Worlds of Design: Improvising the Adventure

When it comes to gamemastering, are you focused on improvising the world or making it believable? Let’s discuss the first.

Picture courtesy of Pixabay.​

I like to categorize as a way toward better understanding. Of the two styles, let’s discuss improvisation by the Game Master in making adventures (and settings). We’ll contrast that style with the second in the following “Worlds of Design” about the “believability” of the adventuring world.

The first style discussion can be seen as the question of how much the GM relies on improvisation. None of these styles are right or wrong; the key to making them work is that everyone (players and GM alike) buys into the style beforehand and that it meshes with their own expectations of play.

(When I refer to “player agency” below I’m talking about how much influence the player can have on the outcome of the game. See “The Tyranny and Freedom of Player Agency”)

Improvisation​

At the one extreme is the style of play as taught by the original Dungeons & Dragons rulesets. The GM improvises a lot, relies on wandering monster rolls and other dice tables, and may rely on the players to help determine where the adventure is going to lead within that adventure. This method is especially good for outdoors adventures as there’s no "dungeon" to make up.

Maximum player agency is achieved in one sense, yet in another they have less agency because so much that happens is random. “Strategizing” depends on logic and causality, and if the GM doesn’t know beforehand what’s actually “there,” then the GM cannot provide clues that would help the players strategize.

Conceivably a GM could use this style to try to tell a story but it might be difficult to impose it on the players. I’d say this is closer to the Situational style than to the Storytelling style.

Situational​

In the middle of the style spectrum is Situational, where the GM sets up a situation with (perhaps) objectives for the players—though there's always exploration and treasure-hunting as objectives—and lets the players "write the story." So there is some structure, but not imposition of activity. Players have strong agency—and if they are too passive, they may fail because they’re expected to be active. (See my first and second column about passive and active players).

This is more like a traditional boardgame than anything else (though most boardgames today are not designed with this in mind). There is a possibility of loss here, and it’s not the GM’s job to make everyone happy. It’s a game, not a storytelling session.

Storytelling​

At the other extreme the GM plans the adventure, and imposes that plan on the players—there is much less player agency. In other words, the GM is storytelling and the players are the listeners, though in RPGs they are more active than typical consumers in other media. A storyteller MUST control what happens in order to make the story work as planned. And a storyteller can reasonably work toward a situation where all the players have a chance to “shine”, where the GM can strongly concern themselves with the happiness of every player. A really good spontaneous storyteller—a rare skill!—can tell stories and still be improvisational.

The key to all this is that the players need to be happy with the style. If the players prefer one style and the GM prefers another then it’s possible that no one will be happy.

Your Turn: How much of an improviser are you?

 

[Ulfgeir]

A mix of both. In some cases it is basically, here is the current situation, you have this endgoal. What do you want to do to get there? Others I have planned more so that certain things has to happen in a certain order. Or at least, have planned that if this thing happens, then that response should happen.

Of course, no plan ever survives contact with the players.

 

[prabe]

I improvise a lot. I tend toward providing a situation, and seeing how the PCs handle it--often including what they decide their goals are. To the extent I do any consideration of possible future events, I mostly limit myself to what will happen if the PCs don't intervene, which gives me a baseline to change as the PCs act on the situation; sometimes there are obvious things to consider, though, and I do consider those.

 

[deganawida]

I guess I'm situational, with a tendency towards improv.

 

[Oofta]

I tend to improvise quite a bit, using the definitions above I'd say I'm situational. Maybe. I try to develop interesting environments, obstacles, a general idea of locations. Then I add in NPCs, organizations and motivations/goals for them, who's who and what are their goals? A beast may simply be hungry or defending territory, a succubus could be tempting a vulnerable city leader while convincing another of falsehoods in order to achieve her goals. The military commander of a local garrison may despise a local crime lord but is being pragmatic and knows there's little they can do.

Then I throw in various moving pieces to the plot, often with just the vaguest outlines. The PCs always get a choice on what threads they pick up or ignore, although I do ask that they decide which particular thread they're pursuing at the end of a game session so that I can prep for the next session.

When planning a session, I decide what encounters would be likely and consider possible triggers and events that would cause the encounter. I usually plan out an extra encounter or two so that I can improvise something on the fly. I usually don't plan out the physical environment in detail unless the PCs goal is to infiltrate a specific location. Usually I just make up the battlefield on the fly and fit it to whatever scene is appropriate.

So no random monsters, it will be monsters that make sense for the depth of Gloom Swamp. Perhaps they will encounter bullywugs defending their territory, which may lead to a fight. Perhaps they'll convince the bullywugs they can help by taking out that young black dragon that's threatening the bullywugs. Maybe they'll bypass the bullywugs somehow but still fight the dragon when it come across them while hunting or I'll go to encounter "C" because it makes the session more fun. Heck, maybe the bullywugs get their fluff changed and turn out to be malformed humans under a curse because I had a moment of inspiration to set up a higher level nemesis that could be an ongoing threat.

So I almost never have a set story in mind. I have threats, opportunities, my best attempt at a living breathing world that will respond to the PCs actions or lack of action. Small arcs may be fairly story driven because if you have to escape the prison of the giant lords, there's only going to be so many options. But once the group escapes? Up to them if they want revenge against their captors by following up on a possible alliance they learned about from a fellow prisoner or continue to pursue the golden McGuffin they were originally after before they ever knew there were giants in them there hills.

The PCs give the direction, I just set the stage.

 

[GMMichael]

+1 for cross-referencing. I've never thought of OD&D as extremely improvisational, but there it is.

. . . Of course, no plan ever survives contact with the players.

Truth. In this sense, every GM is an improviser. And you don't always need a plan, because . . .

Back in '94, AD&D2e went (almost) full-improv thanks to Kirk Botula's villains handbook. The "power matrix" was the GM's tool to combine a map, NPCs, and dependencies - with no plot - and let the chaos ensue from there.

Balancing this knowledge with my storyteller tendency is a lifelong struggle.

 

[Xetheral]

I improvise heavily, but I rarely rely on random generation. Instead I'm introducing improvised content deliberately, creating new content on the fly that (ideally) is: (A) a plausible or logical extrapolation from existing content, and (B) would cause the players' choices to lead to a good story. So (when/if I'm doing a good job) I feel like my approach avoids the loss of agency from random generation that the OP associates with Improvisation and avoids the loss of agency the OP associates with Storytelling. Players at my table can express their agency by making logical plans based on extrapolation from already established content, and they can express their agency by not being in confined in their choices to the bounds of a predetermined story.

One trick that I find essential when improvising content is to deliberately include details you don't need yet. This serves multiple purposes: (1) it gives you established content that later-improvised content can tie back to; (2) any such details you don't eventually tie back to subverts Checkov's Gun, making the game world feel more like a real place instead of a backdrop or setting; (3) it gives the players a much wider range of content to latch on to to drive the adventure forward; and (4) ensures that established content is created faster than it can be completed.

 

[practicalm]

I am definitely situational. There are beats that the antagonists are going to hit if the players don't do anything.
But the beats will adapt to player actions (the villain's plans never survive contact with the players)

If I use random encounters, the encounters have to be tied in to the overall story (or could be a start of a new story).

For example, my early campaigns may have 1-2 short term villain plots, 1 medium term villain plot, and 1 long term plot. As the players progress, plots multiply. Some are handled by NPCs, some the villains win, some the players handle. Plots can intertwine if they make sense to some are just there for color.

(For example a group of Ogres that really wanted to be pirates was a random encounter, the players could have defeated them, but instead the players gave the Ogre's access to a ship and a really large crocodile and thus they became a pirate ally sailing the seas and causing trouble for one of the villains chasing the players. )

 

[Krachek]

I prepare my game in order to improvise!
some npcs, some loose plot, some lore, some monsters and encounter build.
But the final render is usually much improvise.

 

[Jaeger]

Improvisation​

At the one extreme is the style of play as taught by the original Dungeons & Dragons rulesets. The GM improvises a lot, relies on wandering monster rolls and other dice tables, and may rely on the players to help determine where the adventure is going to lead within that adventure. This method is especially good for outdoors adventures as there’s no "dungeon" to make up.

I don't get that at all. Those tables are a GM aid; To be used at certain points when following the dungeon and wilderness play procedures. i.e. Wandering monster rolls and other dice tables are not rolled on without context, and the GM is to use situational judgement when interpreting their results.

Using your definitions - RPG play has always been a combination of a Situational set-up, + GM Improvisation.

Maximum player agency is achieved in one sense, yet in another they have less agency because so much that happens is random. “Strategizing” depends on logic and causality, and if the GM doesn’t know beforehand what’s actually “there,” then the GM cannot provide clues that would help the players strategize.

This only holds if Improve is done in isolation. RPG play is always a combination of Situational + GM Improv.

Situational​

In the middle of the style spectrum is Situational, where the GM sets up a situation with (perhaps) objectives for the players—though there's always exploration and treasure-hunting as objectives—and lets the players "write the story." So there is some structure, but not imposition of activity. Players have strong agency—and if they are too passive, they may fail because they’re expected to be active. (See my first and second column about passive and active players).

Active players is the default for RPG play.

You can get away with a passive player or so in a group of otherwise active players.

This is more like a traditional boardgame than anything else (though most boardgames today are not designed with this in mind). There is a possibility of loss here, and it’s not the GM’s job to make everyone happy. It’s a game, not a storytelling session.

RPG play to be truly engaging and dynamic there must be visceral in-game consequences for the PC's.

RPG's are derivative of wargames. The possibility of loss and failure are intrinsic to playing the game.

Those are good things.

Storytelling​

At the other extreme the GM plans the adventure, and imposes that plan on the players

RPGs are not Movies, novels, or comics. They are their own unique game media.

In my opinion: Trying to impose the dramatic structure of those media on a gaming session is counter productive to the goals of the RPG play paradigm; which is to maximize player agency of the PC's within the game setting.

When I was a much younger GM, I did my share of "storytelling" i.e. railroading PC's because of what I thought would make for a cool "game" session. To my everlasting shame.

Fortunately, I was able to identify that defect in my GM skills early on, and dedicate myself to eliminating it.

Making my games entirely (By the definitions given in the OP) Situational + Improv, with maximum player agency (with further situations emerging from PC actions within the game world).. My GM abilities have gotten much better; maximizing both player and GM engagement during the game session.

Divesting myself of the "GM as a Storyteller" conceit is the best thing I ever did to improve as a GM.

The "power matrix" was the GM's tool to combine a map, NPCs, and dependencies - with no plot - and let the chaos ensue from there.

^This^ The creation of situation play + GM improv is what makes RPG's great.

Unfortunately many early RPG's took this stuff as a common knowledge, and never explicitly laid it out in their rule books. This lack of practical 'how-to' for first time GM's is still a failing of most RPG's to this day!

I had to learn this stuff independently myself. But Everything is all the better for it.