There is an inherent tension between the desire to run an extended roleplaying game campaign over several adventures, all the attendant preparation required, and the need to just prep a few scenes for the next session. My problem has always been how to resolve that tension.

In the past, I have tried to set up and prep the grand vision for my campaign. I have decided on the win-state for the players and vigorously plotted out the steps towards that end. Inevitably, after many weeks of preparation, the players have turned left at Coventry and the whole thing unravelled.
Once I realised that my players probably don’t want or are not likely to follow my grand vision, I decided to embrace the way of the improvisational GM and simply begin with an initial problem and let the players decide how to resolve it. I trusted in the strategy of simply stringing together scenes in response to the players.
Eventually, while following the path of improvisation was easier for me, everyone else eventually realised that we were (to steal the Angry GM’s phrase) just “dicking around” in a sandbox and the game collapsed. What was missing from my improvisational approach was any sense of the game going anywhere. What exactly was the point?
This is why, for many years now, I have found myself creating bounded environments – such as large dungeons to explore – and running games with a clear set of default goals. In the dungeon, if you don’t know what else to do then you look for treasure. How do you do that? Well, you pick a door and enter the next room.
Bounded environments like dungeons don’t necessitate interesting characters. In fact, they generally suit a very optimised and somewhat restricted set of character abilities. This places the focus on the game system rather than on the characterisation or the description of the world. Thus, I often begin to get bored because I don’t enjoy too much focus on the “game” in RPGs.
Today, I am trying to reconcile the tension by setting up a serious potential villain with a plan to cause mayhem in the fantasy world. I’ve set up their first moves and the potential consequences if nobody tries to stop them. I’ve decided that the villain will be subject to the same game rules as the player characters – no “plot armour” or such – but that (for ease) a lack of opposition will allow their success.
The invitation at the table is to discover and then challenge the villain.
This approach gives the game a clear point – there is a major villain to thwart – without prescribing the approaches to that challenge nor the outcome. My hope is that this will lead to some engaging game play which exercises my improvisational skills, doesn’t limit the player’s sense of tactical infinity, and still offers a focused environment within which to play.
Most of all, my hope is that interesting characters arise to challenge the villain. Instead of optimised character builds, I hope the players can imagine people who will rise to tackle the situations which they face in interesting ways. Along the way, I look forward to exploring the interactions between the world, the villain, and the character group.
Game on!

Interesting. Sometimes I have a big plot or thrust in mind, but not always. And I never imagined hearing players complain of dicking around if I let them take the lead. After all, how many “player-directed” and “player agency” articles do I have to read? To my mind, I discuss what type of game we all want. I work out a setting. I may have some interesting events to run in the background, but I generally look for setting stability – I’ve run decade-plus campaigns more than once. I am prepared to drop adventure hints, I have NPCs with agendas to introduce, and I’m open to in-character PC ruminating about possible goals. “Someday I’ll make them pay,” or “a Merchant Prince is something to be.” (Apologies to Mr Lennon.) But what I don’t do is set up a Sauron, or even a Sheriff of Nottingham. But I recognize that there are fans of every plot type.
LikeLike
I find that “villain” can be interpreted as a variety of things worth opposing, but the basic procedure is the same. I tend to use systems where I can easily assess if it is ok to assume success for the events caused by that thing or if I would prefer to roll (BRP, Ubiquity, etc) but for this type of open-ended play, I hear you on this~
LikeLiked by 1 person