Ladino's Rogue Thoughts

Not All Improv is the Same

In the spirit of providing lenses with which to examine our RPG practices from different perspectives, here’s one possible way to categorize different forms of improvisation during a session.

X axis; Y axis

Because we all love a bidimensional graph, let’s work with two axes that define different types of improv at the game table: “based on prep vs. primarily reactive” and “obvious game impact vs. obscure game impact”. This will give us four nice quadrants representing different possible "areas" of improvising new facts, elements or changes to the fiction.

[If I ever upgrade to an image-having blog, this is where an image of the graph goes - but you can imagine a graph with two dimensions yourself, I'm sure, really doesn’t matter where you put each axis and their orientations]

Like most things I write, this framework generally assumes separate GM and player roles and focuses on GM improvisation. But it might still hold up for GMless / GMfull / Solo games if you abstract the ideas here sufficiently? One hopes, anyway.

But to explain the axes in a bit more detail:

Prepped vs Reactive refers to the fact that some improvisation is built on top of a strong foundation of pre-established truths and prepared tools; while other times the improv is more reactive to things that have been established in the fiction right then and there, to the input of other players, and no tools were devised to help flesh them out in advance.

For example: coming up with NPC dialogue for an NPC you’ve just made up, vs. one you’ve already established a simple motivation for, vs. one that has an extensive backstory and has been roleplayed before, are at three different points in the ‘prepped’ to ‘reactive’ axis.

Most everything that is improvised at the table has some amount of established information working as its foundation, even if it is just the preceding fiction up to the point when the improvisation happens. Similarly, everything that is improvised is also to some extent brought to bear reactively, in response to prompts made by other players. Still, the amount of forethought given before the session started to the element that is being improvised affects the experience of improvising it.

It shouldn’t be assumed, however, that having more prior information and tools always makes it easier: having to keep too much information in mind while trying to improvise new details can be a strain on a GM’s cognitive load. It is also worth noting that, when non-GM players get to improvise new details, the further in the “based on prep” (in this case, based on information previously established by the GM) side it is, the more likely someone is to forget a piece of established information and inadvertently contradict it (which may or may not be a problem).

Obvious vs. Obscure is about the degree to which you can foresee how the improvised elements you’re changing or adding to the fiction will have a significant impact on the current fictional positioning and how it will develop forward - usually because it intersects with the players goals, or the obstacles posed to them achieving those goals.

For example: deciding whether there are any rats living in the basement of a location of interest to the players, when that wasn't already defined in the prep, involves different considerations when they’re introduced simply as "scene dressing" vs. when they might serve as informants and tools to a PC who is well-established as being able to speak with and command animals.

When an element will have an obvious impact on how the situation develops (whether it impacts a problem that the players are meant to solve or the direction a narrative they’re trying to author goes) any improvised truths about said element have the added pressure of being considered from a “game design” perspective (for whatever you want the game to do). So an element that everyone can predict how it will help or hinder the players overcome an obstacle or help of hinder the players hit a dramatic beat both impact the improvisation in similar ways: namely, making the table aware that it is a choice that will have repercussions.

Some people report that they have no problem acting neutral with regards to this design consideration, being able to ignore (when desired) the effects the improvised element will have on the fiction going forward and decide upon them on the basis of some other factor. But my experience, which no doubt isn’t unique to me, is that the specter of how the element will affect the situation going forward is always in the back of my mind, becoming yet another plate to spin while trying to come up with something for the table - which is why I propose it here as a relevant aspect that will impact the improv.

On the other hand, improvisational elements that don’t have an obvious impact on what choices and consequences will be available to the characters may still end up being very impactful! But because that impact can’t be easily anticipated, that makes it easier to forego any consideration of how the element will affect the affordances, fictional positioning and choices available - for good and ill, since that sometimes means inadvertently making choices that make up for a bad gameplay experience (however that is defined for a given group).

Let’s look at our quadrants

Combining these two axes, we get four major quadrants of GM improvisation (I could have set up a D&D alignment-style grid instead, but I’m not coming up with nine whole ass types of improv - you do it if you feel like it):

Based on prep with obvious impact: takes player goals, affordances and fictional positioning into account from the outset to make sure the impact of the GM’s improvisation satisfies some desired principle or expectation of play.

Maybe it creates tools that allows the GM to disclose the improvised choice to a random output, to prevent the GM from “tipping the scales” for or against the players; maybe it is meant to specifically allow or disallow a given player tactic in the interest of ‘balance’, or to ‘promote a feeling of power’, or to ‘ensure all character options are valid’. The principles themselves vary, but the technique of trying to prepare in advance to make sure aren’t forced by your own improv to break them is usually at the heart of this quadrant of improv.

E.g. after rolling a random encounter of 2d6 gerblins, disclosing the decision of which ones are using bludgeoning weapons to a 1-in-6 chance, because you know one of the PCs has resistance against such weapons.

Based on prep with obscure impact improv: this is usually the realm where the GM looks for help in creating, mid-session, details that bring the world to life.

While it usually covers elements that are low-stakes in terms of their expected consequences to the progression of the fiction (we don’t usually expect an NPC’s accent will be the key to dealing with a situation, for instance), they tend to be details that enhance the suspension of disbelief if properly used during the session - which justifies prepping them specifically. Thus, a GM who isn’t proficient in providing certain details off-the-cuff, and finds that this undermines their games, may well go looking for ways to facilitate their improvising when needed - thus arriving at this quadrant of improv. Of course, it is always possible that the elements that are improvised based on the prep will end up being impactful to the fictional positioning - but that was not the a point of consideration when it was being prepped.

E.g. a table of names and distinctive traits to roll or pick from whenever the PCs engage with a heretofore undefined NPC. A simple indication that a room is a bedroom, allowing a GM to improvise all the furniture that is in there.

Reactive with obvious impact improv: this is where the potential conflicts of interest inherent in the GM position can really rear their ugly heads, since the GM has to make an improvised decision that will clearly affect the PCs’ position in the fiction, but doesn’t have much prepared as a foundation from which to make this decision.

Even if a GM comes up with a way to disclose that decision to chance, they are deciding the odds after already knowing what the stakes are - so what odds do you pick? Call this “how do you decide whether to throw paper or scissors if you’ve already seen rock?” problem, or maybe the tyranny of human intuition.

All that said, it is in no way impossible to do the improvisation in this quadrant. It just means that you’re making impactful on-the-spot decisions about how the fiction will progress going forward, by whatever metric of your best judgement you deem best for the game. Anything that you and your table are comfortable disclosing to your judgement, despite being decisions that clearly impact the challenges of the game, can comfortably live in this quadrant of prep.

E.g. the players want to shock a monstrous snail at a distance by targeting the trail of slime it leaves behind, assuming it will act as a good conductor and carry the current all the way to the beast. If nothing in the GM’s prep suggests an answer one way or the other, deciding on the slime’s conductiveness falls on this quadrant: if it works, it is a significant tactical advantage; if it doesn’t, a wasted action that may have serious consequences.

Reactive with obscure impact improv: this is when the GM is giving answers and painting the picture of what’s happening in the fiction, and the possible consequences and applications of whatever is being established or changed aren’t immediately pressing.

The details that live in this quadrant of prep are usually the ones a GM is most comfortable creating on the fly, the kind of description/narration/answers that are believably true to the fiction and to the players, and that get to be set up as undefined opportunities for the players.

E.g. the players decide to have their characters enter a pub to discuss the next steps of their plan; the GM has nothing specific about pubs prepared, but adjudicates the general layout of the place, as well as a general sense of the staff and patrons, based on their best judgement.

Further Reading

#improv #lenses #theory #tools