I have been thinking about endings on TTRPGs for a while now. Some coherent thoughts have agglutinated around an unexpected work.
Coffee & Chaos is a small game that we were playing for our main critical analysis articles1. The game is about taking characters and dropping them into “coffee house AU” and other fanfic tropes and applying the pressure to produce that type of cooperative storytelling with determined hyperfocus. Sure, there is supposed to be some coffee shop management going on, but that part of the game is not fun at all — it feels like a chore that your characters — displaced into employees — need to get through and will get through as fast as possible to get to what the game really wants you to do: taking breaks.
The game does that by having cutlery — three kinds used to solve different problems (knives, spoons and forks) with different approaches and the kind that matters (tea spoons). Players go through their cutlery, and when they run out, to refresh the other cutlery, they take a break with any other number of players. This is where the game does the thing it wants to do. Then you go back to the chore-like game, get rid of the cutlery, use tea spoon… and so on and so on.
A funny thing happens, however. There are no rules beyond this loop. There is no way to end the game or to close the coffee shop for another day. You will run out of cutlery and those precious tea spoons. Catastrophes will keep piling up, things get worse and worse, work becomes more nightmarish and things go worse and worse and you cannot succeed — you can only alleviate the crisis for another moment. And you cannot take breaks, you are forever there.
Of course, nobody plays that. You playing for the tea spoons moments, silly. So, the moment there are no (tea) spoons, the game is obviously over. If you are playing this game, the assumptions of what you are trying to get from it needed to play the game in the first place also assure you will “play it wrong” in this way. Nobody will, realistically, play into the Coffee Shop Abyss of Doom.
There are two ways one can react to this: one, is to consider this a failure about ending the game — the game does not have something as simple as telling you to close the shop when running out of cutlery does not stop you from continuing to play deep into the point where it stops being fun and becomes miserable; the other is to regard the game as having a well-established ending and everything about it guiding you to that ending, and being so obvious based on assumed buy-in that one cannot fault such a small game to devote time and design to what it is not2.
These two most likely reads made me think a lot about endings and come to some thinking about that. TTRPGs have a problem with endings, one of the most difficult and yet important part of any artform of storytelling. Most curious, many games that say they are played to and ending and say they work towards it, fail more often than not. And here it is, this small game that “does not even have an ending”, that succeeds at giving itself an ending by the way everything comes together and delivering you over and over again to a natural endpoint to the story. So even if you consider this gamed flawed at ending itself, how come it still outperforms most cooperative collaborative storytelling at endings?
The Forever Game
The truth is, most TTRPGs don’t really think about endings or even consider that worth thinking about, much less problematizing; there are plenty of “ending enough” in them. At some point you will get to max level, get all options, a campaign will end, or some thing that it may serve as an ending and the game ends. Most of the time, this has an attitude that the game, if let to its own devices would go on forever.
Taking this attitude to endings leaves to the situation we are most likely to experience: games will be at the mercy of forces outside of the artform, and those are the ones determining how a TTRPG ends.
The most common is, a game — and the collaborative cooperative storytelling narrative of it — ends when you are no longer able to play it. Pressures of post-late stage capitalism tragedies aside, it also avoids having to think or drive to an ending. And in fact, considering the franchise-logic and database-narrative that dominate IP enclosure capitalism, endings are undesirable.
The fact that you can play a narrative forever, that there is no end, comes from something in which there is no interest to explore, but a feature of TTRPGs. Games are sold as Forever Games, where you can be a franchise of yourself, going forever and ever and ever and ever. This comes on all levels: from the enclosure-capture of Hasbro that prevents you from playing any other game — or even acknowledging they exist, to database-narrative in actual plays and streaming, to even indies taking it as a given even if not advertised — like how BitD assumes characters retire and burnout, but the Crew will keep growing until it swallows the city; or Wanderhome presents the same never-ending cycle of tourism which you can bring into any other Wanderhome game with all their database-narrative into another game.
So, if even when we think about it, game endings being the way they are is not a bug but a feature, no wonder it is such an underdeveloped pillar of storytelling.
One Shot Containment
Of course, everyone has been sitting on the edge of their seats with the obvious question in the tip of their tongue:
“What about one-shot games?”
It is pretty obvious that those games have to consider endings: they cannot do any serious design without being aware driving to the ending will be a concern of most of the cooperative storytelling. So it comes to no surprise that some of the best endings and the most tech for them comes from one-shot games. Of course, just because something is really important to get right, that does not mean many do.
Coffee & Chaos, as described above, does this well. Knows it will end, drives to that. For a more widely known example, this is part of the outstanding success of The Quiet Year: it tells you how to end, when it ends, and spends at least the entire Winter deck building towards it. Final Bid has the movie structure laid out and is pretty good at making all bids pay off towards the climax, and then it is a wrap. Final Girl flawlessly builds towards the final confrontation. Trophy Dark will never go beyond the final ring. Dread has the recurring falling tower.
Weird as these things are always in TTRPGs, the ending of a one-shot is such an obviously important part of them that ends up being neglected — something that will just happen because it is a one-shot game. Apollo 47 entire mechanic is “yes and” that someone at some point will stop, I guess? Maybe? Kingdom, Microscope and so on assume at some point you will believe you just went wide or tall enough and stop. Lasers and Feelings just keeps stumbling ahead from bit to bit and then at some point have to end. Belonging Outside Belonging spends a lot of effort in setup and none in preparing to a wrap. Any Grant Howitt microgame.
Still, most of those just tend to have endings as the point to work towards, end up not being as helpful in creating that cooperative storytelling as they hoped. In those games, the success and failure of an ending — and the one-shot experience, is left to the skill and system-mastery of players and GMs3. Fiasco may have guidelines for how to flip in the 2nd part and what happens in the end, but the first scenes and the wrapping up of all last threads in the last scene are often uphill battles4; sure I know exactly what happened, but often there is a sour taste and more satisfaction from the epilogues than the ending. Thousand Year-Old Vampire has good entry endings and builds well enough for those, but any mechanic-induced ending is very underwhelming without a lot of mental gymnastics and self-narrative chops. Starcrossed can often delay the toppling of the tower, and things feel like having to go through a pre-relationship-checklist/ relationship-permission form rather than things happening when they feel right for the story5. Fall of Magic loses a lot of steam over the other side of the scroll and/or the islands take too much narrative oxygen; this detracts from the farewell to the Magus if one is not careful, as things can drag or the islands being the actual ending of the game.
“PbtA Does This”
Whenever I bring up how difficult endings are to accomplish well in this artform, this is considered a thing “solved” by PbtAs. And while there is a lot from the PbtA framework helps endings, it is not an assumed given and needs a lot of consideration6.
First of all, the universalization is pretty weird. Because PbtA is a framework, not a system; so even successful games need to do pretty unique thing with the systems they build within its framework. So, let’s go with the things more universalized: play culture trends within the framework.
Most of PbtA play is de facto “one-shots”, over 4-12 sessions. It tends towards a single arc, natural endings structure. That, indeed, puts them in a position where it is easier to hit satisfying endings — but also more important to get those endings right. That is in no way intrinsic to the framework, and needs work; in a game like Night Witches or Blackout, that is structured for that and guides you along with its systems, it is quite easy to have good endings and stop-points. But a lot of times, because it is assumed to be built-in in PbtA/one-shot games, you may end up fighting the systems with little support.
Another big trump card in play culture is “Playing To Found Out”. Likewise, it is not inherent, but it definitely worked great for that in Apocalypse World: playbooks are roles already where they need to be, a clock of threats and some obvious questions and the game runs itself like a movie. Unfortunately, “Playing To Found Out” is taken as its own answer, and a dogmatic mantra that does not reflect what it needs to work and what you are trying to find out. The systems of a PbtA game should showcase different questions, and give natural feeds, and guide those questions — when to ask, why to ask, and what to do with the answers you get to build upon next questions. This is seen as such a PbtA-thing that it is not seen as needing time to think about it and assumed anyone playing these games already has the skill and system mastery to do it unsupported. Good use of “Playing to Find Out” is essential for a satisfying ending with PbtA: picklists for playbooks and other starting choices should leave you with both obvious follow-ups and Big Questions you should work the rest of the story be to Find Out — and having that found should be the ending. If the game works well with this aspect of the play culture, I can see how one will get so many satisfying endings within framework.
However, that is not what I see pointed out as the reason for success in delivering endings in PbtA: instead, it is because endings are “hard-coded” in advancement mechanics: you have points where you “retire” characters, or substantially change them into effective retirement with a narrative landmark you buy with an advance and/or playbook change.
Unfortunately, I have to disagree with this position. If this is a tool for endings, it is a tool that says it is for endings but ends up not particularly helping — and sometimes, being yet another uphill thing to wrap around in a way to ending or dare I say, lead to Forever Games. It does communicate the value of endings — but it does not get you there. Sure, your character may retire in a game, but the game is no closer to ending and what do you do now; maybe come back with another playbook or another character. Sure, a Moment of Truth is significant, powerful and THE narrative moment of your character, but how can you setup the second one in a game and it hit significantly? And how often one has to fire Moments of Truth at the final opportunity before a game just so you don’t “waste” it?
But that’s on orthodox PbtA; how are things in the heterodoxy? We have a lot of interesting thoughts about endings. For example, Flying Circus has two pages of rules to Destinies and Retirement — what your pilot ending should be. It is an improvement over advancement-based because of how much it involves building up towards: as you get some money for savings and if you go through thresholds of personal development. While this avoids some pitfalls, it fails to materialize endings in ways that are not very PbtA: player characters advance too fast mechanically, end up with a lot of money that they either are struggling to spend/is all being poured in new equipment when they actually buy it, and are much slower at getting narrative or personal landmark developments.
The best ending experience has to be the Legacy lineage of PbtA games. These games are always ending and always working towards the ending; something one may not expect from games that benefit from longer games. Characters are defined by their progression in their personal narrative: over your narrative life, you will have four roles toward your faction; once you fulfill that, your story is over. So, every character is always working to set up the next role, which is weaving their ending closer and closer. Legacy games characters tend to be quite narrative efficient: they have less spotlight than average PbtA characters, but tend to have more memorable, character-defining scenes. Furthermore, we are also playing for the end of the current era: the building of wonders, the satisfaction of scarcity and dealing with threats. These are also natural ending points: they are satisfying ends, but characters and factions will be drastically changed by the next era. One way or another, there have been endings.
More likely than not, our games end as abruptly as this article.
We ended up not going with it because it is the worst case scenario for critical analysis: does what it says it does, no more, no less, and in a self-evident way that does not benefit from further in-depth analysis.
A popular joke about pop culture games, especially the market monopolist, is that for all the detail level they go into combat, they have rules about how combat begins but never how it ends; a similar scenario then would happen: after rolling initiative, you are forever living in 6 seconds intervals, glued to the grid. This is obviously not a serious point anyone makes, as funny as a joke it may be; I don’t find it would be more useful in any other game.
Which they do not get many opportunities to practice and really could use as much guidance as possible; after all, how many endings do they get to play?
Albeit the latest version of the game has improved the awkward first couple of scenes.
The game mastery answer in this case being “topple the tower ASAP”.
Not to mention my displeasure at considering anything in art as “solved”.