TL;DR: I’m all for playing to discover what happens, but sometimes I’d rather the surprise be the path traveled, not the destination. There’s a lot of flavor text you can skip about me and abstract algebra, so you may want to jump to the practical part.

If you like what you read on The Polverine, please share it with your gaming group, on Facebook, Twitter, or Discord. I’m not asking for myself, more for a friend—really, a bunch of friends: the devs behind Fari. And if you really like what you read, do me a small favor: support Fari on Patreon.

A Rant

When I started playing TTRPGs, improvisation was somewhat frowned upon in my gaming groups. Our GMs would sometimes improvise when things were dragging on, and they needed to break the monotony. Which, more often than not, amounted to roll on a random encounter table. Other than that, improvisation was equated with a lack of preparation. We expected GMs to have something for us in store. We were passengers on their ship, enjoying the cruise. I’m sure there were already plenty of GMs who improvised stuff for their players. But not our GMs. Our GMs had to be prepared.

Things changed somewhat when one of the two groups I was involved with began to play in earnest with a generic homebrew system. I was tasked to adapt the rules to a cyberpunk-syle universe (which I did) and to run a game or two (which I also did). That became the permanently-split party game I mentioned in the last episode. In that game, improvisation was the rule, but it was edgy. Improvisation was a means to an end—advance the plot—and we agreed to it because it was the lesser of two evils (the other being railroading). 

Fast forward thirty years or so. My old gaming groups have long disbanded. I have a new real-life one and am involved in a few online games. And in all the games I’m a player in that are more than one-shots, improvisation seems to be expected (with one exception, but I’m neither positive nor optimistic about it [Sept. 2021 Edit: Gut feeling spot on, the GM nuked the game server with no warning after 3 sessions.] This post is not the place for a deep analysis, but figuring out what happened does not require one either. Apocalypse World happened, and PtbA games have filled the niches D&D did not care for. And with the unprecedented success of the Avatar Kickstarter, it’s clear as day that PtbA is the way of the future. And with it, improvisation. And now, for some LaTeX humor.

\begin{rant}
I’m painfully aware that my system of choice, Fate, is fighting a rear-guard battle. And it’s also true of last-gen, plot-points-centric systems like Savage World, Cortex+ and 2D20. SW has a popular IP with decades of a cult following (Deadlands), so I’d wager it’s here to stay for at least another decade. Cortex+ and 2D20 are churning intellectual properties with a varying level of consistency, and that seems to work well enough to maintain a player base. As for Fate, well, that’s a story for Reddit. But the short of it is that Evil Hat has seen the writing on the wall before Margaret Weiss Publishing (Cortex+) and Modiphiüs (2D20): Monster of the Week and Blades in the Dark run on PtbA. I’m ready to bet that the mainstream entertainment media will call Magpie Games (Avatar) a “disruptor” at some point. Still, they just really combined the two trends of being PtbA and latching on IPs.

Now, I love PtbA systems as a player, but not as much as a GM. For one thing, I love to roll for my NPCs. And for another, I still like to have a strong plot. At first, I didn’t like that PtbA had brought back classes (with the playbooks). But I came around and embraced the idea when I realized that it was yet another new soup in the old pot where character templates are cooked. And I have plans to bring them even closer (hint: I’d use a mechanics from Cortex+ that the last edition of SW has also pilfered). But if I could have GM rolls and plot points wrapped up in a PtbA system, I’d transition to it today. I just haven’t found out yet how. I’m still searching.
\end{rant}

If you love solid plots, the best advice about improvising hardly applies to you. Take PtbA games, usually choke-full of great tips. The chapter on GM-ing from the original Apocalypse World still being my favorite—you can download the first edition for free there. (Still, you know what I think you should think about my preferences, right?) Well, they’re geared towards making-the-plot-as-you-go, where “you” is not the GM alone but the whole group. But what if you’re all for playing to discover what happens, only you’d rather the surprise be about the path traveled, not the destination? Well, you’re stuck in the middle ground between improv and railroading. Better start planning for it.

Applied Abstract Algebra

Planning for improvisation may read like an oxymoron, but it really isn’t. I like to think of it as a problem of lattice theory because I can’t think in words. I’m much better at ideating abstract algebra structures, but that’s just me. Now, before you start thinking I’m just making shit up to sound geeky, have a look at Fig. 1, and you’ll begin to get what I mean. And when I say “begin to get,” it’s not because I have a poor opinion of your understanding abilities. It’s because of a feature of the dashed lines that I left implicit. So, let’s spend some time with them, shall we?

In Fig.1, dashed lines represent accessibility: if a box B has a dashed line connecting it to a box B’, B has access to B’ (equivalently, B’ is accessible from B). Since the boxes represent scenes, accessibility represents the existence of a story path. Now, as depicted, it seems that to go from Main Plot Step #n to Main Plot Step #n+1, you have to go to at least one side-quest, and that to go from Side Quest #1 to Side Quest #3, you have to go through Side Quest #2. But it’s not necessarily the case if you think about accessibility somewhat more abstractly and assume that it is transitive: if a box B has access to a box B’ and B’ has access to B” then B has access to B” too. In plain words, there’s a possible story path from Main Plot Step #n to Main Plot Step #n+1, and there are multiple story paths going through multiple side quests in any order. 

So far, there’s nothing in Fig. 1 about improv. If you are familiar with computer RPG quest design, you’ll realize that Fig.1 might as well be how one would plan fully scripted side-quests with Main Plot Step #n+1 as a point of no return. And if you’re a TTRPG GM, you might as well do the same and plan fully scripted side-quest scenes on index cards that you’d pop in the order your players got through them, and that would not be improv either. How do we get from here to improvisation, then? Well, easily, really.

Planned for Improv

In Part 1, I insisted on how Fate co-opts vocabulary that could describe any RPG gameplay and turns it into game mechanics. I’m not going to repeat my argument here. I’ll just pop the Fate stuff and assume that readers can do the mental gymnastics to turn it into other-than-Fate advice. Also in Part 1, I presented the structure for side quests you need to plan for. Fig. 2 is pulled from that post, and once we have it, we just need some guidelines.

All-Game Features

Let’s take a tour through the features of the index card from left to right and top to bottom. I’ll stick to what applies to any game if you interpret the Fate vocabulary as non-game specific. First, with the main Side-Quest index card.

• [Support For] What makes the side quest a side quest. You can visualize it as a label that you’d put on the dashed arrow between the last main-quest scene and the scene for that side quest. It answers the question, why would the PCs want to do this? 
• [Goal] What makes the side quest a quest. You can view this as the end-point of the scene for the side-quest. Answers the question, what should the PCs strive to achieve here?
• Side-Quest Completion. This one’s optional. It’s most useful when the side-quest is structured as a challenge-in-the-challenge (see Part 2 for some improvised challenges) or a contest (I’ll come to improvised contests in a future post). But you can also add a section whenever players add new stuff that prompt a new skill roll, as a visualization of how long they should expect to be off-track from the main plot.

Anything right of the main card is player input. You may add some GM secret sauce (particularly secrets), but you should let the players in charge of the narrative. Consider what follows as “best practice” recommendations to stay in control of the game.

• Location. A description of the principal area the side-quest is going to take place in. Try to keep it one location, whether it’s small (e.g., a pawn shop) or large (a city), using “zoning” and making the zones into scene aspects (e.g., the back alley, the marketplace, etc.). The secret aspects are there primarily for balance and a bit more Fate-specific. Still, it’s a good place-holder for whatever springs to the GM’s imagination when it snowballs on the players’.
• Obstacle. Anything that could stand in the way of a PC must be overcome to complete the goal and is not an NPC—or at least, an NPC worth description (the lookout in the back-alley of the pawnshop doesn’t need more details than the lock on the backdoor). The aspect should say everything there is to know to overcome the obstacle.
• NPC. If the location is the place, NPC #1 is its face. You could have multiple NPCs, but keeping the side-quest well-contained is best achieved by limiting your players to one NPC deserving of their own aspects and secrets.

Now, if you’re not a Fate GM, you may want to skip the next sub-section, which is mainly concerned with how to set (and balance) the difficulty of your side-quest based on Fate mechanics. Then again, assuming an agile mind and a good grasp of your favorite game’s mechanics, you could still get something out of it, so you may want to keep reading.

Making a Side-Quest Challenging

If you remember Part I, the side quest triggers when a PC has to take an Overcome action in an ongoing Challenge, you offered them to play the role as side-quest, and they took the offer. As mentioned earlier, it’s their show–but you’re still the executive producer. They’re in charge of the story, and you take care of the mechanics.

1. Ask the player who initiated the scene for the location and make a public aspect out of it. Let them ad-lib details and get a rough mental picture of possible zones (e.g., the pawnshop has two zones, the shop proper, and a back alley). 
2. Ask the other players for details about the main NPC and the NPC’s relation to the location. Make aspects out of what they tell you, distribute them between public and secret (for instance, a back-alley door with a combination lock is public, while the combination code is a secret aspect).
3. Determine, per table consensus, which of the public aspects favor the PC‘s task, which are neutral, and which are obstacles (here, you can turn your back-alley door into a Block).

There’s no shortage of good advice on how to adjust the difficulty in Fate. Oftentimes, it’s implicit level-scaling. For my part, I prefer to take the Adjective Ladder as a reference point, but that’s a topic for another day. There’s enough theory in that post already, so I’ll stick to the guidelines I follow and tell you what I’d do if I were you (because that’s what I do) rather than claiming that it’s what you should do.

1. I’d start an Average (+1) difficulty rating because Average is, well, average. Then again, you may prefer Fair (+2)—because Fair’s fair. 
2. I’d add +1/+0/-1 for each public aspect that opposes/does not affect/favors the PC’s task. 
3. I’d use the difficulty from step #2 as a guideline for the NPCs’ skills and the ratings of obstacles.
4. Next, I’d check my ratings against the guidelines for low, moderate, and high difficulties (Fate Condensed, p. 42) and consider the need to further adjust for good narrative and/or drama.

Now, I would not consider Step #4 as a license to fudge numbers. I would not make a back-alley combination lock Superb (+5) because my PC has a Great (+4) Burglary, and I want the side-quest to be challenging. That makes little narrative sense. The lock is undoubtedly meant to keep the Average (+1) curious out, as well as the Fair (+2) B&E specialist, and perhaps even a Good (+3) one. But that’s it. It’s a pawnshop, not a bank, and there’s no reason that the pawnshop owner would plan for that one occasion a Great (+4) burglar wants something from their shop.

If I think my player needs a challenge, I prefer to adjust the difficulty with secret aspects that, undiscovered, complicate the task. The lock may be a Good (+3) one, but what if I add an Average (+1) trap to it? If the player does not tell me they are looking for something odd, only passive opposition counts. And what’s the passive opposition of, say, a suede glove versus a poison dart? An just like that, the whole thing could end up with a Success on the Burglary roll and a Moderate Poisoned consequence. I have other tricks, but this post is way too long already, so I’ll wrap things up for today.

Wrapping up

The plot-versus-improvisation is a made-up dichotomy that I use as a strawman. I tend to consider that a good TTRPG session should stand the middle ground between total improvisation and straight-up railroading. Then again, that’s me, and again, I told you multiple times how much stock I think you should put in my opinion. The take-home point from this post is that you can stand that middle-ground with minimal prep.

And that’s all for today, folks.