character, who is very fast, tries to run away to his allies. We go to a resolution procedure, and my character trips. The monster catches up to him and kills him. “I can’t believe he tripped! What were the odds of that?” “One in twenty, as you saw.” 

“Game” here doesn’t just mean published rules text, though it does also mean that. Mostly though it means whatever set of techniques, distributions of authority, constraints on authority, and resolution procedures your play group in fact uses. Following are some possible parts of a game that might not be in any rules text. They’re in the game if the play group does them.

• Write a little two-sentence bio.
• Make two characters at once, because one will likely die soon.
• Since your character died, do you want to play that guard until the end of session?
• Feel free to add details about [this place], your character is from there.
• Nat 20, nice! Tell me how you do it.
• We’re playing this one seriously. No murder-hoboes.
• I actually use the dungeon stocking tables from OD&D.
• We’ll play flashback scenes whenever something interesting prompts it.
• Don’t bother rolling for that, it’ll obviously work. (Or: Don’t bother rolling for that, there’s no way it’ll work.)

And so on. You can see how various the components of a game might be.

A reward cycle can be long or short, and shorter cycles can be nested inside longer cycles.

Example of nested reward cycles: Adam organizes the play group to conquer a town. His plan is good, he has enough force to succeed. One player makes a bold maneuver, breaking his troops off and charging through the town hall, in turn dividing the enemy troops. Shortest reward cycle. Adam’s whole army wins the day. Reward cycle. Several characters improve in formal game-mechanical terms. Reward cycle. But ultimately taking the town was an expensive endeavor, in lives and gold, which cost is compounded by bad weather. Reward cycle. But by conquering the town he was ultimately able to rule it, once he built his forces back up. Longest reward cycle. The short cycle here took about five minutes to resolve; the longest, several real-life months.

I want lots of nested reward cycles in my games. If I’ve only got immediate, short reward cycles, I’m not establishing anything permanent. I’m only seeing payoffs from the past five minutes of play, which is nothing. (You’ll get this when you play “conch-passing” games, where anything can and does happen, based solely on the whims of the current speaker. Not recommended.) If I only have long reward cycles – if I only get excited when I level up, or my character makes the occasional lasting impact on the world – I’m going to be bored most of the time waiting for that payoff. (In the 1980s and 1990s, this was a very common dysfunction kind of play.)

Reward cycles aren’t the same as gameplay loops.

In the 1977 Holmes D&D rules, the DM describes the situation, the caller asks a question or states their action, we follow any necessary resolution procedure, and the DM describes the result, which is in turn a new situation. In some of these gameplay loops we’ll find a reward cycle. Sometimes we’ll find multiple reward cycles within that single loop. Or we might find none, for quite some time.

In old-school D&D, an individual round in combat is rarely the event of a full reward cycle. Usually we’re more interested in combat as a whole. (In which case, one might wonder, is the handling time of D&D’s combat worth its cumulative effect on the reward cycle? Historically many designers have tried to juice up D&D’s combat model by adding detail or making its results more unpredictable.)

There’s more to a game than its reward cycle. Tunnels & Trolls and old-school D&D can have nearly identical reward cycles, but are nevertheless very different games.

Of course there are all sorts of reward cycles for all sorts of games, not just games about stepping up to a challenge. In some games, we want a reward cycle that triggers when I, the player, make a big thematic statement, when I show what my character cares about.

Example of a reward cycle without challenge: Last week I played a game in which my character tried to psych out her husband’s affair partner. I was making a statement: this is what is important to me, right now, and it’s worth using my force of will to get it. We move to the game’s resolution procedure, and we find out that my character has failed. She’s humiliated. Reward cycle. I put my values on the line, and I got a big return.

Next time you’re playing, pay attention to these nested reward cycles. How quick are they coming? How satisfying is each cycle? Are we all on the same page about the pace of the game, and the things we’re rewarding? If everybody thinks a particular maneuver is great, does it have enough effect?

When you notice someone making an important maneuver, reinforce it, socially and fictionally. You don’t have to give them what they want – I don’t think you should ever give the bastards an inch – but you should mark that you’ve heard them, and let the fiction reflect that.

I’d like to hear about some reward cycles from your recent games. They’re my favorite part of actual play reports. They’re the reason I don’t like in-fiction play reports; I care more about the people playing.