I was listening to Tales of Mystery and Imagination, the first record by The Alan Parsons Project, from 1976. It’s an odd, lovely album. Every track is a reading of Poe: “The Raven”, “The Tell-Tale Heart”, “The Cask of Amontillado”, and a long instrumental suite for “The Fall of the House of Usher” that moves through the story without a single word. That suite doesn’t tell you what happens. It tells you how it feels to be inside a house that is about to fall on you.

Somewhere in the middle of it I realised that what I love in Poe isn’t the horror itself. It’s the narrators. They’re always explaining. They’re always calm, reasonable, eager to convince you of their sanity, and they’re always wrong about something. The story is in the gap between what they tell you and what actually happened.

I wanted a game that lives in that gap. That’s where Evermore came from.

Print + Ditigal on RPG Trader

Digital on itch.io

A city that remembers for you

Evermore is a coastal city in perpetual fog: crumbling mansions with bricked-up windows, lamplighters who never speak, an asylum on the outskirts. Names repeat across generations. Nothing in it is quite stable, and that’s deliberate. The city is a mirror of the Protagonist’s mind, so it can only be as reliable as they are.

One player is the Protagonist, haunted by a sin, a loss, or a delusion. The other is the Narrator, who plays the city, its people, and the voice that leads the Protagonist down. The Narrator is allowed to lie. Whether a memory is true is not decided by either player alone.

Why tarot

A die gives you a number; a tarot card gives you a picture, and pictures are what memories are made of. The Five of Cups is a figure mourning three spilled cups, blind to the two still standing. You don’t need to know what the card “means” to know what it’s doing to your story.

What really sold me was reversal. A reversed card is the same image, upside down: the same thing, inverted. That became the rule the whole game hangs on. Every card you draw is logged and returned to the deck turned end over end. When it comes up again, whatever it brought into the story comes back, changed. The person who blessed you now accuses you. The letter you burned is sitting in your desk. I call it the Rule of Evermore, and it’s the closest I could get to the way Poe’s stories double back on themselves.

Truth as a vote

Each Act circles a single Memory Fragment: a locked attic, a childhood song in the street, a reflection that doesn’t match. At the end of the Act, both players secretly choose whether it was True, Distorted, or False, and reveal at the same time. If they agree, that’s canon. If they don’t, a card decides, and the draw leans slightly toward the Protagonist. Memory is theirs to lose.

This is the heart of the game: neither of you knows the story until you’ve told it. The Narrator can plant a lie and then watch the Protagonist believe it. The Protagonist can decide a horror never happened, and sometimes be right.

Underneath all of it there’s one card nobody looks at. Before play starts you draw it and put it face down under the Guilt Track. It’s the Buried Card: what truly happened. It sits there for three hours. At the end, at the Inhumation, you dig it up.

Guilt only goes up

The engine of the game is Guilt, and Guilt never decreases. It rises when you refuse to pay the price a scene asks, when you see proof of what you did, when you hurt the one person in the city who still knows you. At 6, 9, and 12 the Protagonist breaks: hours vanish, a suppressed memory forces its way up, or the body starts to fail. You can also choose to stop fighting it, and break on your own terms in exchange for one true answer.

This is where the game taught me something. Before release I ran the rules through a Monte Carlo simulation: 10,000 to 20,000 games for each configuration, with Protagonists who always pay the Toll, always refuse it, or mix the two. What it showed me was that in the version I had written, every Protagonist died. Every single one. The way the Acts were gated by Guilt meant you entered the last Act already past the line, so nothing you chose along the way mattered.

That’s a very Poe outcome, but it isn’t a game. I rebuilt the clock (Guilt now ticks up by one only in scenes that gained no Guilt any other way) and moved the endings. Now the choice between paying a price and refusing it actually shapes whether the Protagonist survives, survives broken, or goes into the ground. In the simulation the median Protagonist now ends at Guilt 14 if they always pay, 15 if they mix, and 17 if they always refuse: the typical Protagonist who pays or bargains survives broken; the one who always refuses goes into the ground. Doom is still on the table. It just has to be earned.

The Raven

Once per game, the Narrator can say “Quoth the Raven…” and ask a single question that cuts to the heart of it. Did you push him, or did he fall? The Protagonist has to answer in character, completely honestly, and the answer is permanent: it overrides everything, even what the cards decided. It’s the one moment in the game where the narrator can’t explain their way out.

Playing alone

I also wanted the game to work solo, because so much of Poe is one voice talking in a closed room. Alone, the deck becomes the Narrator, and you write the game as a Manuscript: a first-person confession. Two rules give it its character. When a Break makes you lose time, you’re not allowed to write it in your own voice; you write it as a found document, a physician’s note or a police report, telling you what you did. And each time you choose to break, you’re allowed to lift a corner of the Buried Card: first its suit, then its orientation, then the whole thing. You can learn the truth early, if you’re willing to pay for it.

Debts

The games I learned from are written all over this one: For the Queen by Alex Roberts, for showing me how much a card prompt can carry, and The Quiet Year by Avery Alder, for showing me that a deck can be a calendar and a fate at once. The safety tools adapt the X-Card and Lines and Veils, renamed for the city: the Veil, the Sealed Tomb, and the Shroud. And of course there’s Poe, whose stories always end in the ground.

Where it is

Evermore is out now on itch.io and rpgtrader.com, as a PDF, a print-friendly edition, and an EPUB, with printable play aids for both duet and solo play. It’s released under CC BY-SA 4.0, so if it sparks something of your own, take it and build.

If you play it, I’d love to read your epitaphs.

— Roberto

Ensemble 2e comes out on Friday, November 6, as a PDF on DriveThruRPG and a softcover on Amazon, both on the same day.

Ensemble is the group game of the Loner family: a rules-light roleplaying game for two or more players, with a Game Master or without one. It shares the Oracle, the tags and the six outcomes with Loner 4e. This post is about why I rewrote it, and about the idea that holds the second edition together.

Why a Second Edition

I rewrote Ensemble because the first edition was too wordy. It spent its pages on the philosophy of the game and of its design, and it was not a manual to play with. Loner 4e gave me the occasion to rethink, from the ground up, how the framework could be presented in its group version too.

Ensemble started from Freeform Universal by Nathan Russell, and the book still credits that debt. But the 2e is not Freeform Universal reworded: it is a long reflection on what a fiction engine is. I wrote it while the theoretical framework behind it was taking shape, the Fiction Engine Framework I have now published online, and the book is its practical application. The framework holds the theory. The book holds only what a group needs at the table.

What a Fiction Engine Is

The whole game rests on one principle: fiction precedes mechanics. A player describes what the character does, the world reacts, and the dice come in only when the outcome is uncertain and failure matters. Everything else is conversation.

That principle is also the test for every rule. A rule stays in the book only if it moves the fiction forward. If it does not, it is not there. This is what the book calls a fiction engine.

Four commitments follow from it. The game is portable: a handful of six-sided dice and something to write with. It is rules-light: one core mechanic covers almost everything. It is tag-based: characters, places and things are described in words, not numbers. And it is emergent: there is no pre-written storyline, and no one at the table, the GM included, knows how a session ends.

From the Framework to the Table

The first edition explained these ideas. The second edition turns them into procedures.

The dice pool. When the outcome is uncertain, the player asks a yes-or-no question and builds a pool: one Action Die, a Chance Die for each tag that helps, a Risk Die for each tag that hinders. Unlike Loner, the pool has no cap. Its size is limited only by honest fictional justification, so the argument about what helps and what hinders happens in the fiction, not in a modifier table. The pool puts advantage and disadvantage physically on the table.

Reading the Oracle. The six outcomes tell the group what happens and how clean it is, not how or why. The book spends a whole section on that step, because it is the hardest part for a new GM: before narrating, look at what is already in the scene, the tags, the details, what each NPC wants. The answer is almost always there. A good result does not invent something new; it tightens a thread that was already live.

Prepare instability, not outcomes. The chapter on preparing adventures says as much about what not to prepare as about what to prepare: no outcomes, no scripted NPC responses, no description of places before the characters arrive. The GM prepares a situation that cannot stay still, and the Oracle and the players do the rest. Between sessions, the work is not prep for the next one but maintenance of a living world: update the NPCs, the places and the open threads, so the next session starts where the fiction actually is.

With a GM or without. In Classic mode a GM shapes the world. In GM-less mode the group shares narrative authority, by rotating a scene host or deciding together, and the Oracle settles any interpretation the table cannot agree on. The same dice pool and the same tags work in both.

What Is in the Book

Next to the complete rules and the optional modules (Harm & Luck, Status Track, Challenge Tracks and Leverage), the book has a part for the Game Master, a guide to preparing emergent adventures, two walkthroughs of a first session (one with a GM, one without), a sample scenario, The Whispering Tower, and the Adventure Maker, seven D66 tables for settings and adventure premises. Eight diagrams show the core procedures, from the dice pool to GM-less play; the first edition had none. A rules summary, a glossary and a character sheet close the book. It has 106 pages.

If You Already Play

If you own the first edition on DriveThruRPG: the 2e arrives as a new version of the same product. On November 6 you receive it as a free update, with no new purchase.

If you play Loner 4e: your characters and the Loner Adventure Packs carry over to Ensemble with a few adjustments, and the book lists them: the dice pool replaces the Chance and Risk comparison, and the Twist Counter grows on every Yes, but. Oracle tables, setting material and character traits need no changes.

For now: mark November 6, and if you have questions about Ensemble, ask them here in the comments.

Last week I told you the date. Today it is here: Loner: Prime Time Hero is out, a solo roleplaying game set in the 1980s of network television, where every session is an episode, somebody needs help that nobody else will give, and the hero has a car, a skill, and a code they don’t break. This post is less about the release and more about what the game is and where it came from.

DriveThruRPG

Amazon

The Shows I Wanted to Play

I have always loved the 1980s shows with larger-than-life heroes: MacGyver, Knight Rider, Miami Vice. I grew up in the mid-1980s, and I watched these series when they first aired in prime time and again in reruns. I wanted a game that let me play in that atmosphere.

These shows had a shape that is easy to recognise and hard to explain. A stranger is in trouble. The police, the company, or the government will not help. The hero arrives, takes the case, gets knocked down once, improvises a way back and closes the case before the last commercial break. Next week, a new town and a new case, and nothing that came before needs to be remembered for the new episode to work.

That shape is what makes the genre a good fit for a Geared Towards Loner game. The genre is not only a look. It is a structure, and a structure is something a solo game can use.

Why an Episode Works for Solo Play

When I play alone, the Oracle decides what happens. Nobody decides when a story should turn, slow down or end. Emergent play is the point of Loner, but a session with no shape can drift: too many scenes of investigation, a climax that never arrives.

A television episode solves that without writing the story in advance. Prime Time Hero lays the classic structure over a normal Loner session: a cold open, a title sequence, three acts and an epilogue. It is not a script. The Oracle still decides what happens. What the structure gives me is an answer to the most common question in solo play: what kind of scene should I frame next?

The acts read off the Case Track, the track Loner already uses for problems too big for one scene. Act I ends when the hero knows what the case is, and that is when I open the track. Act II ends when the track has one empty box left, and I close that scene on a cliffhanger. The first scene of Act III answers it, and the act ends with the confrontation, when the track fills or when the villain’s own countdown fills first. The episode needs no extra bookkeeping: the same boxes that measure the case tell me where I am in the story. Acts only move forward, so a box I lose makes the case harder, not the episode shorter.

Where the Inspirations Show Up in the Rules

The book ends with a list of inspirational media, and almost every item on it left a mark on a rule or a table.

MacGyver is the model for every improvised gadget in the book. When the hero builds a way out from what is in the room, I say what the gadget must do and what it is made of, ask the Oracle whether it works, and write it as a tag: a microwave turned into a short-range EMP, a car stereo retuned as a radio jammer. It is Gear for one scene. The answer of the Oracle reads into the gadget itself: on a Yes, but... it works once and dies, or works and leaves something behind.

Knight Rider gave me two things. The first is the vehicle chase, where the car is a character with its own tags (Turbo-Charged, Bulletproof Glass, Handles Like a Boat) and the road gets a vote through traffic, weather and terrain. The second is Gadget Show Mode, a variant for a series built around a talking car, a black helicopter or a lab on wheels: one piece of Gear gets a full character sheet, with its own Frailty and a Goal that may not match the hero’s. The more the episode leans on the machine, the more often the machine is the problem.

The A-Team is the Team Format: the hero leads a crew of specialists, each an NPC with a sheet, who grant Advantage when their tags fit the job. Magnum P.I. is the Tropical Paradise variation, Miami Vice the darker end of the decade and the Urban Jungle. Remington Steele and Hart to Hart bring the charm and the glamorous parties, and Airwolf the stolen super-helicopter.

The films widened the frame. Raiders of the Lost Ark and Romancing the Stone for set-piece action and a hero who makes it up as he goes; Beverly Hills Cop for the outsider bluffing through the rich side of town; Die Hard for one building, one hero and every tool in reach. To Catch a Thief and Charade are older than the decade, and they are on the list for style.

The books go further back. John D. MacDonald’s Travis McGee, Elmore Leonard’s LaBrava, Peter O’Donnell’s Modesty Blaise, Leslie Charteris’s The Saint and Doc Savage, the pulp ancestor of the gadget hero. The prime-time hero did not start in the 1980s. Television gave an old figure a weekly schedule.

And the music: Mike Post and Pete Carpenter, Jan Hammer, Harold Faltermeyer, Giorgio Moroder. It is not in the rules, but it is the sound of the night drive and the final chase.

A Heightened 1980s, With Real Headlines

The world of the game is the 1980s as television showed them: green screens, car phones, designer suits, explosions a little bigger than they should be. Heroes have a code: violence is a last resort and innocent people come first. The good guys are supposed to win. The Oracle decides what it costs them.

Under the gloss, the decade itself is there. The setting chapter has a timeline of real headlines from 1980 to 1989, each one read as a source of cases: the Strategic Defense Initiative and the contractors who want a slice of its budget, the sinking of the Rainbow Warrior, Black Monday and the people who knew it was coming, the fall of the Berlin Wall and a generation of spies suddenly out of work. The six factions and the foes come from the same place: corporate raiders, arms dealers, rogue agents, a cult leader, an art forger, and a Category Four storm, because in this genre the weather can be the villain of the week.

What Is in the Book

The complete Loner rules. One Oracle, two dice, tags instead of numbers. Prime Time Hero needs no other Loner book, only four six-sided dice and something to write on. It opens with what a solo RPG is and how it differs from a gamebook, and it closes with a quick start guide for the first case.

The rules the genre adds. The episode structure on the Case Track, improvised gadgets, and vehicle chases.

Ways to run a series. Recurring characters, story arcs, a reputation that becomes a tag (Local Legend, Known to Interpol), the Team, Rival and Guest Star formats, and three difficulty variants: Pulp Hero, Gritty and Gadget Show.

The world. The heightened decade and its timeline, locations from tropical paradises to American hotspots, 6 factions and 36 foes.

36 D66 adventure tables. Cases, cold opens, episode titles, cliffhangers, climax settings, epilogues and more, so an episode can start from nothing.

A character sheet at the back of the book.

What I Hope Happens Next

If you play an episode, I would like to read about it in the comments, or wherever you found this post.

Somebody needs help. The case is yours.

Loner: Prime Time Hero comes out next Tuesday, October 27, as a PDF on DriveThruRPG and a softcover on Amazon, both on the same day.

It is the next Geared Towards Loner game: a standalone solo RPG set in the action-adventure television of the 1980s. Every session is an episode. The hero takes a case, somebody needs help that nobody else will give, and the Oracle decides what the win costs.

I have always loved the 1980s shows with larger-than-life heroes: MacGyver, Knight Rider, Miami Vice. I wanted a game that let me play in that atmosphere.

What I built around the genre is the episode itself. The Case Track that measures progress on the case also tells me which act I am in: I open it when Act I ends, its last empty box marks the Act II cliffhanger, and it fills in the Act III confrontation. Next to it sit rules for improvised gadgets, which become Gear for one scene, and for vehicle chases, where the car is a character with tags of its own. The back of the book holds the generator tables: cases, cold opens, episode titles, cliffhangers, climax settings, epilogues.

If you already play Loner: Prime Time Hero runs on the same engine, Oracle and Twist Counter included, and it is complete on its own. It needs no other Loner book, only four six-sided dice and something to write on.

If you are new to Loner: the book opens with what a solo RPG is and how it differs from a gamebook, and it closes with a quick start guide for the first case.

For now: mark October 27, and if you have questions about the game, ask them here or wherever you found this post.

The first wolf I ever really looked at was dead.

I was a teenager, wandering the villages at the foot of Monte Amiata, in Tuscany, with no plan beyond soaking up old walls and the slow weight of time on them. One August evening, in the cool after vespers, I drifted through the stalls of a village fair. I don’t remember anything that was for sale. I remember the wolf.

It was a grey wolf, shot for reasons nobody explained, then stuffed and put on display for the crowd. It stood there, stiff, in its pointless cage. I stopped and looked at it for a long time. Not at what Blake would have called its fearful symmetry, but at its calm muzzle and the way its gaze went past all of us, somewhere far off. I was furious at whoever had turned an animal like that into an object. And I was sure, against all reason, that it had gotten away anyway, that the thing in the cage was only a relic and the wolf was running somewhere I couldn’t follow.

I’ve seen a few wolves since, behind zoo fences. They were alive, but they weren’t any easier to reach.

Last winter I finally let it take over. I spent the cold months reading about wolves: novels, natural history, and above all the accounts of the biologists who study them. What struck me most in those books wasn’t the wolves. It was how little the people following them actually get to see. Months of GPS pings, camera traps, tracks in snow and scat under a hand lens, and then one sighting at dawn that lasts thirty seconds. Out of fragments like that they build a story about a family of animals, and they keep rewriting it as the pack changes behind their backs.

That’s where this game came from. It’s a small solo journaling game about watching a wolf pack. It’s finished, or as finished as these things get, so I want to write down how it came together. It’s not a pitch. These are the notes I’d want to read if someone else had made it.

itch.io

RPG Trader

The premise

You’re a wolf biologist with funding for a few seasons and a pack to follow. That’s it.

What drew me in is how little of wolf research is actually looking at wolves. It’s GPS pings, camera traps, tracks in snow, scat. Direct sightings are rare. The researcher builds a story out of fragments, and the story is often wrong. A wolf you think is hunting alone because it’s low in the pack is actually scouting for a place to leave to. The tracks you’ve been crediting to one animal belong to another with the same coat.

So the question I kept asking was: can being wrong be the point of a game? Not a failure state, not something to avoid, but the thing the game is actually about.

Things happen while you’re away

Most journaling games give you control of the story. You decide what happens, and the prompts help you decide. But a pack doesn’t wait for you. Pups are born, someone gets hurt, a young male walks off over the ridge, and you aren’t there.

The answer I landed on is the part I’m happiest with. Between seasons you draw two or three cards face down and slide them under suit markers without looking. These are sealed events: births, deaths, departures, rivals. In the first half of the next season they stay closed, and you work from what you think you know. From the eighth move on, flipping a card of a marked suit opens the event waiting under that suit.

Two details make it work.

Discovery doesn’t match the event. The suit you flip has nothing to do with what’s sealed under it. You find out about a death while looking for den sites, and about a newcomer while reading boundary marks. I tried versions where it lined up neatly, and they felt staged. Real news arrives sideways.

The event lands where you’re least sure. When a sealed event falls on “one wolf”, you skip any wolf you actually saw this season, by sighting or collar. If some wolves only have guesses or tracks this season, the event picks among those. That way it finds the thin spots in your knowledge. This rule did more than anything else to make the game feel like research and not a random table. You don’t find out that a random wolf left. You find out that the wolf you’d been confidently writing about for weeks wasn’t who you thought.

In early drafts you looked at the cards, wrote the events on slips of paper, and folded them shut. It’s a lovely ritual, but you always half-knew what was coming. Now the cards stay face down and you resolve them only when they open. The folded slips survive as a variant for people who prefer writing first.

Knowing how you know

Every entry in a wolf’s dossier gets a tag for how you know it: [C] collar or camera, [S] sighting, [T] tracks, [I] inference.

It’s bookkeeping about what you actually know, made into a rule. When a sealed event contradicts your notes, you look for the [I] and [T] entries first, cross them out, and write what you now believe. If everything you have on that wolf is [S] or [C], you’re in a harder spot. Either the wolf changed, or you identified the wrong animal. The game doesn’t tell you which.

Identifying a wolf takes a roll, and it’s harsh on purpose. Tracks need a 5+. Wolves that share a coat are easy to confuse, and all pups look alike. If you fail, you still write your best guess, but it goes down as [I]. So the dossier slowly fills with confident-looking entries that rest on very little, and that’s exactly what the sealed events find.

You cross out, never erase. Your mistakes stay on the page, because they’re part of the record.

Two documents

The game produces two things. The Pack Dossier is clinical: one page per wolf, tagged entries, designations like M-03 and F-02. The Field Journal is free. It’s where you’re frustrated, where you notice the same tawny wolf on the ridge again, where you wonder whether he’s alone by choice.

I wanted both because research is both. The dossier is where contradictions become visible. The journal is where you get to be a person about them.

Smaller decisions

No alphas. The game calls the parents the breeding pair. A wild pack is a family, and the alpha idea comes from captive wolves and has been largely abandoned by the people who study them. The pack still has an order: who eats first, who’s pushed to the edge. That order is something you watch shift, not a fixed rank you assign.

Inherited notes. At the start, a previous researcher hands you three notes about the pack. One of them is wrong, and you don’t know which. I like that you begin the study already inheriting someone else’s mistakes.

Not watching has a cost. If you spend three or more moves that turn up nothing new (going back to places you already searched, walking across known ground), you draw an extra sealed card. The wolves had a life you didn’t see, and more of it happened off the page. An earlier version measured how much of the grid you’d explored. This one measures attention, which is closer to what I meant.

Echoes. Each card’s first season goes in a log. When you meet the same card in a later season, you don’t read the prompt as new. You answer one question: what is different? The same single-file line of tracks in Autumn means something else when the pups are walking in the middle of the line, not at the back.

Calibrating loss. Half the deck is life and half is change and loss. Death leans toward the face cards, so it’s rarer than injury or dispersal. If the pack lost a wolf every season, you’d have no pack by the fourth. If nothing bad ever happened, nothing would be at stake. There’s also a floor: a pack of four or fewer won’t get smaller, unless you choose the variant that removes the floor.

The look

I wanted the book to look like a field notebook you inherited, not a game. That means:

• a small digest format;

• warm off-white paper, dark charcoal, and one dull green used sparingly;

• a cover that’s only contour lines and a title, plus four faint wolf prints in the corner;

• no wolf art anywhere.

The dossier, the map and the card log are drawn like forms you’d fill in with a pencil. If someone flips through it and thinks “this feels calm, and a little lonely”, I’ll take that.

The title

It was called Wolf Study for a long time, which sounded like a textbook chapter. Where the Wolves Are is a question the game asks for its whole length and never answers. There’s an echo of Where the Wild Things Are, turned around: you never get to go there. You stay outside, looking in.

The last prompt

At the end of a study you write a final report. The last question is: What question will you never answer?

I think it’s the most important line in the game. There’s always one. Did M-03 leave on his own, or was he pushed out? What happened to the spring pups? Why did the breeding female keep going back to that ridge?

You don’t find out. The study ends, and the wolves continue.

If you have been playing the beta on HackMD, this is the same system you already know, now laid out, edited, and illustrated the way it always deserved to be. If you are new to Loner, here is what is actually in each book.

Core Rules

The engine on its own. Tags, dice, an Oracle that answers yes, no, or something worse. This is the one you need to sit down and play tonight, no other books required.

PDF

Softcover

Companion

The why behind the how. Fiction first, emergent narrative, fictional positioning, explained and shown in play rather than just asserted. If Core Rules taught you the moves, Companion is where you learn why they work.

PDF

Softcover

World Builder’s Guide

Tools for building settings that push back: factions, NPCs, and locations with teeth, plus the Oracle and D66 tables that keep a setting as unpredictable as the story running through it.

PDF

Softcover

Character Builder’s Guide

A different way to build a Protagonist: lifepaths, shifting tags, identity that can be rewritten mid-game instead of locked in at the start.

PDF

Softcover

Omnibus

All four of the above in one volume, for anyone who would rather own the whole system as a single book than four separate ones.

PDF

Softcover

Hardcover

If you owned 3e on DriveThruRPG, your library has already been updated. The 4e files are sitting there waiting for you.

I will be around all week to answer questions and troubleshoot anything that looks off in the files. Thank you for waiting this out with me.

Go roll something.

When attempting to map the boundary between a Dungeon Crawler and a Solo Tabletop Role-Playing Game, the most common response (and unfortunately the most reductive) invariably invokes “fiction.” It is often claimed that the former is a mere exercise in grid tactical combat and spatial mapping, whereas the latter is “more story-driven.”

This distinction is fundamentally flawed.

Both experiences generate fiction. When a fighter in a dungeon crawler survives on a single hit point thanks to a desperate critical roll before slipping through a side door, that is a story. When a solo roleplayer determines the fate of a political faction through a series of oracle prompts, they are likewise experiencing a sequence of dramatic events. The presence of a narrative arc is an emergent property of play, not a mechanical differentiator.

The real divergence lies not in the product (the resulting tale), but in the mechanical and epistemic foundations of the game design. It resides in the architecture of the decision-making process, the nature of uncertainty, and the cognitive mode in which the player engages with the system.

To understand this operational rift, we must analyze four key vectors: the optimization of resource management versus the generation of meaning, the function of probability, the structural nature of the adversary, and concrete implementations across notable systems.

1. Decision-Making Architecture: Resource Optimization vs. Interpretive Response

The Dungeon Crawler: A Closed Optimization Problem

At its core, the dungeon crawler functions as a closed system of risk management.

The fundamental act of playing a dungeon crawler consists of evaluating cost-to-benefit ratios within a rigid perimeter of explicit rules. The player interfaces with an internal game economy where resources are discrete, quantifiable, and strictly scarce:

• Hit points, stamina, and spell slots;

• Torches, rations, and light sources;

• Inventory slots and encumbrance metrics;

• Action economy per round and grid positional advantages.

In this operational state, every choice is a calculation of expected value and risk mitigation. The player asks:

“Is it worth expending my last scroll of invisibility now to bypass this sentry, or should I reserve it for an unknown threat deeper down, risking a fight that might drain our remaining healing supplies?”

The decision vector is explicitly teleological: survival, spatial progress, and victory over environmental constraints. Efficiency is the primary virtue. In games like Descent or Gloomhaven, the system presents brutal trade-offs where no perfect choice exists, only tactical compromises. The player operates much like an engineer or a chess player: analyzing the board state, calculating known variables, estimating probabilities, and selecting the path with the highest statistical yield.

┌─────────────────────────────────────────────────────────┐
│              DUNGEON CRAWLER LOOP                       │
├─────────────────────────────────────────────────────────┤
│ Board/State → Resource Check → Risk Assessment          │
│     ↑                                     │             │
│     └────────────── State Change ─────────┘             │
└─────────────────────────────────────────────────────────┘

The Solo TTRPG: An Epistemic Conversation Loop

In a solo TTRPG, the focus shifts radically. The primary task is no longer solving a mathematical puzzle, but navigating a cycle of meaning creation.

The solo player does not act merely as a tactical problem solver, but as an intermediary between the character’s internal motivations and the external system’s prompts. The vital resource is not a potion counter or an inventory slot, but interpretive honesty.

The core question driving a solo TTRPG session is never “How can I win this efficiently?”, but rather:

“Given what this character cares about, their flaws, and this unexpected turn of events, what is their most authentic response?”

If a player makes a tactically disastrous decision in a dungeon crawler, it is a gameplay error. In a solo TTRPG, a tactically disastrous decision may be the absolute correct move if it aligns with the character’s ethos or current psychological state. The cognitive loop is conversational and iterative:

1. Intention: The character acts based on motives, fears, or drives.

2. Interrogation: The player queries a random generator (an oracle, a probability matrix, or a prompt list) to introduce an unscripted reality.

3. Integration: The player interprets the outcome, updates the situational fiction, and returns to step one.

While the dungeon crawler requires the player to beat the system under set constraints, the solo TTRPG asks the player to discover where the narrative trajectory leads, regardless of whether the protagonist triumphs or perishes.

┌─────────────────────────────────────────────────────────┐
│               SOLO TTRPG LOOP                           │
├─────────────────────────────────────────────────────────┤
│ Intent → Oracle/Prompt → Interpretive Integration       │
│   ↑                                         │           │
│   └──────────── Fictional Shift ────────────┘           │
└─────────────────────────────────────────────────────────┘

2. The Nature of Uncertainty: Resolution vs. Prompting

Both genres utilize dice, cards, tables, and randomizers. However, the cognitive function of these components stands in sharp contrast.

Resolution of Performance: The Tactical Die

In dungeon crawlers, randomness introduces mechanical friction. It tests performance within fully defined, pre-existing parameters.

When you roll a d20 to strike an enemy with an Armor Class of 16, the bounds of outcome are completely known prior to the roll:

• A result of 16 or higher yields a hit (inflicting numerical damage).

• A result of 15 or lower yields a miss (wasting the action or resource).

Uncertainty is confined to execution performance, not directional outcome. The roll checks whether a tactical choice succeeded. Randomness is an obstacle to be tamed through character builds, positional bonuses, gear stacking, and resource expenditure. The player constantly seeks to minimize variance in favor of determinism.

Disruption of Expectation: The Oracle

In a solo TTRPG, the randomizer (traditionally called an Oracle) does not evaluate performance; it serves to shatter the player’s confirmation bias.

When playing without a Gamemaster, the human mind naturally leans toward familiar tropes or convenient paths. The Oracle acts as an engine of alterity, simulating an external mind that responds outside the player’s direct agency.

If a player asks an Oracle: “Is the guard at the gate open to a bribe?” and receives an outcome like “No, and...”, the system is not testing a Skill Modifier. It is delivering a dramatic prompt. The player must integrate that negative modifier with an additional complication:

• No, she refuses the bribe, AND she recognizes the insignia on your cloak from a past conflict.

• No, she doesn’t take bribes, AND her captain just stepped out of the guardhouse behind her.

Here, the dice do not measure tactical success. They enforce a conceptual pivot. The uncertainty concerns what happens to the situation, not merely whether your stats pass a check. The generator strips control away from the player, forcing immediate adaptation over long-term strategic calculation.

3. Structural Opponents and Game State

The definition of the “adversary” and the representation of the world state mark another fundamental divide between these two forms of play.

The Explicit Adversarial Environment

In a dungeon crawler, the Game State is explicit and measurable. It consists of board tiles, monster stat blocks, movement ranges, line-of-sight rules, and tracking sheets.

The opponent is the mechanized environment itself. The game functions as a dynamic puzzle driven by enemy movement charts, encounter decks, and trap triggers. Victory conditions are explicit (clear the room, escape the level, kill the boss), as are defeat conditions (party death, running out of turns before a doom tracker hits zero).

If you strip all thematic fluff from a pure dungeon crawler (replacing miniatures with tokens and rooms with abstract nodes) the underlying mechanics remain entirely functional. The theme is an aesthetic skin over a mathematical model.

The Emergent Fictional Environment

In a solo TTRPG, the Game State is largely qualitative and contextual.

The adversary is rarely a static stat block on a grid; it is the friction of circumstances. A solo TTRPG session may feature no combat, no dungeons, and no physical maps. The obstacle could be an ethical dilemma, a missing piece of evidence, or a tense negotiation with an NPC whose true motives were generated only moments prior.

There is no built-in “win condition.” The session concludes not when the board is clear, but when the situation reaches a natural dramatic resolution. Failure is not a “Game Over” requiring a hard reset, but a vector for narrative velocity: a failed roll or an unfavorable oracle response escalates tension, thrusting the character into unforeseen predicaments.

4. Analysis through Notable Games

To see this theoretical divide in action, we can examine prominent titles across both sides of the spectrum, as well as hybrid designs that bridge the gap.

                  THE DESIGN SPECTRUM
                  
  [PURE TACTICAL] ◄─────────────────► [PURE INTERPRETIVE]
  
   Gloomhaven        Four Against       Ironclad /        Ironsworn
  Dungeon Degens       Darkness        Loner / Colostle   Thousand Year
                                                            Old Vampire

Pure & Hybrid Dungeon Crawlers

• Gloomhaven / Frosthaven (Isaac Childres):

  The quintessential modern dungeon crawler. Gloomhaven completely isolates tactical optimization. Players manage a hand of cards, balancing action initiative, movement ranges, and elemental infusions. The game state is fully deterministic aside from a tightly controlled modifier deck. A solo player here acts strictly as a tactical commander managing two or more characters to solve a complex positional equation.

• Four Against Darkness (Andrea Sfiligoi):

  A lightweight, pen-and-paper dungeon generator. While minimalist, Four Against Darkness is firmly a dungeon crawler. You roll on procedural tables to draw rooms, spawn monsters, and resolve combat using simple math. The goal is resource conservation (managing bandages, spells, and lantern oil) to survive the depth and exit with loot. Narrative emergence is secondary to procedural survival.

• Dungeon Degens: Hand of Doom (Locus Pocus Games):

  An example of heavy atmospheric immersion built on a dungeon crawler backbone. Despite its rich dark-fantasy lore and punk art style, the core loop is an attrition-based board game. Players navigate tracks, manage corruption, fight encounters using dice stats, and attempt to stop a sprawling doom counter. The focus remains on systemic survival within an oppressive world.

Pure Solo TTRPGs & Oracle Engines

• Ironsworn / Starforged (Shawn Tomkin):

  The gold standard of solo roleplaying systems. Ironsworn uses a Move-based framework derived from Powered by the Apocalypse mechanics combined with a robust built-in Oracle system. Action rolls do not just yield binary pass/fail results; they output Strong Hits, Weak Hits, or Misses. A Weak Hit forces a narrative compromise, while a Miss triggers a “Pay the Price” table. The engine constantly forces the player to interpret mechanical state changes into evolving fictional realities.

• Thousand Year Old Vampire (Tim Hutchings):

  A prompt-driven solo RPG focused on memory loss and tragedy. The player tracks resources (Traits, Experiences, Marks, and Resources), but the game is played by rolling dice to move through a numbered list of evocative prompts. The player writes journal entries responding to the mechanical loss of memories as their vampire ages through centuries. It is entirely an exercise in interpretive framing; there are no tactical maps or combat grids.

• Mythic Game Master Emulator (Tana Pigeon):

  Not a standalone game, but the foundational engine for solo RPG play. Mythic provides a framework for running any system solo by using a Fate Chart, a Chaos Factor, and Random Event tables. It formalizes the conversation between player and system, translating simple Yes/No questions into surprising plot twists through word-pair association tables (e.g., “Attract” + “Elements”).

Hybrid Systems: Bridging the Divide

• Five Parsecs from Home / Five Leagues from the Borderlands (Ivan Sorensen):

  These titles sit squarely on the boundary. They combine a tabletop skirmish game (hex/grid-based movement, line-of-sight, tactical cover, dice-based shooting mechanics) with a solo campaign framework featuring procedurally generated events, character downtime, persistent injuries, and story beats. The tactical battle is a pure crawler/skirmish experience, but the campaign layer operates as a solo RPG engine.

• Ker Nethalas: Into the Midnight Throne (Alex T. Albrecht):

  A dark, survival-focused solo RPG that blends heavy dungeon crawling mechanics (grid mapping, inventory degradation, tactical combat turns) with solo RPG components (domain management, event tables, and emergent narrative prompts). It demonstrates how detailed tactical sub-systems can be fused with solo roleplaying sensibilities.

5. The Cognitive Shift: A Practical Comparison

To illustrate how these structural differences manifest at the table, consider how each genre handles an identical scenario: A player character finds a locked heavy iron door at the end of a corridor.

Scenario A: The Dungeon Crawler Approach

1. Systemic Check: The player checks their sheet. They have 2 Lockpicks remaining and an Agility rating of +3.

2. Resource Math: The door requires a Lockpicking difficulty of 14. A failed check breaks 1 Lockpick. Alternatively, the Barbarian can force the door with a Strength check of 16, but doing so triggers a Noise Roll on the Wandering Monster Table.

3. Calculation: The player calculates that wasting a lockpick carries a 35% failure risk, while forcing the door has a 60% chance of spawning a wandering patrol, consuming 2-4 hit points across the party.

4. Action: The player chooses the lockpick to conserve party HP for the upcoming room on the map grid.

Scenario B: The Solo TTRPG Approach

1. Fictional Context: The rogue reaches the iron door, fleeing from a city guard after an assassination attempt.

2. Oracle Interrogation: The player rolls on an Oracle to ask: “Is the door locked?” Outcome: “Yes, AND...”

3. Prompt Interpretation: The player interprets the “AND”: The door is locked, AND someone is actively sliding a heavy bar into place from the inside.

4. Character Response: The rogue doesn’t pull out lockpicks: there is no time, and someone is waiting on the other side. The player considers the character’s core drive (“Never get trapped in small spaces again after the prison break”). Instead of picking the lock or quietly retreating, the rogue screams a password they overheard three chapters ago, pretending to be an agent of the cult inside.

5. System Roll: The player rolls a Move (e.g., Compel or Face Danger) to see if the bluff succeeds, completely shifting the narrative vector from a stealth break-in to a high-stakes social deception scene.

6. The Gray Zone: Mechanical Hybridity and System Bleed

While the theoretical boundary between these two formats is clear, design in practice is rarely sterile. A significant portion of modern solo gaming exists in a hybrid gray zone, where the procedural grid of the dungeon crawler intersects with the open-ended interpretive frameworks of solo TTRPGs.

                    THE HYBRID SPECTRUM
                    
   Tactical Engine                       Interpretive Layer
 [Spatial/Resource] ◄─── Bleed Zone ───► [Oracle/Contextual]
         │                                       │
         ▼                                       ▼
   Skirmish Rules                         Fictional Anchors
 (Five Parsecs/Leagues)                 (Ker Nethalas/Notequest)

This intersection occurs primarily in two ways: tactical engines adopting interpretive layers, or narrative engines anchoring themselves to rigid spatial procedures.

Skirmish-RPG Hybrids

Titles like Five Parsecs from Home and Five Leagues from the Borderlands (by Ivan Sorensen) explicitly bridge this divide.

During the tactical phase, the game operates entirely as a miniature skirmish dungeon crawler: hex/grid-based movement, line-of-sight checks, cover modifiers, and dice-based hit resolution. The mental mode required is pure tactical optimization.

However, the campaign layer surrounding these combat encounters operates through procedural solo RPG engines. Events between missions, character injuries, trade decisions, and persistent rivalries are generated via tables that require interpretation. The skirmish battle ceases to be a self-contained board game and becomes a high-stakes resolution engine for an ongoing narrative.

The Procedural Dungeon-Journaler

Conversely, systems like Ker Nethalas: Into the Midnight Throne or minimalist dungeon-crawling journalers push from the opposite direction. They use heavy procedural generation (drawing rooms on graph paper, managing torch burn rates, rolling on item tables) but leave open space for character motives and dramatic prompts.

The gray zone is defined by a shift in player posture:

• System-driven emergence: You roll on a table, spawn a monster in room 4B, and resolve the fight using grid tactical rules (Dungeon Crawler stance).

• Player-driven projection: During or immediately after that fight, you use an Oracle to ask why the creature was guarding a shattered shrine, integrating the tactical result into a broader campaign arc (Solo TTRPG stance).

This hybridity demonstrates that the distinction is not a matter of components (miniatures, graph paper, or dice) but of cognitive gear-shifting. A player can run a pure dungeon crawl, pause the game after a narrow escape, switch mindsets, and use a solo RPG oracle to determine the emotional and political fallout of that battle.

Conclusion

Distinguishing between a dungeon crawler and a solo TTRPG by pointing to “the story” misses the mechanics of how games function. Story is an emergent byproduct present in both modes.

The distinction lies in the nature of the player’s engagement:

• The Dungeon Crawler engages the analytical mind. It is a game of systemic efficiency, resource allocation, and spatial problem-solving under controlled uncertainty.

• The Solo TTRPG engages the generative, adaptive mind. It is an exercise in narrative friction, where randomizers act as an external consciousness to strip away player control, forcing authentic reactions to unexpected events.

Understanding this division allows designers to build tighter systems and helps players select the exact type of friction they want at their table. Whether that means outsmarting an intricate board of tactical variables or stepping into the dark with an oracle to see what awaits.

I have been sitting on this news for a while, and it is finally time to say it plainly: Loner 4e launches in two weeks, on September 25.

All four core books land the same day: Core Rules, Companion, World Builder’s Guide, and Character Builder’s Guide, alongside the Omnibus that collects all four into one volume. No staggered rollout, no waiting on one book while the others catch up. Everything, at once.

This edition has been shaped by two years of feedback from people who actually played the beta, not just read it. The core loop is tighter, the Oracle and Twist mechanics have been reworked, and the system finally reads the way it plays: light enough to disappear behind the story.

If you’re buying on Amazon

The third edition print books stay listed and purchasable through September 23. After that, they come down to make room for 4e. If you have been meaning to grab a 3e copy for any reason, a gift, a reference, whatever, that is your window.

If you already own it on DriveThruRPG

You do not need to do anything. DriveThruRPG updates existing purchases in place, so your 3e files become 4e files automatically the moment the new edition goes live. No repurchase, no separate download link to track down.

What’s next

Over the next two weeks I will be sharing layout previews and a closer look at what actually changed under the hood. For now: mark September 25, and if you have questions about the transition, ask them here or wherever you found this.

I have a soft spot for small games that do a lot with very little.

Not “small” as in shallow. I mean small in the sense of sharp: a few rules, a clear table procedure, and enough trust in the players that the game does not feel the need to simulate every hinge, bruise, bullet, and awkward pause.

Over the years, I have found myself returning again and again to a particular family of games: games where the fiction is not decoration over a machine, but the actual fuel of play. You say what your character does. The table looks at the situation. If the outcome is uncertain, the rules step in. Then the rules hand you back a changed situation.

That sounds obvious, maybe. Isn’t that what all role-playing games do?

Well, yes and no.

Many RPGs use the fiction as a prompt for entering a mechanical sub-system. You describe swinging a sword, sneaking past guards, or pleading with a queen, and suddenly you are inside a different machine: combat rounds, initiative ladders, action economies, damage types, skill DCs, social influence procedures, resource spends. These tools can be fun. I like many games built that way. But the more machinery you invoke, the more play can drift away from the living situation at the table and toward operating the system.

The games I want to talk about here pull in the opposite direction.

They are interested in rules that stay close to the imagined moment. Rules that help answer: What happens now? What changes? What does this cost? What new problem or opportunity appears?

That is what I have elsewhere been calling a Fiction Engine: a way of playing where the rules exist to turn fictional intent into fictional change, with as little dead weight as possible. The framework I am working from puts it more strictly: procedures should either create new fiction or support the conditions that make meaningful fiction possible; outcomes should move, complicate, or clarify the situation rather than pause it inside unrelated bookkeeping.

This article is not a taxonomy for experts. I am not trying to hand out official certificates. I want to compare six games that orbit this idea in different ways:

The Pool, Fudge, FU, Loner, Ensemble, and PDQ.

I should be transparent from the beginning: Loner and Ensemble are my own games. That means I am not a neutral observer there. But it also means I can speak openly about what I was trying to do with them, what I borrowed, what I changed, and where I think they sit among their ancestors and cousins.

The Pool: risk, authorship, and the joy of giving the player the mic

The Pool, by James V. West, is one of those games that looks almost impossibly small until you realize how much design is hiding in the space it leaves open.

Its basic loop is direct. A player states what their character is trying to do. If there is conflict or uncertainty, dice are rolled. If the roll succeeds, the player has a choice: take a die for future use, or claim a Monologue of Victory, narrating the outcome themselves within the limits of established continuity. If the roll fails, the GM narrates what happens.

That single choice is the whole game’s magic trick.

Do I strengthen my future position, or do I seize the story now?

The Pool is not “fiction-first” because it has no mechanics. It absolutely has mechanics: the dice pool, gambling dice, risk, gain, loss. But those mechanics are not there to model physics, combat tempo, or task granularity. They are there to create a moment of pressure around narration.

The Monologue of Victory is especially important. A successful roll can hand temporary narrative authority to the player. They can describe not merely how their character performs an action, but what follows from it: consequences, details, sometimes even developments beyond the immediate body of the character, as long as they respect continuity and the other characters.

That makes The Pool a strong example of a fiction-generating game. The roll does not produce “three successes,” “eight damage,” or “a partial tactical state.” It produces permission to alter the story.

But The Pool is not invisible. The Pool of dice is a very visible meta-resource. You are not pretending that the dice are “really” fatigue, morale, positioning, or skill. They are a risk economy. You gamble. You lose dice. You build dice. You decide when to cash in. So if someone’s ideal is a game where every procedure feels diegetic, The Pool will not quite fit.

For me, that is not a flaw. The Pool is not trying to hide the table. It is saying: we are all authors here, and sometimes the rules decide who gets the pen.

So I would call The Pool a minimalist fiction engine of risk and authorship. It is not primarily about simulating the world. It is about deciding who gets to say what happens next, and what they must risk to earn that privilege.

Fudge: a toolkit that can become a fiction engine

Fudge is a different animal.

Where The Pool is a tight little procedure, Fudge is a toolkit. It is famous for describing characters with words rather than numbers: Superb, Great, Good, Fair, Mediocre, Poor, Terrible. A character can be a Great Swordsman rather than a 14th-level fighter. The text presents itself as rules-light, customizable, and focused more on role-playing than on endless tables and rolls.

That alone makes it attractive if you care about fiction-facing play. Words like “Great” and “Fair” are immediately readable. They tell the table something in ordinary language. You do not need to translate a stat block before you understand the character.

Fudge also has a healthy instinct about when not to roll. It explicitly says there is no need to roll when an action is automatic, and no need to roll when an action has no chance. Dice belong in the middle ground, where the outcome is genuinely uncertain. It also encourages the GM to keep die rolling to a minimum and not call for rolls for mundane actions.

That is very close to the spirit I am interested in. The fiction leads. The dice enter only when needed.

But Fudge can go in many directions.

It has attributes, skills, gifts, faults, supernormal powers, scale, combat options, wounds, healing, different levels of character creation detail, and optional sub-systems. None of that is bad. In fact, it is part of why Fudge has been so influential. You can tune it toward cinematic action, gritty realism, fantasy adventure, supers, or almost anything else.

But that flexibility means Fudge is not automatically a fiction engine in the stricter sense. It depends how you configure it.

A light Fudge game where players describe intent, the GM sets a difficulty from the fictional situation, the dice answer uncertainty, and the result immediately changes the scene? That is very much in the zone.

A heavily configured Fudge game with many skill lists, tactical modifiers, wound calculations, weapon factors, detailed combat pacing, and multiple optional procedures? That can still be fun, but it begins to feel more like a traditional universal RPG engine.

So Fudge sits at an interesting point in the lineage. It is fiction-facing, but not necessarily fiction-generating by default. Its descriptive trait ladder and roll-only-when-needed attitude are deeply compatible with the kind of play I am describing. But it remains a toolkit. The group has to choose how much machinery to bring to the table.

If The Pool says “risk narration,” Fudge says “describe the world in ordinary language, then build the amount of system you need.”

That is powerful, but it is also less opinionated.

FU: the clean grammar of “Yes, and / No, but”

FU, or Freeform Universal, feels like someone took a lot of the instincts behind fiction-first play and expressed them in the clearest possible grammar.

The heart of FU is the closed question. When something uncertain happens, you ask a yes/no question: “Do I leap the chasm?” “Do I punch that jerk in the nose?” “Does the tavern wench fall for my charm?” Then you roll, and the result gives one of six answers: Yes, and... / Yes / Yes, but... / No, but... / No / No, and...

That table is small, but it changes everything.

A binary pass/fail roll often answers only one thing: success or failure. FU’s result table answers success or failure and momentum. Did you get more than you wanted? Did you succeed at a cost? Did you fail but find another way? Did things get worse?

This is why FU feels so close to my idea of a fiction engine. The result already speaks in story language.

FU goes further by saying that qualifiers like “and” and “but” can add Conditions to characters or Details to scenes. A character becomes angry, confused, tired, injured. Curtains catch fire. Windows break. Animals run off. Machinery stalls. These are not abstract points floating above play. They are changes to the imagined situation, and they can matter later.

Modifiers also come from the fiction. Descriptors, gear, conditions, and scene details can make an action easier or harder, adding dice and letting the player keep the best or worst result. Again, the rule asks: what is true right now that helps or hinders?

One of FU’s most elegant decisions is refusing to create separate procedures for every kind of conflict. The text says there are no opposed rolls, contested actions, damage rolls, or rolling to hit as distinct systems. The same “beat the odds” roll handles driving through a crowded mall, arm wrestling a giant, shrugging off damage, sword fights, political debates, wars, or gun battles.

This does not mean every moment feels the same. It means the scale of the question changes.

“Do I hit him?” is a close-up question.

“Do I win the duel?” is a wider question.

“Do my forces storm the castle?” is wider still.

The same core procedure can zoom in and out because it is not pretending to be a physics engine. It is a consequence engine.

FU is probably the cleanest ancestor of both Loner and Ensemble. It taught me that a resolution table can be more than a probability device. It can be a sentence starter for the next piece of fiction.

If I had to summarize FU in one line, I would say: FU is the grammar of fictional consequence made playable.

Loner: FU’s spirit turned toward solo play

Loner began with a question that kept bothering me: what happens when you take that FU-like grammar and need it to work without a GM?

In a group game, when the dice give you “No, but...” or “Yes, and...”, other people help interpret. A GM can reincorporate details, make the world push back, introduce pressure, and keep the story moving.

In solo play, you need procedures that do some of that work without taking over your imagination.

That is where Loner lives.

Loner is a minimalist solo RPG built around one protagonist, descriptive tags, and an Oracle. You ask closed questions, interpret the Oracle’s answers, and let unexpected developments pull the story in directions you did not plan. The rules describe the game as improvisational, emergent, and open-ended, with no pre-written story.

The kinship with FU is obvious: the six outcomes are still there. But the dice work differently. Loner uses a Chance Die and a Risk Die. Compare them: Chance higher means Yes, Risk higher means No, equal means Yes, but and advances the Twist Counter. If both dice are low, you add “but”; if both are high, you add “and.”

I like this because it makes the roll feel like two forces meeting: possibility and resistance.

But the real core of Loner is not just the Oracle. It is the tag system.

Loner says tags describe what is true in the fiction. They are not numbers or passive bonuses. They provide context, and only matter when they are relevant to the specific moment. The same tag can help in one scene and hinder in another. A frailty like being crippled might impose disadvantage while climbing a wall, but might earn sympathy when asking a guard for help.

That principle matters a lot to me. I do not want tags to become disguised stats. I want them to be reminders of fictional truth.

Loner also adds tools that solo play needs: scene framing, scene closing, scene transitions, dead ends, relationships, the Twist Counter, Meanwhile scenes, and the Living World. These are not meant to replace imagination. They are meant to stop the solo player from getting stuck in either total control or total randomness.

The Twist Counter, for example, tracks accumulating pressure. A twist is not just a bad roll; it is something larger changing — the situation, the goal, or who is in play. The Dead Ends procedure tells you what to do when a lead collapses: stop interrogating the closed path and shift scale. The Living World procedure makes sure NPCs, factions, locations, and unresolved events remember what happened between adventures.

These are all procedural, yes. Loner is less “purely minimal” than FU. But the procedures exist because solo play needs scaffolding. Without a GM, you need the game to help you ask better questions, stop over-questioning, move scenes along, and let the world act offscreen.

That is why I think of Loner as a solo fiction engine. It is not just a task resolver. It is a small machine for keeping fiction alive when there is only one person at the table.

Ensemble: bringing the engine back to the table

Ensemble is, in many ways, my attempt to bring Loner’s procedures back into group play without losing FU’s conversational clarity.

The text is not shy about its intention. It says directly that Ensemble is built on the principle that “fiction precedes mechanics,” and calls itself a “fiction engine”: every rule exists to move the fiction forward, and if a rule is not doing that, it is not here.

That is as explicit as I can make it.

Ensemble keeps the same six-result structure: Yes, and... / Yes / Yes, but... / No, but... / No / No, and... But unlike Loner, it uses a dice pool: an Action Die plus Chance Dice and Risk Dice. Advantages and disadvantages cancel one-for-one. If Chance Dice remain, roll and keep the highest. If Risk Dice remain, roll and keep the lowest.

This makes advantage and disadvantage physically present on the table. You can see the situation gathering weight.

The danger, of course, is that dice pools can tempt players into shopping for bonuses. So Ensemble insists that tags are qualitative, not quantitative. A tag only matters if it is relevant to the current fictional moment. “Bow Mastery” does not give +2 forever. It matters when the character is actually doing something where bow mastery changes what is possible.

This is also why Ensemble does not use a hard cap like Loner. In Loner, I wanted solo play to stay fast and avoid tag-counting. In Ensemble, I wanted the group to be able to recognize multiple distinct fictional factors when they genuinely matter. But the limit is honesty. If three tags are just three ways of saying the same thing, they should not all become dice.

Ensemble also expands the social side. Relationship tags are earned through play, not assigned as static background. If a bond is tested, strained, or deepened, it can become a tag: Trusted Ally, Sworn Enemy, Uneasy Truce, Owes Me a Debt. These tags then matter when they bear directly on a situation.

That is important because a fiction engine should not only care about competence. It should care about the changing web of commitments, debts, fears, loyalties, and reputations that make campaign play meaningful.

Like Loner, Ensemble uses scene framing and scene closing. Before a scene begins, establish where the characters are, who is present or likely to be present, and what the characters intend to do. A scene ends when the goal is resolved, blocked, or abandoned.

That sounds simple, but it is one of the most useful play procedures I know. It keeps the table from drifting. It also prevents the common habit of rolling and rolling after the scene has already done its job.

Ensemble also has Classic and Masterless modes. In Classic play, a GM manages the world and NPCs. In Masterless play, authority is shared and the Oracle fills gaps. That dual mode matters because I do not think “fiction engine” means “GMless” or “GMful.” It means the rules are accountable to the fiction, whatever authority structure the group uses.

If FU is the seed, and Loner is the seed adapted for solo play, Ensemble is my attempt to grow it into a table-facing campaign game.

PDQ: descriptive, clever, and a little more mechanical under the hood

PDQ, the Prose Descriptive Qualities System by Chad Underkoffler, is fascinating because it shares many instincts with these games but expresses them through a more traditional resolution skeleton.

Characters are made of Qualities: prose, descriptive, do-it-yourself traits that can stand in for attributes, skills, advantages, flaws, powers, or incapacities. That is immediately attractive. A Quality is not just a narrow skill. It is a meaningful phrase with a “penumbra,” a shadow of related uses.

The text gives examples. “Gunplay” is not just shooting; it can include cleaning and repairing guns, knowing where gun dealers and shooting ranges are, understanding firearm history, and so on. “Teacher” might be too broad in some games, so it could become “College History Prof,” “High School Chemistry Teacher,” or “Grade School Teacher,” each bringing a different shadow of implied knowledge.

This is excellent design. It asks the table to understand what a character trait means, not merely what bonus it grants.

PDQ also supports automatic success in simple situations. The GM compares the task’s Difficulty Rank to the character’s Quality Rank. If the character’s rank is higher, the task succeeds without a roll; if the difficulty is equal or higher, it becomes a complicated situation. Again, this keeps dice out of places where they are not needed.

Where PDQ diverges from FU, Loner, and Ensemble is conflict.

PDQ has a more elaborate conflict procedure: initiative, attacker and defender, opposed 2d6 + Quality rolls, damage based on the margin of success, Failure Ranks or Damage Ranks, and downshifting Qualities as a character loses capability.

Now, this is not a bad system. In fact, it is clever. PDQ defines damage broadly: physical, mental, emotional, or social damage is a loss of capability. A debate, seduction attempt, chess match, race, or sword fight can all wear down a character’s ability to continue. That is much more flexible than hit points.

Even better, players choose where to apply damage by lowering their own Qualities. This creates strange and sometimes funny abstraction. If you take a punch in a fight, maybe your “Accountant” Quality drops because you are rattled, bruised, distracted, or simply less able to function as your best self. The text openly acknowledges that this is not one-to-one simulation; it is conflict abstraction.

I admire that. PDQ understands that all conflict is pressure on the character as a whole, not just meat being removed from a body.

But compared to FU-style games, PDQ’s roll result does not directly speak in fictional momentum. It tells you who wins the exchange and how many ranks are lost. The fiction still matters, and the Qualities are beautifully descriptive, but the conflict procedure introduces more bookkeeping and a more traditional rhythm: act, oppose, compare, damage, recover.

So I would not put PDQ in the same bucket as FU, Loner, or Ensemble. I would call it a descriptive, fiction-facing engine with a more mechanical conflict core.

It is closer to Fudge than to FU, but with a sharper idea of broad character-defining phrases.

So where do they land?

If I were arranging these games by how directly they turn fictional intent into fictional change, I would place them like this:

FU is the cleanest expression of the core grammar. Ask a question. Roll. Get a story-shaped answer. Add a condition or detail. Move on.

Loner takes that spirit and builds solo procedures around it: Oracle, tags, twists, scene structure, dead ends, and a living world.

Ensemble takes that same foundation into group play, making the principle explicit and adding campaign-facing tools for relationships, scenes, long-term arcs, and shared or GM-led authority.

The Pool is different but powerful: it is less about graded consequences and more about risk and narration rights. Its Monologue of Victory remains one of the boldest small procedures in RPG design.

PDQ is strongly descriptive and flexible, with Qualities that behave like rich fictional handles, but its conflict system is more attritional and procedural.

Fudge is the broadest toolkit. It can absolutely be played in this family, but it can also become more traditional depending on how much of the optional machinery you use.

That ranking is not a quality ranking. It is not “best to worst.” It is about design posture.

If I want the most elegant little question-answer fiction machine, I reach for FU.

If I want solo emergent play, I use Loner.

If I want that approach at a table, especially for a campaign, I use Ensemble.

If I want player authorship as a charged reward, I look at The Pool.

If I want descriptive characters with a little more crunch and broad applicability, PDQ is still a joy.

If I want a universal toolkit I can tune myself, Fudge remains one of the classics.

What I learned from comparing them

The main lesson, for me, is that “fiction-first” is not one thing.

There is the authorship route, where the key question is: who gets to say what happens? The Pool lives here.

There is the descriptive trait route, where the key question is: what does this character or situation mean in ordinary language? Fudge and PDQ live here.

There is the consequence grammar route, where the key question is: how does the answer change the situation? FU lives here, and Loner and Ensemble grow from it.

The games overlap, of course. But each one reveals a different design value.

The Pool reminds me that risk can be about voice, not just success.

Fudge reminds me that ordinary language can carry a whole game if you trust it.

FU reminds me that a result table can be a narrative engine.

PDQ reminds me that broad descriptors can be mechanically robust without becoming lists of micro-skills.

Loner reminds me that solo play needs procedures not to replace imagination, but to surprise it.

Ensemble is my attempt to put all of that back into a social space: to let a group play with fictional clarity, light rules, consequence-rich results, and a world that remembers.

I do not think every RPG should work this way. Sometimes I want tactical maps. Sometimes I want gear lists. Sometimes I want combat procedures with teeth and weight and positioning puzzles.

But when I want a game where the imagined situation stays bright and close, where every roll immediately asks “what changes?”, these are the games I keep coming back to.

And if there is a single practical test I use now, it is this:

After a rule is used, can everyone at the table clearly say what is different in the fiction?

If yes, the rule is doing its job.

If no, maybe the rule is only making noise.

That, more than any label, is what I care about.

With this issue I start a new campaign using SELLSWORDS, my 24XX/5e SRD mash-up to play high fantasy withouth the burden of a crunchier ruleset.

Player Character: Ink “Corvain,” Weary Barbarian

Origin: Tiefling Company: Lantern and Blade

Demeanor: Weary — the kind of tired that isn’t about sleep. Ink’s seen enough fights that nothing gets much reaction anymore, but the rage still comes when it’s needed.

Skills

• Melee Combat — d8

• Intimidation — d8

Tiefling Traits

• Ember-blood — a scrap of inherited infernal magic. Ink’s eyes flare red-orange when their temper spikes, and they can will a small flame to life in their palm, no bigger than a candle, at no cost, when there’s no real risk in it.

• Fireproof hide — a stubborn resistance to fire. Burns that would cripple someone else leave Ink singed, not scarred.

Signature Item

The Widow’s Reckoning — a notched, blackened greataxe that hums faintly when Ink’s fury rises. An extension of that fury, as the rules put it: not elegant, just honest.

Gear (₲2 starting gold)

• The Widow’s Reckoning (greataxe, bulky)

• Leather armor (breaks once for defense)

• Traveling gear (rope, torches, rations — untracked)

• ₲2 remaining

Spark: Protect + Omen

The job that finds Ink isn’t posted on any board — it comes through Lantern and Blade’s usual channel, a name they trust, urgent enough to skip the paperwork.

A hedge-priest in the village of Thornhollow has been having the same dream for six nights running: a black comet splitting the sky over the village chapel, and beneath it, the chapel doors standing open with nothing but dark inside. She’s convinced it’s not a dream but a warning, and she’s paid Lantern and Blade to have someone stand watch over the chapel through the seventh night, when she’s sure the omen comes due.

Half the company thinks she’s touched. Ink took the job anyway — coin’s coin, and a night’s watch is easy work.

(Situation rolled Trouble — something’s already wrong.)

Ink arrives at Thornhollow an hour before dusk, greataxe slung over one shoulder, and finds the chapel doors already standing open. Not forced. Not broken. Just open, like something walked out through them and didn’t bother closing up after itself.

The Thornhollow Chronicle

S1 – Thornhollow Chapel, dusk

[PC:Ink|Barbarian|Melee Combat d8|Intimidation d8]
[L:Thornhollow Chapel|doors open|undisturbed]

@ Ink inspects the chapel
=> Empty. Doors open, unforced. No struggle. [Thread:The Priest's Warning|Open]

Ink steps through the doorway with his axe already loosened in its harness, moving with the wary quiet of someone who’s learned not to trust silence. The chapel greets him empty, its doors thrown wide as if whoever left had no reason to hurry. No overturned pews, no blood, no sign of struggle. And yet that perfect stillness settles over him like a cold hand on the back of his neck.

@ Ink inspects the altar
=> Wax melted upward before curling down. Fresh symbol scratched into the stone: a circle split by a jagged line.

He crosses to the altar, and the wax gives the first wrong note — it hasn’t dripped downward as it should, but climbed upward, curling back on itself before falling. A small detail, but wrong enough to stiffen his shoulders. Beneath his calloused fingers he finds a symbol freshly scratched into the stone: a circle split by a jagged line, like a wound someone wanted left open on purpose.

@ Ink follows the traces leading outside
=> Footprints cross the yard, end abruptly at the treeline. Flattened grass in a wide circle.

Outside, footprints cross the yard in a straight, unhurried line — then stop dead at the treeline, as if the person who made them simply ceased to exist. Around the spot, the grass lies flattened in a wide, deliberate circle, too clean to be the work of wind or animals.

? Something interesting's visible or heard-able around there?
d: d8=3 -> Yes, but...
=> A complication is coming, not yet revealed.

? What kind of complication?
d: d6=5 -> Misleading
=> A "fallen" black stone, tooled rather than natural — made to look like something it isn't.

At the center of the circle, something glints — a black stone half-sunk in the flattened grass. Ink crouches to look closer, and the shape stops him: this thing didn’t fall from anywhere. Its edges are too even, its surface too deliberately dulled, as if something worked hard to make a crafted object look like a lump of nothing.

@ Ink takes the stone and returns to the chapel
=> The stone's warmth doesn't fade on the walk back.

He pockets it anyway — better in his hand than left for something else to claim — and starts back toward the chapel. The stone should cool against his palm with every step. It doesn’t. If anything, it holds its warmth like a coal banked in ash, patient and unbothered by the night air.

@ Ink prepares for the night, hidden
d: Situation d6=3 -> Signs
=> The stone's warmth turns to a slow, deliberate pulse. One altar candle gutters out on its own.

Ink tucks himself into shadow near the altar, axe across his knees, and waits. The stone’s warmth changes rhythm — no longer steady, but pulsing, slow and deliberate, like something breathing in its sleep. Across the chapel, one candle on the altar guts out with no draft to blame.

@ Ink waits until anything shows up
d: Situation d4=3 -> Signs
=> The pulse cuts off cold. The stillness changes shape.

The pulse stops. Not fades — stops, all at once, like a held breath finally released. The stillness in the chapel shifts, thickens, takes on a shape Ink can’t name but can absolutely feel pressing against the edge of the dark.

? Something happens that night?
d: d10=6 -> Yes, and...
=> A hooded figure appears in the doorway, unnaturally still. [N:Hooded Figure|wants the stone|unknown obstacle]

A figure fills the doorway. Hooded, motionless, wrong in a way Ink’s eyes can’t quite parse — no sway of breath, no shift of weight, nothing that marks it as alive.

N (Hooded Figure): “You’re not who left that for me to find.”

PC (Ink): “Ah! You are interested in this?” (points at the stone) “Well, then...”

@ Ink tries to shatter the stone with his axe
d: Melee Combat d8=6 -> Success
=> The stone shatters. Black smoke recoils toward the doorway. The figure staggers, hood slips — its face unfinished beneath.

Ink doesn’t wait for an answer. The axe comes down and the stone shatters under the blow, black smoke boiling out and recoiling toward the doorway like something yanked on a leash. The figure staggers, its hood slipping back — and beneath it, no face. Just the suggestion of one, unfinished, as though whoever made this thing ran out of time or reason to bother.

N (Hooded Figure): “That wasn’t yours to break.”

@ Ink attacks the hooded figure
d: Melee Combat d8=4 -> Setback
=> The axe passes through like fog. Cold lashes back, numbs Ink's sword arm. [PC:Ink|hurt: numbed arm, rolls hindered d4 for grip-based actions]

He presses the attack, but the axe meets nothing — passes through the figure like fog through fingers. Cold lashes back up the blade and into his arm, numbing it to the shoulder.

@ Ink uses Ember-blood
d: d6=6 -> Success
=> Flame catches along blade and hand, drives the cold out. The figure recoils further, afraid for the first time.

Gritting his teeth against the numbness, Ink calls on the ember-blood running in his veins. Flame catches along the axe head and his own knuckles, burning the cold out of his arm in one sharp wave. For the first time, the figure recoils — not staggering this time, but flinching, genuinely afraid.

N (Hooded Figure): “Fire-blood. That’s not — that wasn’t in the reckoning.”

@ Ink uses Intimidation on the figure
d: Intimidation d8=1 -> Disaster
=> The figure snaps instead of yielding — full cold blast to the chest. [PC:Ink|badly hurt: needs rest, physical rolls hindered d4]
=> The figure vanishes with the last of the smoke.

Ink pushes his advantage, snarling a threat meant to break the thing’s nerve entirely. Instead, it snaps — lashing out with a full blast of cold straight to his chest that drops him half to his knees, breath torn away.

N (Hooded Figure): “The reckoning still comes. This only delays it.”

The last of the black smoke curls away, and the figure goes with it, leaving Ink alone in a chapel gone cold and empty again.

@ Ink goes in search of help
d: Situation d4=2 -> Trouble
=> The priest's cottage stands open too. No answer inside. [N:Priest|missing]

Still aching, Ink drags himself out to find the priest — someone who might know what he just crossed axes with. Her cottage stands open the same way the chapel did: unforced, undisturbed, and utterly empty. No answer comes to his call.

@ Ink eats rations and searches for a pallet for the night
=> Rests at the inn. [PC:Ink|no longer hurt]

With nothing left to chase in the dark, Ink retreats to the village inn, forces down cold rations, and collapses onto a pallet. Sleep comes hard-won but complete — by morning, the numbness and the ache are gone.

S2 – Thornhollow village, morning

@ Ink searches for villagers
d: Situation d6=1 -> Trouble
=> Villagers avoid the priest's cottage on sight. [N:Old Man|reluctant]

Daylight doesn’t make the village any friendlier. Every villager Ink approaches goes quiet or finds somewhere else to be the moment the priest’s cottage comes up in conversation — all except one old man, who lingers just long enough to be caught.

PC (Ink): “Speak, old man. Where is she now?”

@ Ink presses the old man for answers
d: Intimidation d8=6 -> Success
=> Gone three nights. Half the village thinks she wandered off; the other half won't say what they think. [Thread:What Took the Priest|Open]

A hard look and a harder tone loosen the old man’s tongue. The priest has been gone three nights, he admits. Half the village has settled on “wandered off” as the comfortable answer. The other half won’t say what they actually believe — and their silence tells Ink more than words would.

@ Ink investigates the priestess' house
d: Situation d6=5 -> Calm
=> Finds a week of notes on the dream, a sketch matching the altar symbol, and the line "It isn't a warning. It's a summons." A pouch holds three more dull, unshaped stones. [E:Reckoning Clues 2/?]

Inside the cottage, Ink finds what the priest left behind: a week’s worth of notes chronicling some recurring dream, a sketch that matches the symbol carved into the altar stroke for stroke, and one line underlined twice — “It isn’t a warning. It’s a summons.” Tucked in a pouch nearby sit three more stones, dull and unshaped, twins to the one he shattered the night before.

@ Ink takes the notes and the stones, leaves the village to return to his Company
=> The road to Lantern and Blade's hall is two days on foot. [Thread:The Priest's Warning|Open]
[Thread:What Took the Priest|Open]

Ink pockets the notes and the stones and turns his back on Thornhollow. Two days of road stretch between him and the hall of Lantern and Blade, his Company — two days to sort through what he’s seen, and what he still doesn’t understand.

? Meanwhile, do events set in motion?
d: d10=7 -> Yes, and...
=> Something moves in Thornhollow while Ink is on the road. Unrevealed.

Behind him, unseen, something in Thornhollow begins to stir the moment the village gates fall out of sight.

Introduction

The Warlock of Firetop Mountain, the book that launched the Fighting Fantasy line in 1982, is remembered today as a genre-founding text: the first mass-market gamebook to fuse the branching structure of Choose Your Own Adventure with the resolution mechanics of tabletop role-playing. It is also, on replay, a startlingly hostile piece of design. Its dungeon is not a simple corridor of binary choices but a genuinely non-linear space, full of loops, dead ends, and a small number of deliberate chokepoints that exist purely to keep the book’s branching structure printable. Its combat system, built on the Skill/Stamina/Luck triad, is statistically punishing even for a character rolled at or near the maximum. And its win condition, three specific keys found in three specific locations and carried to a final chest that will not open without them, cannot plausibly be discovered on a first attempt. The book does not merely tolerate failure; it requires it as a precondition of completion.

This essay takes as its starting point a set of observations made about exactly this text: that Firetop Mountain is built around mandatory trial-and-error, that its combat is more lethal than its reputation suggests, and, most interestingly, that despite anticipating repeated failure as a structural feature, the book and its contemporaries never formalized any mechanism for checkpointing progress. Every reader who has played these books, it turns out, independently invented the same workaround: keeping a finger wedged between pages, or maintaining a mental map of which forks lead to death, and restarting not from paragraph 1 but from whatever point they judge safe. This is a fascinating case of a design gap being silently patched by an entire generation of players using the same ad hoc technique, without the designers ever acknowledging or formalizing it. What follows is an attempt to unpack why that gap exists, what it reveals about the design culture of early gamebooks, and what it implies for anyone, myself included, in the context of RPG design work like Heroic Evoke, building punishing, exploration-based systems today.

Trial-and-error as the actual genre

It is worth being precise about what kind of failure Firetop Mountain traffics in. There are, broadly, two ways a branching narrative can punish the reader. The first is narrative failure: you make a choice, the story goes somewhere worse, but you are not ejected from the fiction: you simply experience a different, lesser outcome. The second is systemic failure: you die, or you reach the end without having satisfied a hidden condition, and the book tells you, in effect, that this playthrough does not count. Firetop Mountain is overwhelmingly the second kind. Wrong turns do not degrade the story; they end it. And because the win condition depends on collecting specific items scattered behind specific doors, “playing well” in the sense of making sensible in-fiction decisions is not sufficient to win. You can behave exactly as a sensible adventurer would, avoiding obviously dangerous rooms, retreating from unfavorable fights, husbanding resources, and still arrive at the vault without the third key, because nothing in the fiction told you that a particular closet, easily skippable, contained something essential.

This is the sense in which the golden path is undiscoverable on a first attempt: not merely difficult to find, but epistemically inaccessible. The information needed to route correctly through the dungeon is distributed across paragraphs the reader has, by definition, not yet read, and there is no in-fiction signal, no foreshadowing, no NPC hint reliably present on a blind run, that would let a first-time reader infer the correct sequence. The book is not testing tactical judgment so much as testing memory of a previous failed attempt. Structurally, this makes Firetop Mountain closer to a maze with a single correct solution overlaid on a combat gauntlet than to the “choose your own adventure” branching-story format the surface presentation suggests. The branches are real, and the prose response to each choice is often well-crafted and specific, but the game underneath is a memorization exercise disguised as an exploration narrative.

The combat system compounds this. The received wisdom about Fighting Fantasy, repeated often enough that it functions as folklore, is that a character rolled with maximum or near-maximum Skill can “walk through” the book, and that stories of narrow escapes or unexpected deaths are the result of poor play or bad luck rather than the system itself. Actual play does not bear this out. The 2d6-based attack roll compared against a fixed Skill score produces win probabilities that degrade sharply as the number of consecutive fights increases, and Firetop Mountain in particular stacks several fights back to back with limited opportunity to recover Stamina between them. A character with Skill 12 (the maximum) still loses individual combat rounds at a non-trivial rate, and Stamina, once spent, is not easily replenished mid-dungeon. Running a full campaign of the book, as opposed to a single isolated encounter, exposes how close to death a “well-built” character can come, repeatedly, even when played with full knowledge of the correct route. The claim that skill investment alone insulates a player from death is a comforting simplification that does not survive contact with the dice.

Why this matters for design, not just nostalgia

It would be easy to treat all of this as an artifact of an unsophisticated early design era, interesting only as history. But the trial-and-error structure of Firetop Mountain is load-bearing, not accidental sloppiness. A gamebook is a single physical object with a fixed number of paragraphs. It cannot dynamically adjust its difficulty, cannot present a hint system that reads the player’s history, cannot do anything a modern branching-narrative video game can do with save states and adaptive text. Within those constraints, requiring repeated attempts is one of the few ways to extend a book’s playable lifespan and to make a genuinely non-linear structure worth the printing cost of the wrong branches. If the correct path were discoverable on a first read, most of the book’s paragraph budget, the ones describing wrong turns, would be content nobody experiences on their winning run. Punishing exploration and demanding memorized replay is, in that light, a way of ensuring the reader consumes more of what was written. The difficulty is not merely thematic (dungeons are dangerous, so the book should be dangerous) but economic and structural: it is what makes non-linearity worth the paper.

This reframes the “why didn’t they include save points” question. It is not that Livingstone and Jackson failed to imagine a Souls-like checkpoint system and stumbled into trial-and-error by accident. It is closer to the reverse: trial-and-error was the design goal, and a formal checkpoint mechanic would have partially undermined it, by letting players resume from deep in the dungeon with full knowledge but without having to hold that knowledge themselves. The tension is between two different currencies of progress, player knowledge versus formalized bookkeeping, and the gamebook format, lacking any bookkeeping tool beyond a pencil and the reader’s own memory, defaults to the former. Every reader who kept a finger between pages was not circumventing the design; they were completing it. The “official” absence of a checkpoint mechanic and the near-universal, independently-reinvented presence of an unofficial one are two halves of the same system.

The Dark Souls comparison, and where it breaks down

The comparison to bonfires is illuminating precisely because it clarifies what a gamebook cannot do. A Souls-style checkpoint is a piece of world simulation. Enemies respawn, but the player’s build, equipment, and accumulated knowledge persist. Crucially, the game remembers things a gamebook physically cannot: which enemies you have already fought, which shortcuts you have unlocked, which items you have picked up, all without requiring you to hold that state in your own head or on a scrap of paper. A gamebook has none of this infrastructure. Every piece of persistent state, inventory, wounds, which doors are now known to be traps, has to be tracked manually by the reader on a character sheet, and the book itself has no way of “knowing” that you have been here before. A formal checkpoint system in a gamebook would therefore have to be something much cruder than a bonfire: probably a set of named paragraphs the reader is instructed to write down, functioning as bookmarks, from which they are told they may resume after death without penalty.

This is not technically difficult to design: the essay’s opening observation is correct that it is a strange omission, not an impossible one. A short table of “safe waypoints” tied to specific paragraph numbers, explicitly sanctioned by the rules text (”if you die, you may resume from the last waypoint you reached, retaining your equipment and any keys found, but starting combat-fresh”), would have cost almost nothing in production terms and would have removed the read-between-the-lines quality of the finger-in-the-pages technique. That nobody in the early Fighting Fantasy line formalized this, despite an entire readership converging on the same informal solution, says something about the design culture of the moment rather than about any technical constraint.

Three explanatory threads are worth separating here, because they are not equally strong.

The first is the “no available model” explanation: that gamebooks predate the video-game save state as a widely understood convention, so the designers had no genre precedent to draw from. This is only partially convincing. Save states existed in some early computer games by the early 1980s, and more importantly, gamebooks did not need to borrow from video games at all: a “you may resume from here” instruction is a purely textual convention, no more technically demanding than the branching structure itself. The absence of a cultural template explains why the idea might not have occurred to the designers unprompted, but it does not explain why nobody added it once the trial-and-error problem became obvious through actual play.

The second thread is the tabletop RPG inheritance: gamebooks descend directly from Dungeons & Dragons-style play, where character death is often final and the campaign continues with a new character rather than a rewound one. This explanation is stronger, because it identifies a genuine structural transplant. In a tabletop game, permanent character death is bearable because the table persists: the GM, the other players, the ongoing campaign world all continue regardless of any one character’s fate, and a new character can be woven into that continuity in ways that cost the group very little. A solo gamebook has no table to fall back on. There is only the reader and the book. Transplanting “character death is permanent, roll a new one” from a group activity into a solitary one removes the very thing that made the convention tolerable in its original context. The designers imported the rule without importing the social scaffolding that made the rule work, and the result is the “multiverse” absurdity correctly identified in the source material: a reader who has died six times, and is now on a seventh nominal protagonist, but who is functionally the same continuous agent, holding the same accumulated knowledge, simply wearing a new character sheet as a formality.

The third thread, the fear that an official checkpoint would make the book “feel” too easy or too short, is the most speculative but also the most design-relevant. If accurate, it suggests the designers were, consciously or not, using unnecessary friction as a proxy for value: a book that takes twenty attempts to complete feels like it contains more game than one that takes three, even if the actual authored content is identical either way. This is a design trap that persists well beyond gamebooks: the conflation of “time spent failing” with “value delivered” shows up constantly in modern game design discourse, usually under less flattering names. It is worth naming here because it is the one explanation that implicates the designers’ priorities rather than their tools or their genre inheritance, and it is the one most directly relevant to anyone building a punishing system today: length achieved through repetition is not the same thing as length achieved through content, and conflating the two produces exactly the kind of dissonance described below.

The metagame dissonance

The most interesting failure mode identified is not mechanical but experiential: the split between the player and the character produced by repeated death-and-restart. On a nominal narrative level, each new attempt is a new protagonist entering the dungeon for the first time, with no in-fiction memory of previous attempts. On the actual experiential level, the player piloting that new protagonist knows exactly which door leads to a giant spider and exactly which fork conceals the second key, because they died learning that on a previous run. The character is notionally naive; the player is not. This produces a strange puppetry effect, in which the fiction’s claim that “you have never been here before” is constantly and silently contradicted by the reader’s behavior, which is that of someone navigating from memorized notes.

Most branching-narrative media that involve failure and retry have to solve this problem somehow, and it is worth noting how differently other media have handled it, if only to sharpen what makes the gamebook case distinctive. A video game like Dark Souls solves it by not pretending the amnesia is total: your character explicitly retains memory, items, and levels across deaths, and the fiction is built around a curse of undeath that makes repeated dying part of the premise rather than a contradiction of it. A visual novel with a “true ending requires multiple playthroughs” structure often leans the other way, explicitly narrativizing the loop: the protagonist is looping through time, or the multiple playthroughs represent different timelines being compared by an outside observer. What both of these solutions share is that the fiction absorbs the metagame: the game finds an in-world justification for the fact that the player, not just the character, is accumulating knowledge across attempts.

Classic Fighting Fantasy gamebooks do neither. They offer no fictional frame for repeated attempts at all: no reincarnation myth, no “the mountain remembers your death and rearranges itself,” nothing. The rules text simply says: if you die, start again as a new adventurer. This leaves the dissonance fully exposed rather than absorbed, and it is precisely because nothing in the text acknowledges it that players had to solve it themselves, silently, through the finger-in-the-pages technique, a purely mechanical hack with no narrative justification offered or needed, because the book never asked for one.

Implications for design work in the same space

Anyone building an exploration-heavy, failure-punishing system today inherits this same design fork, whether the medium is a gamebook, a dungeon crawler, or a solo tabletop game built around procedural risk. Three implications seem to follow directly from the analysis above.

First, if a system is going to require trial-and-error as its actual mode of play, as opposed to trial-and-error being an unintended side effect of high lethality, that requirement should be stated honestly, in the rules text, rather than left implicit and discovered by the reader through repeated failure. A rules line that says, in effect, “you are expected to die and restart; here is how restarting works” costs nothing and removes an entire category of reader confusion about whether they are “playing it wrong.”

Second, the choice between narrative death-and-restart and a formal checkpoint mechanic is not a binary between “hardcore” and “casual” design; it is a choice about where the burden of continuity is placed. Placing it entirely on the reader’s memory (the classic gamebook approach) externalizes bookkeeping onto the player at the cost of producing exactly the character/player dissonance described above. Placing it in the system, even minimally, with a short list of sanctioned waypoints and a rule for what persists across a death, removes the dissonance without necessarily removing the difficulty. A punishing system does not need permadeath-with-total-narrative-amnesia to be punishing; the punishment can live in resource loss, position loss, or narrative complication at the checkpoint, rather than in the fiction of a wholly new, wholly ignorant protagonist.

Third, and most specifically to combat-heavy dungeon material in the Fighting Fantasy tradition: reputation is not evidence. The received wisdom that a maxed-out character trivializes the book is exactly the kind of folklore that survives because nobody actually plays the full campaign back to back and tracks the near-deaths. Designers working in this tradition, including in a rules-on-Fighting-Fantasy project like Heroic Evoke, inherit both the genre’s difficulty curve and its reputation for being easier than it is. Actual play data (or, absent that, actual full playthroughs, tracked honestly) is the only reliable check against inheriting the folklore along with the mechanics. If a system’s community consensus says “high Skill trivializes combat” and actual multi-fight sequences say otherwise, the system is being misjudged by its own players in a way that will eventually surface as unexpected frustration, exactly the frustration the original voice memo describes experiencing firsthand, running a Fighting Fantasy campaign and repeatedly coming close to a death the received wisdom said should not have been possible.

A closing observation

The most durable image from this whole discussion is not the dungeon, or the keys, or the spider: it is the finger wedged between two pages. An entire readership, working in total isolation from one another, arrived independently at the same unofficial patch for the same unacknowledged design gap, and it took decades of players comparing notes to discover that this was not an idiosyncratic cheat but a near-universal convergent solution. That convergence is itself the strongest possible evidence that the gap was real: when a large, disconnected population of users independently invents the identical workaround for a system, the system is missing a feature, not testing the users’ resolve. The interesting design lesson is not “gamebooks were too hard”: difficulty was clearly intentional and, in its own terms, well-motivated by the format’s constraints. Rather, the designers mistook the absence of an acknowledged solution for the absence of a problem, when in fact the problem was simply being solved silently, off the page, by every single reader, the same way, without being told to.

Yesterday I wrote about rocketpunk as a genre: the science fiction that takes the rocket equation seriously enough to let it hurt, that trades warp drives for delta-v budgets and force fields for spin gravity. I didn’t mention at the time that I’d already spent the better part of a year building a game inside that genre. Today I’m putting it in your hands. It’s called Loner: Rocketpunk, and it’s a solo tabletop RPG set in a 2026 where the Cold War never left orbit.

PDF on DriveThruRPG

Softcover on Amazon

I want to talk about why I made it, and what it actually is, before I talk about what’s in the box.

The Premise I Couldn’t Shake

The setup is simple to state: Sergei Korolev survives a routine surgery in 1966 that killed him in our timeline, and the Soviet space program doesn’t lose its chief architect at the exact moment it needed him most. The N1 rocket gets the reliability engineering it never got in reality. The competition with Apollo doesn’t slacken after 1969, it sharpens. Both superpowers keep building, keep funding, keep racing, because neither one can afford to be the side that stopped first.

Nothing else about the twentieth century needs to change for the rest to follow. Lunar bases by the 1980s. Mars settlements by the 2000s. Asteroid mining around Ceres and Vesta by 2011. A hybrid UN-style regulatory body, the Orbital Nations Alliance, trying to keep order across a solar system that’s been claimed a dozen different ways by governments and corporations who don’t fully trust each other. By 2026, where you play, there are people who were born on Mars. They’re five years old. Nobody quite knows what that means yet, for them or for the species.

I didn’t invent this idea, not really. It’s the premise of For All Mankind, and it’s been sitting in the back of science fiction for decades, this quiet what-if about a future we were, for a few years in the 1960s, actually building toward. What I wanted was to hand someone the controls of that future and let them live in it for an evening, alone, with four dice and a pencil.

Why Solo, and Why Rocketpunk Specifically

I play a lot of solo RPGs. There’s something about sitting with a game by yourself, letting an oracle answer questions you don’t know the answer to, that produces a particular kind of story you can’t get any other way. No one is performing for you. No one is managing pacing on your behalf. You ask the dice a question, and then you have to live with whatever they tell you.

Rocketpunk turned out to be an unusually good fit for that mode of play, and it took me a while to understand why. Most solo RPGs are journaling games or dungeon crawls, built around exploration or introspection. Rocketpunk fiction is built around constraint: limited fuel, limited air, limited time before the next resupply. A solo player is already working under constraints of their own, the ones that come from being one person doing a job a whole table usually splits between six. Put those two kinds of pressure together and they reinforce each other instead of fighting. Every choice in a Loner session already carries weight because it’s yours alone. Every choice in a rocketpunk story already carries weight because the delta-v doesn’t lie. The overlap felt less like a design decision and more like something I noticed was already true.

What’s Actually in the Book

I built Loner: Rocketpunk around four commitments, and I held to them stubbornly enough that they shaped almost every other decision in the text.

Portable. You need four six-sided dice, two colors, and something to write with. That’s the whole kit. I wanted a game you could run from a jacket pocket on a train, not one that needs a shelf.

Rules-light. There’s one core resolution mechanic. You learn it in the first session and you’ll have it memorized by the second. I kept cutting subsystems out of early drafts because I noticed I was building them to feel productive, not because the game needed them.

Tag-based. Characters, ships, locations, NPCs, even a malfunctioning airlock, are all described the same way: with descriptive tags instead of numeric stats. Your protagonist isn’t a spreadsheet of attributes. They’re a Radiation-Scarred Salvage Operator with a Debt to a Crime Syndicate and a Jury-Rigged Thruster Pack, and those descriptions do the mechanical work that numbers would do in another system. When a tag is relevant to what’s happening, it shifts the odds. When it isn’t, it’s just true about the character and stays out of the way.

Emergent and open-ended. There’s no plot in the book for you to walk through. The Oracle answers yes-or-no questions, weighted by how likely you judge the answer to be, and the story gets built out of your questions and its answers, session by session. Nobody wrote the ending in advance. Including me.

Everything else in the book, the character creation tables, the timeline of the divergence from our own history, the shipyard full of named vessel classes, the situation tables for when you need a scene to start moving, exists to serve those four commitments rather than to pad the page count. If a table didn’t earn its place, I cut it.

The Parts I’m Proud Of

The tag system is the thing I’d point to first. I spent a long time trying to make a numeric skill system work for solo play before I gave up on it entirely. Numbers are precise, but precision wasn’t actually what the game needed. What it needed was for a character’s history to matter mechanically without turning character creation into an accounting exercise. Vacuum Specialist tells you something a “+2 to EVA checks” never will, and it still does exactly the same job at the table when you roll.

I’m also proud of the setting’s restraint. It would have been easy to let the alternate history run wild, to pile on speculative technology until it stopped resembling anything we could plausibly have built. I kept pulling it back. Analog electronics stay analog because vacuum tubes actually are more radiation-resistant than early solid-state components, and in a timeline where nobody skipped a decade of hardening research, that fact still matters in 2026. The bulky suits stay bulky because that’s what life support looks like when you’re not allowed to invent your way past physics. Every retro detail in this book is there because it’s still true, not because it’s aesthetically pleasing. The aesthetic came out of the truth, not the other way around.

What I Hope Happens Next

I don’t have a launch trailer for this. I don’t have a marketing plan. What I have is a game I wrote because I wanted to play it, and a genuine curiosity about what happens when other people sit down with it and start asking the Oracle questions I never thought to ask.

If you play a session, I’d like to hear what your character’s frailty turned out to cost them. I’d like to know which tag ended up mattering more than you expected when you wrote it down. That’s the whole reason I built this the way I did: not to hand you a story, but to hand you the tools to find out what your story is, four dice and a pencil at a time.

The future we almost had is yours now. Go build it.

There’s a particular kind of science fiction that doesn’t dream in warp drives or laser swords. It dreams in delta-v budgets, in the groan of a hull under spin gravity, in the terrible arithmetic of how much reaction mass you’ll need to slow down again once you’ve sped up. This is rocketpunk, and even if you’ve never heard the term, you’ve probably felt its gravity anyway: through For All Mankind, an Atomic Rockets deep-dive at 2am, or a tabletop game where your ship’s engine has a real name and a real exhaust velocity.

Rocketpunk is what happens when science fiction takes the “science” half of its name seriously enough to let it hurt.

What Rocketpunk Actually Is

The term itself is younger than the genre it describes. It’s most associated with the blog Rocketpunk Manifesto, written by Rick Robinson in the mid-2000s, which used it to describe a specific flavor of “in-system” science fiction: stories confined to our own solar system, powered by extrapolations of real rocketry rather than invented physics. No warp drives, no wormholes, no convenient artificial gravity fields humming away in the background. Just thrust, mass, and time: the three things that actually govern how you get from Earth orbit to the Jovian moons.

Where space opera asks “what if we could go anywhere, instantly, and it were easy?”, rocketpunk asks “what if we could only go where the rocket equation allows, and it were brutally hard?” That single shift in assumptions changes everything downstream of it. Distance becomes a character. Fuel becomes a plot device. A trip to Mars isn’t a scene transition; it’s an ordeal measured in months, with its own dangers, its own economy, its own psychological toll.

This is science fiction with its shoes on the ground, even when the ground is four hundred million kilometers away.

The Rocket Equation Is the Genre’s God

Every rocketpunk story lives or dies by one piece of math: the Tsiolkovsky rocket equation, which relates a spacecraft’s change in velocity to how much propellant it burns and how efficiently it burns it. It’s not exciting to look at. But its consequences ripple through everything the genre cares about.

Because propellant is heavy, and heavy things are expensive to launch, every gram of reaction mass a ship carries is a gram it had to justify. That means ships in rocketpunk fiction don’t casually change course, don’t outrun pursuers with a dramatic burst of speed, don’t loiter indefinitely waiting for the plot to catch up. They commit to trajectories the way a diver commits to a dive. Once you’ve burned the fuel for a transfer orbit to Ceres, you’re going to Ceres, on a schedule set by orbital mechanics rather than narrative convenience.

This produces a very particular kind of tension. Rocketpunk stories are full of characters doing the math out loud: checking whether they have enough delta-v for a rendezvous, whether a rescue burn will leave them stranded, whether the numbers on the fuel gauge are lying by even a percentage point. It’s a genre where a spreadsheet error can be as lethal as a villain.

No Force Fields, No Free Lunch

Compare this to the assumptions baked into most popular space fiction. Star Wars ships bank and swoop through vacuum like fighter jets in an atmosphere, because nobody involved was trying to model orbital mechanics; they were making a war movie in space. Star Trek ships have inertial dampeners and artificial gravity as background furniture, technologies so taken for granted they’re rarely even named on screen. These aren’t criticisms; those shows are doing something else entirely, and doing it well. But rocketpunk draws its line in a different place.

In a rocketpunk story, gravity mostly comes from spin: a rotating hab ring or a tethered counterweight, with the attendant Coriolis effects that make a bucket of water behave strangely and a thrown ball curve wrong. Radiation is a constant hazard, not a plot-convenient obstacle that shows up for one episode. Vacuum kills you not with a suit breach and a soundtrack sting, but with cold, mundane speed. Ships look less like sleek arrows and more like ungainly assemblies of fuel tanks, radiators, and habitat modules bolted around a drive stage, because that’s what you get when form has to follow the actual function of surviving in space rather than looking heroic silhouetted against a nebula.

The absence of technobabble shortcuts is the point. When you can’t invent your way out of a problem, the problem becomes the story.

A Genre With Deep Roots

Rocketpunk didn’t spring up whole in the 2000s; it’s a return trip to a much older tradition. You can trace a direct line back to the “hard SF” writers of the mid-20th century: Robert Heinlein’s juveniles, with their meticulous descriptions of orbital rendezvous and lunar economics; Arthur C. Clarke, an actual satellite-communications pioneer who wrote fiction the way an engineer sketches blueprints; the whole tradition of magazine SF that treated the solar system as a frontier you’d have to earn, mile by expensive mile.

The visual language has its own lineage too. Chris Foss’s paintings of bulbous, industrial spacecraft, all pipes and tanks and sun-bleached hull plating, shaped how a generation pictured what “realistic” spaceships should look like: not aerodynamic, because there’s no air to be aerodynamic in, but functional in the way an oil tanker or a drilling rig is functional. The Terran Trade Authority books of the 1970s did something similar, presenting invented spacecraft with the deadpan technical specificity of an aircraft recognition manual.

More recently, the genre found a second wind. For All Mankind is probably rocketpunk’s most visible modern ambassador: an alternate history in which the Space Race never ended, built almost entirely on the premise that if we’d simply kept funding the hardware, the rest would follow. Its drama comes from EVA checklists, docking procedures, and the political will to pay for delta-v, not from any invented physics; the show’s tension is generated by the same unglamorous math that governs the rest of the genre. Andy Weir’s The Martian is rocketpunk distilled to a single, obsessive question: given what we actually know about rocket science, botany, and chemistry, how does one man survive Mars with nothing but competence and duct tape? The film Gravity spent ninety tense minutes proving that orbital mechanics alone, without a single alien or laser, can carry an entire thriller.

The genre’s prose tradition never really went away, either. Allen Steele’s Near Space novels, beginning with Orbital Decay, read like they were written by someone who’d actually worked a shift on a space station: unionized construction crews, budget overruns, and the unglamorous labor of building infrastructure in orbit. Ben Bova’s Colony, part of his sprawling Grand Tour sequence, works the same seam from the other direction, following the political and economic strain of building a permanent settlement at L5. Both writers treat spaceflight the way a labor novelist treats a factory floor: as a workplace first, an adventure second.

It’s worth remembering, too, that rocketpunk was never purely a fiction project. Some of its most load-bearing ideas came from engineers and physicists writing non-fiction, not novelists writing speculation. Gerard K. O’Neill’s The High Frontier is probably the single most influential of these: a serious, technically grounded proposal for building rotating space colonies at the Earth-Moon Lagrange points, complete with mass drivers, solar power satellites, and population projections. O’Neill wasn’t writing science fiction; he was writing an engineering argument. But the cylinders and tori he sketched became the default architecture for “realistic” space habitats in fiction for decades afterward. A rocketpunk writer describing a spinning habitat, however casually, is almost always describing an O’Neill cylinder whether they credit him or not.

Why the Constraints Are the Appeal

It would be easy to assume a genre this committed to realism is just science fiction for people who enjoy homework. That’s not quite it. The appeal of rocketpunk isn’t the math for its own sake; it’s what the math forces the story to be about.

When ships can’t just go anywhere, distance becomes consequential again, and so does isolation. A crew stranded past Jupiter isn’t a phone call away from rescue; they’re weeks or months from anyone who could help, and every decision they make has to account for that. When fuel is finite and expensive, economics stops being background texture and becomes the engine of conflict: who controls the mass drivers, who owns the water ice on that asteroid, who can afford the delta-v to enforce a claim nobody else can reach in time to contest it. When gravity is manufactured by spin rather than assumed, the body itself becomes a site of story: bone density loss, vestibular disorientation, the particular vertigo of walking “up” a curved deck.

Rocketpunk trades the infinite playground of space opera for something narrower and, often, sharper: a solar system that behaves like the one we actually live in, populated by people who have to solve problems the way engineers solve problems, under constraints that don’t bend for anyone’s convenience. The drama comes from ingenuity operating at the edge of what physics allows, not from physics quietly stepping aside.

Playing It at the Table

This is also, not coincidentally, a genre that translates unusually well to tabletop RPGs, because tabletop play already rewards the kind of granular, consequence-driven decision-making rocketpunk fiction is built on. A game where your ship’s propulsion type is a mechanical fact, where an ion drive gives you efficient, patient acceleration and a nuclear pulse rocket gives you brutal thrust at the cost of everyone’s nerves, turns “how do we get there” into a real question with real trade-offs, not a formality on the way to the next set piece.

Settings built on this foundation use the same textures the fiction does: debris cleaners and space miners as viable, dignified professions rather than backstory color; a taxonomy of real propulsion concepts (ion, nuclear-thermal, mass driver, solar sail, nuclear pulse) that players learn to reason about the way they’d learn a magic system elsewhere; starship classes defined by mission profile rather than aesthetics. It’s a genre where knowing your ship’s specs isn’t optional trivia. It’s how you survive the session.

Played well, a rocketpunk table has the same texture as the best of the genre’s prose: quiet stretches of real tension while a burn completes, sudden crises when a number doesn’t add up, and a sense that the universe isn’t rooting for you or against you; it’s just indifferent, and that indifference is the whole challenge.

The Future That Might Still Happen

Part of what keeps rocketpunk relevant, even decades on, is that it isn’t really about some other future. It’s about the plausible one: the solar system we could, with enough will and enough decades, actually go build. Warp drives ask nothing of us but imagination. Rocketpunk asks whether we could really do this, with the physics we’ve already got, if we decided it mattered enough.

That’s a heavier kind of speculation than most science fiction bothers with. It’s also, for a certain kind of reader and a certain kind of player, the only kind that feels like it counts.

On May 12, 2019, an anonymous user on 4chan posted a photograph to a thread asking people to share images that felt subtly wrong. The photograph showed a windowless room: yellow walls, damp beige carpet, fluorescent lighting buzzing at a frequency just outside comfortable. No people, no furniture beyond a few scattered chairs, no visible exit. The caption warned that if you were not careful, you could slip out of reality and end up trapped in a place like this, an endless expanse of disconnected, empty rooms, with nothing to keep you company but the hum of the lights.

That single image, later identified as a furniture showroom in Oshkosh, Wisconsin, seeded one of the most durable horror mythologies the internet has produced. Within a few years it had a name, the Backrooms, a wiki, thousands of user-submitted “levels,” and a growing bestiary of entities. In 2022 a seventeen-year-old filmmaker named Kane Parsons turned the concept into a found-footage YouTube series that accumulated tens of millions of views. By 2026 it had become a theatrical film distributed by A24. Somewhere along that trajectory, the Backrooms also became the reference point for a broader aesthetic category that internet culture started calling “liminal spaces”: photographs of empty malls, motel corridors, swimming pool complexes, waiting rooms, all sharing the same quality of a place built for crowds and photographed without any.

This essay is not about the Backrooms as a media franchise. It is about the feeling the original photograph produced, before anyone attached a mythology to it, and about how that feeling gets systematically diluted the moment a mythology is attached. My claim is that the internet has spent seven years describing a very specific and very old horror sensation using the wrong vocabulary, borrowed from an unrelated academic tradition, and that once you correct the vocabulary, you can see a second, sharper horror hiding underneath the first one, a horror that has almost nothing to do with architecture and everything to do with the collapse of your own social existence.

To make that argument I need to walk through three things in order: the word that actually describes what these photographs make you feel, the word that internet culture uses instead and where that word actually comes from, and why the gap between the two matters for how horror fiction gets built.

Naming the Feeling: Kenopsia

In 2009, the writer John Koenig started an online project called The Dictionary of Obscure Sorrows, later published as a book by Simon & Schuster in 2021. The project invents words for emotional states that English has never bothered to name. One of those words is kenopsia, built from the Greek kenos, meaning empty, and opsia, meaning seeing. Koenig defines it as the eerie, forlorn atmosphere of a place that is usually full of people but is currently abandoned and silent: a school hallway at night, an office on a weekend, a fairground out of season. He describes the effect as an emotional afterimage, a room that reads as hyper-empty rather than simply empty, because the population you expect to find there is so conspicuously missing that its absence becomes a kind of presence.

Kenopsia is precise in a way that matters for this argument. It names an atmosphere, not a structure. It says nothing about thresholds, doorways, or the geometry of a space. It is purely about the mismatch between a place’s evident social function and its current emptiness. It should also be kept separate from kenophobia, the clinical fear of vast open or empty spaces, which is a phobia in the medical sense, a fear response, not an atmosphere you can register from a photograph without ever having set foot in the room. Kenopsia is what you feel looking at a picture of an empty school gym. Kenophobia is a diagnosis.

The word that actually spread across the internet to describe the Backrooms and its cousins, however, was not kenopsia. It was “liminal.”

The Liminal Confusion

“Liminal” comes from the Latin limen, threshold. Its modern conceptual life begins with the folklorist Arnold van Gennep, who in 1909 described rites of passage as having three phases: separation from an old status, a liminal phase of being between statuses, and reincorporation into a new one. Sixty years later, the anthropologist Victor Turner expanded the liminal phase into a full theory of social transition, describing people undergoing it as “betwixt and between,” structurally invisible because they no longer belong to their old category and do not yet belong to their new one. In this original sense, liminality is not about empty rooms at all. It is about persons occupying an ambiguous status during a ritual, and by extension, about spaces that mark that ambiguous transition: the threshold of a doorway, a corridor between two rooms, an airport departure lounge between two countries.

When the “liminal space” aesthetic exploded on Reddit and TikTok around 2019 and 2020, alongside the rise of the Backrooms, it borrowed this vocabulary but not this meaning. Nobody posting a photo of an empty hotel corridor on the liminalspace subreddit was making a claim about ritual status. They were reaching for a word that sounded academic and evocative to describe something closer to kenopsia: a familiar transitional space, drained of the people who would normally pass through it. The word “liminal” survived the transfer. Its actual content did not. What took hold was a folk usage that fused two genuinely separate ideas, the anthropological threshold and the atmospheric emptiness, into one loose umbrella term, and horror discourse inherited that fusion wholesale.

This matters because it leaves “liminal horror” permanently ambiguous about what, exactly, is supposed to be frightening. Is the threshold itself unstable, structurally wrong, promising two destinations and delivering neither, in the Van Gennep and Turner sense? Or is the space simply the wrong population density for its evident function, in the Koenig sense? These produce different fears through different mechanisms. A corridor that seems to loop back on itself is frightening because your spatial reasoning fails. An empty cafeteria at 3 a.m. is frightening because your social reasoning fails, you expect noise and motion and get silence instead. Popular horror writing about the Backrooms uses both registers interchangeably, and as a result rarely commits to either one long enough to develop its full implications.

There is, however, a more useful way to hold the two traditions together than simply picking one and discarding the other, and it comes from taking Turner’s account of liminal persons more seriously rather than less. Turner describes people undergoing the liminal phase of a rite as “betwixt and between,” stripped of their old social markers and not yet granted their new ones, which makes them, for the duration of the rite, structurally invisible to the ordinary categories of their community. That invisibility is not a bug in the ritual. It is the entire point, and it is always temporary by design. The rite exists precisely to move a person through that phase and back into visibility, into a reincorporated status on the other side. A wedding, an initiation, a graduation, all of them stage a controlled, time-limited disappearance from the social order, followed by a guaranteed return. Read this way, the Backrooms photograph, and the fear it produces, can be described with more precision than the loose internet usage of “liminal” allows: it is a rite of passage that gets stuck in its middle phase and never completes. The threshold is crossed, the old status is shed, and the return never comes. What Van Gennep and Turner describe as a safe, structured, communally supervised passage becomes, in kenoptic horror, a passage with no supervision and no scheduled end. This is the correct technical relationship between the two ideas: not liminal horror as a vague cousin of kenoptic horror, but kenoptic horror as what liminality becomes when its third phase is permanently withheld.

Where Liminal Horror Stops

The Backrooms mythology did not stay with its founding image for long. Within months of the original 2019 post, the concept had accumulated user-generated “levels,” each with its own geography, hazards, and rules, and eventually a large cast of hostile entities with names like Smilers and Hounds. Journalists covering the phenomenon noted this shift directly: what had made the original short film distinctive, according to critics at outlets like Kotaku, was precisely its restraint, the near-total absence of overt threat, with dread produced instead by a diffuse sense of wrongness in the space itself. The subsequent expansion into levels and entities moved the property away from that restraint and toward something structurally closer to a survival game: an environment with rules to learn, monsters to avoid, and an implicit objective, find the exit, survive the level, reach the next area.

This is not a criticism of the Backrooms as entertainment. Collaborative fan mythologies grow because growth is what keeps a community engaged, and a shared universe with named entities and mapped levels is more sustainable as an ongoing creative project than a single static image. But it is worth being honest about what that growth costs the original feeling. A space with a monster in it is a space with a problem to solve. The monster gives you an adversary, and an adversary gives you agency: you can run, hide, fight, negotiate. The horror becomes conventional, structured around threat and evasion, which is the oldest and most reliable engine in the medium, but it is not the engine that made the initial photograph unsettling. That photograph had no monster. It had no threat at all in the conventional sense. It had only the wrongness of an empty room that should not have been empty, and the unanswered question of how you would ever get out of it, or tell anyone where you had gone.

The Backrooms is not unique in this respect. Journalists have grouped it together with the SCP Foundation, another crowd-authored horror wiki, under a shared genre label of collaborative online horror. The comparison is instructive, because the SCP Foundation faced the same pressure toward encyclopedic completeness years earlier: an initial handful of unsettling anomalies gave way, over time, to thousands of catalogued entries, each one tidily explained, contained, and cross-referenced. Completeness is what keeps a wiki-based mythology alive as a participatory hobby, but completeness is corrosive to dread. A world in which everything strange has eventually been named, numbered, and filed away is a world that has quietly stopped being frightening in the way it started out being frightening, even while it keeps adding new frights to the catalogue. The same drift happened to the Backrooms almost immediately, and it is worth naming as a structural tendency of the format itself, not a failure specific to any one contributor.

That second half of the sentence, how you would tell anyone where you had gone, is where I think the real subject of this essay actually lives, and it is not covered by either half of the liminal-kenopsia fusion. Both van Gennep’s threshold and Koenig’s atmosphere describe something you observe. A liminal corridor is unsettling to walk through. An empty gymnasium is unsettling to look at. Neither concept, on its own, addresses what happens when you are not just observing the wrongness but are permanently inside it, cut off from any audience who could confirm your experience or come looking for you. That is a different category of fear, closer to abandonment and erasure than to spatial or atmospheric discomfort, and it deserves its own name and its own theoretical grounding.

Not Claustrophobia, Not the Desert Island

Before moving to precedent, it is worth ruling out two easy misreadings, because both are the first objection anyone raises when this argument comes up.

The first is claustrophobia. It is tempting to file kenoptic dread under fear of confinement, but the two do not overlap as neatly as they seem to. Claustrophobia responds to physical constriction, low ceilings, narrow corridors, a shrinking room. Much of the Backrooms imagery is the opposite of constricting: vast, low-lit, open-plan spaces that go on for what feels like miles. The dread survives the removal of any physical constraint whatsoever, which tells you the mechanism is not spatial confinement but social and existential absence.

The second is the desert island or shipwreck tradition of isolation fiction, the Robinson Crusoe model where a person is stranded somewhere genuinely remote from human contact. That tradition generates sympathy and adventure more often than horror, because the isolation there is explicable: you know exactly why no one can find you, and the space itself, however hostile, is at least honest about being unfamiliar and far away. Kenoptic dread depends on the opposite condition. The space is not remote or exotic. It is the office you worked in, the mall you shopped in, rendered unreachable not by distance but by some failure in the world’s bookkeeping that has stopped registering you as present anywhere at all. The horror is not being far from home. It is being unlocatable while apparently still standing inside a place that looks exactly like home.

Two Precedents Before the Internet

The internet did not invent this fear. It only gave it a shared visual vocabulary. Two much older sources describe the same structure with more precision than anything the Backrooms wiki has produced, and a third, from clinical psychiatry, gives it a real-world anchor.

The first source is Sigmund Freud’s 1919 essay on the uncanny, Das Unheimliche. Freud’s argument, stripped of its psychoanalytic apparatus, is that the uncanny arises specifically from the return of the familiar in an unfamiliar form, not from the wholly alien. A monster from outer space is frightening in a straightforward way. A childhood home rearranged just slightly wrong is frightening in a way that reaches further, because your recognition systems fire and then fail to complete the match. This is exactly the mechanism behind the first half of the Backrooms feeling: the moquette, the neon, the office furniture are not alien architecture, they are the most mundane architecture imaginable, which is precisely why their emptiness registers as violation rather than mere absence. Kenopsia, in other words, is Koenig’s poetic name for a spatial application of something Freud had already mapped in detail a century earlier. Naming this mechanism explicitly, rather than treating it as a vague online mood, gives it a theoretical spine.

The second and third sources speak to the sharper fear, the one about isolation rather than architecture, and they predate the internet by decades. Algernon Blackwood’s 1907 story “The Willows” follows two canoeists stranded on a shifting island in the Danube who gradually realize they have drifted into contact with something that does not merely threaten their bodies but threatens their standing as witnesses. The horror of the story is not a creature, it barely has one, it is the characters’ dawning understanding that they may be marked for a kind of absorption that will leave no trace anyone else could ever find or interpret. J.G. Ballard’s late short story “The Enormous Space,” from the tail end of the 1980s, pushes the same idea further inward: a man withdraws into his own house, room by room, progressively shrinking the boundary of what counts as real for him, until the house and his consciousness become coextensive and he effectively erases himself from the world outside without any external event forcing it. Neither story features a monster. Both are far more disturbing, on a careful read, than most contemporary Backrooms fiction, because both understand that the worst version of this fear is not being hunted, it is being forgotten by the fabric of reality while still, somehow, alive to notice it happening.

There is also a clinical name for a version of this experience, which is worth mentioning not because horror fiction needs a diagnosis but because it confirms the fear is psychologically real rather than a purely aesthetic internet invention. Cotard’s delusion, first described by the French neurologist Jules Cotard in 1880, is a condition in which patients become convinced that they are dead, that they do not exist, or that the world around them has ceased to exist. It is an extreme and rare presentation, but it demonstrates that the specific terror of believing yourself to be alive and simultaneously absent from the register of the living is not a metaphor invented for internet horror. It is a documented way a mind can fail, which is precisely why fiction that engages with it seriously, rather than decorating it with monsters, tends to land harder than fiction that does not.

Backrooms folklore has its own term for the moment of entry into this state: noclip, borrowed from video game development, where it describes a glitch that lets a player pass through solid geometry the game was supposed to prevent them from crossing. Within the mythology it usually just means the mechanism by which a person enters the Backrooms. But the metaphor is more useful than the folklore tends to notice. Noclipping through a wall in a game does not just move your character to a new room. It moves them outside the collision logic the rest of the simulation obeys. Other objects in the game world no longer register your presence correctly, because you are technically no longer playing by the rules that make mutual detection possible. That is a remarkably exact description of the fear Blackwood and Ballard were writing about long before anyone had a word for it: not being lost, but being unregistered.

Kenoptic Horror: Toward a Definition

I want to propose a name for this second, sharper fear, one that keeps its debt to Koenig’s coinage visible while marking it as a distinct category from both liminality and kenopsia proper. I am calling it kenoptic horror.

Kenoptic horror is a subgenre of psychological horror in which dread does not come from an active threat, a monster, a countdown, a chase, but from the emotional and ontological weight of absolute emptiness applied to a human-built space. Where liminal horror, in its confused popular usage, stops at the geometric destabilization of a transitional space, kenoptic horror is concerned with something that happens after the transition: the trauma of existential isolation once you are already on the other side of it.

That trauma rests on two pillars.

The first is the distortion of familiarity, the kenoptic void proper. The environment is not alien. It is aggressively, almost insultingly human: carpet, fluorescent light, office partitions, the architecture of waiting rooms and retail floors. The horror comes from the short circuit between the space’s evident readiness for use and the complete absence of the humanity that space was built around. This pillar draws directly on Freud’s account of the uncanny: it is the familiar rendered strange precisely because it remains so recognizable. It is a version of the everyday, stripped of everything that gave the everyday its meaning.

The second pillar is personal oblivion, what the Backrooms folklore calls noclipping and what I would describe more plainly as falling out of phase with the rest of existence. This is the true sentence of the space, and it has nothing to do with dying. The fear is not mortality. It is the certainty of having slipped outside the social and temporal line that connects you to the world you left, becoming a ghost in reverse: still alive, still thinking, still afraid, but permanently unregistered by the architecture that surrounds you and by everyone who might otherwise come looking. You are cut off not from safety but from witness.

Held together, these two pillars produce something the entity-and-level model of Backrooms fiction has mostly abandoned in favor of easier, more game-shaped scares. A well-built piece of kenoptic horror needs no monster, because introducing one converts an ontological condition into a survival puzzle, and a puzzle can be solved. What kenoptic horror asks its audience to sit with instead is a condition with no solution on offer, a space that will never again register your presence no matter what you do inside it. That is a harder, quieter, and in my view considerably more durable kind of fear than anything with teeth, and it is the fear the original 2019 photograph actually contained, before seven years of collaborative mythology buried it under levels, monsters, and a marketing budget.

For anyone trying to build kenoptic horror rather than just recognize it, the two pillars translate into a small set of working constraints. No entity should ever be confirmed on the page or on screen, because confirmation converts an ontological threat into a tactical one, and tactics can be planned around. No exit should be found, hinted at, or meaningfully searched for, because a search implies the possibility of success, and possibility is what liminal-horror-as-puzzle is built on. No rescue should ever arrive, not even an ambiguous one, because a rescue retroactively restores the protagonist to the register of the witnessed, closing exactly the wound the story exists to keep open. What should remain, instead, is the world’s own failure to notice: doors that were always this ordinary, light fixtures that keep working for no one, silence that never resolves into either safety or attack. The fear is not that something is coming for you. It is that nothing is coming at all, and nothing ever will, because as far as the rest of existence is concerned, you were never there to begin with.

A few weeks ago I wrote about a specific stall I hit while running a 24XX campaign. The players were doing exactly what the fiction called for, careful, low-profile infiltration, and the system had nothing to say about it. No risk meant no roll. No roll meant no mechanical output. The situation moved only as far as I pushed it, and I found myself increasingly uncomfortable with that authority.

My diagnosis then was that 24XX is built around a single design decision: roll only to avoid risk. That decision is clean and correct in the contexts it was designed for. The problem is structural. The system resolves pressure. It does not produce it.

I left that article with a practical resolution, switch systems for this campaign, but the design question stayed open. What would it look like to patch 24XX itself, without replacing it with something else?

Download 24XX World Patch for free

Resolution and generation

The Fiction Engine Framework gave me the vocabulary to frame the gap precisely.

A fiction engine commits to one test: every rule must serve the fiction, either producing a fictional state or maintaining the conditions under which fictional states can occur. Administrative overhead that does neither is a candidate for removal. 24XX passes this test cleanly. Its resolution procedure is thin, diegetic, and direct. Roll, read the highest die, the fiction changes.

The gap is not in the resolution layer. It is in what the framework calls generation: the production of independent fictional facts that neither player intent nor referee judgment directly authored. In a standard play session with active risks, the dice fill that role, a disaster produces a consequence nobody chose; a setback creates a complication that has to be addressed. The dice act as a second voice in the conversation.

When risks disappear, in careful play, in quiet scenes, in the spaces between confrontations, the dice go silent. The second voice disappears. Generation collapses entirely into the referee, who must decide moment to moment what the world does next, unassisted. This is what I described in the stall article as carrying the entire structure alone. In the framework’s terms, the feedback loop closes over a single agent instead of two. Emergence of the relevant kind is not possible.

The seed

The fix was already in the SRD.

24XX includes one rule that does not fit the resolution pattern: the Luck roll. The GM can roll a die at any time to test luck, checking for (1–2) trouble or (3–4) signs of it. That is, in embryo, a generative oracle. It produces a fact, trouble or the shadow of it, that nobody authored. The problem is that it is underdeveloped. It is GM-discretionary, GM-triggered, and GM-interpreted, which means in practice it tends to function as the referee rolling for permission to do what they were already planning. It is not a second agent. It is a coin flip with a low threshold for ignoring.

The design task was to promote Luck from an occasional procedural flourish into a genuine generative organ, one that fires structurally, not by fiat, and produces outputs the referee is bound to honor.

The World Roll

The result is what the patch calls the Situation die, though for clarity I will continue calling it the World Roll here.

The trigger is the key design decision. The World Roll fires precisely when no skill roll applies, when an action carries no risk and would simply succeed. This makes it mutually exclusive with the resolution engine. When there is risk, the skill roll already produces fictional change; the World Roll would be redundant. When there is no risk, the skill roll is silent; the World Roll fills the gap. The two engines never run simultaneously, and the fundamental rule of 24XX, roll only to avoid risk, is untouched. The new roll answers a different question about a different subject.

The die is read from the fiction rather than tracked on a sheet. You pick from a scale, d4 for a chaotic situation, d12 for a stable one, based on how settled things are right now. This prevents the patch from turning into a pressure engine. A pressure engine works by making a resource visible and managed: the player attends to the economy, not just the fiction. Here the resource is the fictional situation itself. There is nothing to track between rolls. Each read is fresh, taken from the current state of things. As the situation heats up, you reach for a smaller die without being told to, the escalation is diegetic because the meter is the fiction.

Results follow the same pattern as the existing resolution ladder: 1–2 is trouble (an adverse fact surfaces), 3–4 is signs (trouble approaches without landing, and the situation is now less stable, the next die will likely be smaller), 5+ is calm (nothing new, the action lands). One roll, one fictional state, no chained resolution steps.

The binding rule is what makes this a second agent rather than managed fiat. The surfaced fact must be proportionate to what is established, leave something to act on, and follow from prior events. These three criteria, adapted from a heuristic Chris McDowall introduced in the context of open strategy games, are the discipline that separates a generative oracle from a referee rolling for mood. In the final rules text, McDowall’s acronym (RAT: Reasonable, Actionable, Traceable) does not appear. The criteria are stated as plain requirements in the fiction engine’s own vocabulary, and no attribution is needed because the name is gone. But the logic behind them remains the load-bearing rule of the whole patch.

Combat as a specialization

The same gap that appears in quiet infiltration appears in combat, but it takes a different form. In a fight the problem is not that nothing happens, plenty happens, but that the opposition’s actions have no mechanical source. Enemies have no skill dice in 24XX. The GM describes what they do, and that description is entirely authored. If the player group acts without triggering a skill roll, repositioning behind cover, reloading, calling out a warning, the world is silent. What the enemy does in that moment, and whether the fight escalates or begins to break, depends entirely on the referee’s judgment.

The combat module is the World Roll applied to a tactical situation. The die is read the same way, this time from how much the characters control the fight. Ambushed or scattered is d4. Evenly matched is d6. Holding an advantage is d8. Dominating is d10. The opposition broken is d12. The results are renamed to fit the context: Press (a new risk or obstacle appears, the fight grows harder), Shift (the opposition adapts, repositions, changes tactic), Opening (an obstacle clears, a gap appears, or the opposition is ready to break). But the structure is identical to the Situation die. It is the same engine, applied to a specific register.

The Opponents section, Behavior, Risks, Obstacles, provides the diegetic substrate the combat roll needs. Obstacles are the fiction engine’s stand-in for hit points. Each one is a visible fictional condition, armor, numbers, cover, a locked door, that must be addressed with a stated fictional reason before the fight can end. The total number of obstacles determines how long the combat lasts without tracking a number between rounds. Press results can surface new obstacles. Opening results clear them. The World Roll manages fight length by managing the fiction, not a counter.

Break-off follows from the same logic. When the opposition is cornered, the die would read high, and the roll comes up Opening, the fight has reached its natural conclusion. The enemy breaks according to their Behavior: they flee, yield, or bargain. That conclusion was already implied by how 24XX describes opponents. The combat World Roll makes it mechanically available rather than dependent on the referee’s moment-to-moment judgment about when the enemy has had enough.

The pressure engine temptation

The most important decision in designing this patch was what not to do.

The obvious fix for the generation gap is a resource economy: a heat track, a suspicion counter, a momentum pool. These work. Breathless, which is structurally adjacent to 24XX, solves its pacing problem this way. The die degradation and reset economy create a chain reaction where the resource state keeps pushing toward tension. It rarely stalls.

But Breathless is not a fiction engine. The Fiction Engine Framework classifies it as a pressure engine, a design where a managed resource economy sits above the transparency threshold by design, because the strategic visibility of that economy is the experience the game produces. Adding a heat pool to 24XX would not patch the gap. It would change the design family.

The World Roll stays below the transparency threshold because there is nothing to manage. You read the die from the situation, roll, and the result is a fictional state. The resource is the fiction. No additional layer is introduced between the player and the world.

Style and completeness

The final design question was not mechanical but formal: what register should the patch be written in?

24XX’s SRD is compressed. Each section is a handful of sentences. The prose is imperative and load-bearing; there is no explanatory meta-text. The goal was to write the three new sections, Situation, Opponents, Conflict, in that voice, so they read as native additions rather than commentary.

The first attempt was too compressed. It gave the die endpoints (d4 chaotic, d12 stable) but dropped the three middle rungs. It stated that Signs makes the situation “less stable” without explaining what that means mechanically. A first-time reader would understand the shape of the rule without being able to apply it: they would know that stability determines the die, but not how to read a precarious situation versus a manageable one. They would know that Signs has an escalation effect, but not that the effect is a smaller die on the next roll.

The final version restores all five rungs, named inline with commas, in the SRD’s style, and adds “the next die will likely be smaller” to the Signs result. Those two changes are the difference between a rule that reads clearly and a rule that only reads clearly to someone who already understands it. The rest of the SRD’s compression is earned because the SRD’s core concepts are widely familiar. New concepts have to carry their own explanation, even at the same density.

What the patch is

Three sections. About two hundred words. They sit in the existing SRD without displacing anything.

Situation installs the generative organ, the World Roll, as a structural procedure rather than a GM discretionary tool. Opponents provides the diegetic model that makes Conflict self-contained. Conflict applies the same die and the same result structure to a fight, giving the opposition an independent mechanical voice and giving the fight a natural arc and a natural conclusion.

The patch does not add hit points, initiative order, resource economies, or stress tracks. It does not modify the resolution procedure. It does not contradict the rule that you roll only to avoid risk. What it adds is the other roll: the one that fires when there is no risk, and that lets the world speak without the referee deciding what it says.

I wrote about the Piecepack before, because it crystallized something I care about deeply: the best tabletop tools are the ones that stay open, affordable, and reusable instead of locking you into a single closed ecosystem. DECK# comes from that same place. In a lot of ways, this project is my attempt to carry that philosophy forward and ask how much farther accessibility can go if the starting point is not a custom component set, but a standard deck almost anyone can get their hands on.

That is the real bridge for me from Piecepack to DECK#: not similarity of form, but similarity of intent. I am not trying to recreate Piecepack with different parts. I am trying to continue a way of thinking about game design that puts shared infrastructure ahead of proprietary objects.

Deck# Official Website

Shared structure

What first drew me so strongly to the Piecepack was the sense that it treated components as a common language rather than as the locked content of a single title.

DECK# is built from that same instinct. In the specification, I describe it as an open, standardized substrate that other designers can build on freely, rather than as a universal game that arrives with one prescribed meaning.

That distinction matters to me a lot. A closed game can be beautiful, polished, and deeply satisfying, but it still ends where its box ends. An open system does something different: it invites reuse, reinterpretation, and divergence. It creates continuity between designers who may never collaborate directly, because they are still working on top of the same shared structure.

Substrate, not doctrine

That is why I keep coming back to the idea of substrate. DECK# is not meant to tell anyone what kind of games they should make. The spec defines a graphic vocabulary and a physical annotation structure, but it leaves the meaning of suits, colors, values, and glyphs to the individual game.

That separation between shared structure and local meaning is one of the most important things I wanted to preserve. I think that is also where the Piecepack’s influence becomes clearest. What I admire is not just that it offered a component system, but that it treated openness as part of the design, not as an afterthought.

Open by design

That same principle sits at the center of DECK#. It had to be open in the legal sense, open in the material sense, and open in the design sense, or it would not really be the project I wanted to make.

Materially, that meant starting from the most ordinary object I could reasonably build on: a standard French-suited deck. One of the core goals in the spec is to let people construct and play complex tabletop games using materials available almost anywhere and at minimal cost: one standard deck and two permanent markers, black and red. No proprietary components, no dedicated kit, no dependency on a particular publisher or manufacturing channel.

Why the deck matters

That choice may sound small, but to me it changes the character of the whole system. A standard deck is already culturally familiar, physically compact, easy to carry, easy to replace, and easy to understand at a glance. People know how to shuffle it, hold it, fan it, conceal it, stack it, and place it on a table.

DECK# does not ask players and designers to first acquire a special object before they can begin thinking with the system; it asks them to repurpose one they likely already know. That is the accessibility claim I care about most. Not just affordability, though affordability matters. Not just openness of license, though that matters too.

Accessibility as friction reduction

I mean accessibility as a reduction of friction across the whole chain: sourcing materials, understanding the object, modifying it, explaining it, and sharing it. The more of those steps become ordinary, the more possible it becomes for the system to actually circulate.

So DECK# tries to remove as many barriers as it can. The annotations are additive rather than destructive: they add information to the card without replacing the information already present in the deck. The system is language-agnostic, built from simple geometric glyphs and Arabic numerals rather than text, and those glyphs are intentionally drawable with very few strokes by people without special artistic skill. Even the color vocabulary is constrained to black and red markers for consistency and availability.

A kitchen-table system

I care about these details because they are not neutral. They determine who can plausibly use the system. If a supposedly open framework still depends on polished iconography, custom printing, expensive blanks, or specialized fabrication, then its openness is narrower than it looks.

I wanted DECK# to be something a person could actually make at a kitchen table. That practical reproducibility is, for me, inseparable from the open source stance. The DECK# specification is released under Creative Commons Attribution 4.0, which means people can use, modify, and distribute the specification and the games built on it, including commercially, so long as they provide attribution.

Permission to build

I chose that because I did not want permission to become the bottleneck. If someone wants to build on DECK#, publish with it, teach from it, or push it in a direction I never anticipated, the framework should support that by default.

This is one place where I feel especially strongly that game design should borrow more confidence from other open cultures. In software, in knowledge projects, in maker communities, we already understand that open systems can generate ecosystems rather than just products. Tabletop design sometimes still defaults to thinking in terms of proprietary identity first and reusable infrastructure second. I am much more interested in the opposite order.

Structure still matters

That does not mean the system should be vague. In fact, DECK# tries very hard not to be. The spec is specific about what the inherited deck contributes, what the annotation layer contributes, where marks belong, what remains out of scope, and what every designer must decide when building a game on top of the system. The design guide reinforces that point by treating DECK# as a substrate with its own internal mathematical and structural properties rather than as a loose prompt.

That structure is important because openness alone is not enough to make a system useful. A good open framework needs to give designers leverage. It should reduce some problems, reveal some patterns, and create meaningful constraints that make invention easier rather than harder.

Built-in leverage

One of the things I like most in the design guide is the observation that the glyph distribution is deliberately balanced across the deck, so that a semantic choice assigned to a glyph will tend not to be too rare or too dominant. That is exactly the sort of quiet structural support I want a substrate to provide.

The same goes for the spatial side of DECK#. In the extended profile, cards can become physical spaces on the table, with edge markings defining connectivity and orientation conventions turning a standard card into a tile-like element. The guide describes this as a topological graph emerging card by card rather than as a fixed map, which I think captures the spirit of it very well.

Additive design

Again, the point is not just novelty. The point is that familiar components can be made to do more work without ceasing to be familiar. That additive philosophy is central to the whole project.

DECK# does not reject the standard deck for being too culturally loaded or too historically specific. It accepts the deck as an inherited structure with its own three active axes—value, color, and suit—and then builds around that. I like that because it feels more honest than pretending we are starting from a blank slate. We almost never are.

Working with inheritance

Most game design is really about deciding what to inherit, what to preserve, and what to layer on top. In that sense, DECK# is probably less radical than it first appears. It is not inventing a new object so much as formalizing a way to see an old one differently.

A standard deck already knows how to manage hidden information, ordering, rank, grouping, distribution, and physical handling. DECK# just asks whether that same object can also become a more explicit design substrate through a lightweight, repeatable annotation system.

Continuing the conversation

That question leads directly back to Piecepack for me. What Piecepack modeled so well was the idea that a component system could be bigger than any one title designed on top of it. It could become a durable invitation. DECK# is my attempt to take that invitation seriously in a context where the base materials are even easier to find and the barrier to construction is even lower.

I do not mean that as a claim of superiority. Piecepack and DECK# solve different problems in different ways. Piecepack gives you a family of components purpose-built for combinability. DECK# begins with a preexisting object and tries to unlock more design space from it through structure rather than manufacture. What matters to me is that both approaches take shared possibility seriously.

The designer’s role

That is also why I keep emphasizing that DECK# is not a universal game. It does not prescribe mechanics, genres, meanings, or player counts. It offers a vocabulary, not a doctrine. The spec is very explicit that the semantic mapping belongs to the game designer, not to the substrate.

As a designer, I find that exciting because it shifts the work to the right place. The system can give me a grammar, but it cannot give me taste, judgment, pacing, tension, or coherence. Those still have to be designed. The guide says this directly in a line I keep returning to: DECK# is a substrate; the designer is the author.

Why openness matters

It also helps clarify why openness matters so much here. If the real creative act lives in the mapping, in the rules, in the interplay between structure and interpretation, then the substrate should be as shareable as possible. Closing it off would mean restricting the least personal part of the process while pretending to protect the most personal one.

I would rather do the opposite: keep the foundation open, and let the authored games emerge from it in many directions. I think that is the broader hope behind DECK#. Not simply that people will play the games I make with it, though of course I want that. More importantly, I want the system to be available for uses I would never have predicted.

Designed to travel

I want someone to look at it and decide it is best for tactical play, or narrative play, or solitaire procedures, or educational tools, or something messier that does not fit any obvious label. The moment an open system becomes truly interesting is usually the moment someone else misuses it productively.

That kind of divergence is only possible when the project is designed to survive outside its origin. Open licensing helps with that. Minimal material requirements help with that. Language-agnostic marks help with that. Clear structural rules help with that. All of those choices in DECK# are meant to increase the odds that the system can travel intact while still being interpreted differently.

From influence to permission

So when I think about the road from Piecepack to DECK#, I do not really think in terms of influence as style. I think in terms of influence as permission. Piecepack helped establish, at least for me, that tabletop systems can be conceived as shared, extensible infrastructure rather than sealed artifacts. DECK# grows out of that permission, but tries to push the material side of accessibility further by starting from the humble standard deck and asking it to do more.

That is the project in the simplest terms I can give it. I want a game system that is open enough to be communal, structured enough to be useful, cheap enough to be reproducible, and ordinary enough in its raw materials that it never needs a proprietary product to justify its existence. The more I work on DECK#, the more I realize that all of those goals are really facets of the same one.

I want the foundation to belong, as much as possible, to everyone.

This piece adapts concepts from the Fiction Engine framework and companion essays. Originally developed for TTRPG designers, adapted here for players and GMs navigating table dynamics, consequence, and the hidden architecture of emergent play.

You know the moment. Someone describes what their character does. The dice hit the table. The GM (or the oracle, or the group) responds. And suddenly, the whole table goes quiet for a second, then erupts: “I didn’t plan that, but it makes perfect sense.”

You didn’t script it. The rules didn’t force it. The story just… emerged.

We call it magic. It isn’t. It’s a system. And once you see the conditions that allow it to happen, you’ll never look at your game’s rules, your table’s habits, or your own prep the same way again.

The Secret Isn’t the Dice. It’s the Loop.

Think of a great RPG session like a real conversation. When you’re truly listening, you don’t just wait for your turn to speak. You react to what was just said. You adjust your next line based on the last one. The dialogue builds on itself.

Tabletop play works the same way. The fiction is the conversation. The loop looks like this:

1. A player acts inside the situation.

2. The world responds.

3. The player (or another player) reacts to that response.

4. The fiction accumulates. The chain continues.

For this loop to produce genuine surprise — the kind that feels earned, not random — it needs three things:

• Multiple voices (players, GM, oracle) producing outputs.

• Real sensitivity to what just happened (adjusting plans based on the last roll, not sticking to a pre-written script).

• Accumulated state (a shared memory of what’s already been established, so choices feel weighted and consequences stick).

When those three conditions align, the table isn’t just playing a game. It’s running a feedback network. And that network is where emergent play comes from.

How Thick Rules Break the Spell

Here’s where design choices matter more than we usually admit.

Imagine you’re in the middle of that conversation, and suddenly you have to pause to fill out a form. “Wait, does this count as Acting Under Fire or Defying Danger?”

That’s what happens when procedures get thick. Instead of reacting to the fiction, you’re classifying it. You’re translating “I dive behind the counter as the shotgun fires” into a mechanical category. While your brain does that translation work, the feedback loop pauses. Your attention shifts from the story to the system.

Even worse, thick procedures create what designers call the uniformity effect. Two completely different fictional situations — negotiating with a grieving widow vs. haggling with a corrupt merchant — might both trigger the exact same “Persuade” move with the exact same outcome table. The system delivers consistent dramatic rhythm, but it sands down the specific texture of this situation.

Emergence thrives on specificity. It needs the accumulated weight of prior choices, unique relationships, and irreversible consequences. Procedural taxonomies flatten those into generic inputs. You still get drama. You just stop getting this drama.

Rich Worlds Feed the Loop. Busy Rules Break It.

This doesn’t mean your game should be simple. In fact, the opposite is true.

A sprawling political conspiracy, a character with a decade of scars, a city full of shifting alliances, a treaty signed under duress — that’s complexity in the fiction. And complexity in the fiction feeds the feedback loop. More history means more to react to. More established facts mean more ways the current situation differs from any other. More specificity means more room for genuine surprise.

What breaks emergence isn’t a rich world. It’s complexity in the procedure: the classification steps, the resource trackers, the attentional demands that force you to look away from the table and down at your sheet.

One kind of complexity enriches the conversation. The other interrupts it. Fiction engine designs are built to eliminate the second while welcoming the first.

The Four Ways the Magic Fades (And How to Spot Them)

When emergent play breaks down, it’s rarely because the players “aren’t trying.” It’s because the feedback network is degrading. The framework names four failure modes. Here’s what they look like in plain English:

⚠️ Overhead Accumulation: Too many tiny tracking steps. Every time you have to check a stress track, update a clock, or calculate a modifier, attention bleeds from the fiction to the admin layer. The loop doesn’t break; it starves.

⚠️ Suspended Outcomes: Rolling to see if you can roll to see what happens. When a resolution requires another resolution before the fiction actually changes, the chain breaks mid-cycle. Players stop reacting and start waiting.

⚠️ Authority Drift: The GM or oracle hands you an administrative prompt instead of a fictional event. “Roll Perception to see what you notice” vs. “You hear the floorboards groan behind the door.” The first makes you think about mechanics. The second makes you react to the world.

⚠️ Orientational Divergence: Half the table is playing a tactical puzzle. Half is playing a character drama. You’re not listening to the same conversation. The feedback loops fragment because there’s no shared medium to run through.

These aren’t random bad habits. They’re specific ways the structural conditions for emergence degrade.

How to Protect the Chain at Your Table

You don’t need to rewrite your system to keep the loop running. You just need to recognize what supports it and what starves it.

✅ Treat rules as pipes, not destinations. They should carry the story forward, not stop to be inspected. If you can describe the outcome without consulting a table, you’re probably on the right track.

✅ Keep outcomes moving. Every roll should leave the table with something new to react to. Even a “success” should complicate something you hadn’t anticipated. Even a “failure” should create a new condition, not a dead end.

✅ Use descriptive states over numbers. “Exhausted and bleeding” tells the table how to play. “3 Stress” tells them to check a tracker. Fictional states self-maintain. Administrative states decay the moment you look away.

✅ Stay inside the fiction. When you catch yourself optimizing a mechanic or gaming a resource pool, pause. Ask: “Given what just happened, what would my character do right now?” If the answer doesn’t change based on the last five minutes of play, you’ve slipped out of the loop.

✅ Name the game you’re playing. Emergence requires shared orientation. If your table wants tactical optimization, play for that. If you want emergent story, protect the chain. You can’t run both simultaneously without one starving the other.

The Takeaway

The “magic” of emergent play isn’t a happy accident. It’s the result of keeping the feedback loops intact. Thin mediation isn’t an aesthetic preference for rules-light games. It’s structural engineering for surprise.

When the framework says “outcomes should advance the fiction,” it’s protecting the loop from stalling. When it says “use fictional descriptors,” it’s keeping complexity in the story, not the spreadsheet. When it says “the situation should remain more visible than the system,” it’s stating the exact condition required for emergence to happen.

These aren’t style tips. They’re load-bearing requirements.

Next time you sit down to play, don’t just ask if the rules are fun. Ask if they’re letting the conversation flow. The best stories aren’t processed through procedure. They emerge through it. And now you know how to keep the machinery running.

This question comes up often enough that it deserves a proper answer instead of a reply buried in a comment thread. The shape varies: sometimes it’s a one-line “just combine them,” sometimes it’s a thoughtful case built on real design precedent. But the request underneath is the same: stop publishing two books, publish one.

I’m not going to. Here’s why, in more than a comment box allows.

What’s Actually Already Shared

Start with the part everyone making this suggestion is right about: Loner and Ensemble are not unrelated games that happen to share a publisher. They run on the same engine. Both are built on Tags rather than numeric stats. Both resolve uncertainty through the same six-outcome spread, from Yes, and... through No, and.... Both hold to the same principle that fiction comes first and dice only enter once an outcome is genuinely in doubt. Settings, NPCs, and Adventure Packs cross over with minimal friction; Ensemble even ships with a sidebar explaining exactly how to bring its content into a Loner game, dice conversion included.

So when someone says “these are basically the same system,” they’re not wrong. They are the same system, wearing two different procedures. The disagreement isn’t about whether the engine is shared. It’s about what follows from that.

Where They Diverge, and Why It Isn’t an Accident

Everything that splits Loner from Ensemble traces back to a single variable: how many people are at the table, and whether one of them is a GM.

The dice procedure is the clearest example. Loner resolves an action with a single Chance Die and a single Risk Die, compared directly. Ensemble builds a pool: one Action Die plus however many Chance and Risk dice the fiction justifies, canceled and rolled. Those aren’t the same mechanic in different clothing. A solo player working through a scene alone needs resolution that’s fast and self-contained, because there’s no one else at the table to build tension against. A group wants the pool, because the physical pile of dice makes everyone’s contributions visible at once: gear, frailties, relationships, the GM’s framing of risk. The pool is a social object. A solo session has no use for one.

The Oracle’s interpretation authority diverges the same way. In Loner, the player who asks the question also interprets the answer, because there’s no one else who could. In Ensemble’s Masterless mode, interpretation deliberately falls to someone other than the asker, specifically so no one is grading their own homework. In Classic mode, it’s the GM’s job. This is a structural answer to a structural question about who has standing to say what the dice meant, and that question simply doesn’t exist when you’re playing alone.

Safety tooling follows the same logic. Loner needs one rule: if a scene makes you uncomfortable, change it. You’re the only person whose comfort is in the room. Ensemble needs actual coordination: tools like CATS before play and Script Change during it, because the moment a second person joins the fiction, you no longer get to unilaterally decide what’s fine for the table. Even the Twist Counter’s trigger condition shifts between the two. It builds differently because the pace of complication that feels right for a single protagonist’s story isn’t the pace that feels right for an ensemble cast’s.

None of this is incidental tuning I could shave off to make the two games line up. It’s the design responding correctly to the fact that solo and group play are not the same activity. They have different sources of tension, different authority structures, different needs around safety. A ruleset that pretended otherwise would be worse at both jobs.

What a Combined Chapter Would Actually Look Like

The abstract claim that “the rules would have to branch a lot” is easy to wave away, so it’s worth walking through a specific chapter.

Take Oracle resolution, the most-used page in either book. In a combined text, that page can’t simply explain how to ask a closed question and read the result. It has to first ask the reader to identify their mode, then route them: if you’re solo, roll Chance and Risk directly and interpret it yourself; if you’re in a GM-less group, roll the pool, hand interpretation to someone other than the asker, and resolve any disagreement about who that should be; if you’re in a Classic group, roll the pool and the GM interprets, but only after framing the risk beforehand. Then the Twist Counter has to do the same fork (doubles in solo, every Yes, but... in group), followed by a third fork for safety tooling, because the single-player “change it if it bothers you” rule isn’t sufficient once another person’s comfort is also at stake.

That’s four decision points stacked before a new reader resolves their first roll. Two books avoid this entirely: a Loner reader gets the solo branch as the only branch, written as a straight line. Separating the books pays that routing cost once, at the point of purchase, rather than on every page.

The Ironsworn Comparison, and Where It Breaks

The strongest version of the “combine them” argument points to Ironsworn, which ships solo, co-op, and guided play in a single core book under one license. It’s a fair comparison to raise, and it deserves a real answer.

The analogy breaks at the resolution procedure. In Ironsworn, whether you’re playing alone, with a co-op partner, or with a GM, you roll the same action die against the same two challenge dice. What changes between modes is narration authority: who decides what a result means, who frames the next scene. The underlying math stays the same. That’s a genuinely thin seam, and a single book can hold both sides of it without strain. Most of the rules don’t care which mode you’re in.

Loner and Ensemble don’t have that seam available to them, because what changes between them is the resolution procedure itself, not just who narrates around it. A combined book would stop every reader, in nearly every chapter, with the same question: are you playing solo or group? Dice, Oracle authority, Twist Counter, safety tools. All of it forks. That’s not one source of truth; it’s two books stapled together with a sorting instruction at the top of every page, and it asks new readers to figure out which paragraph is actually theirs before they can use it.

The “one core text” benefit is real when the variation between modes is thin. It doesn’t transfer when the variation is the core loop itself.

What Staying Separate Actually Buys

There’s a practical case for keeping them separate beyond the design argument.

Each book can be tuned to its actual table size without hedging. Loner doesn’t need a paragraph explaining what to do if a second player shows up, because that paragraph already exists in Ensemble, where it belongs. Ensemble doesn’t need to carry solo-specific Oracle interpretation rules that ninety percent of its readers will never touch.

Onboarding gets cleaner too. Someone who only ever intends to play solo should never have to wade through facilitator rotation, GM presentation duties, or group safety coordination to find the one rule that’s actually theirs. And each book evolves on its own schedule. A revision to how Ensemble handles group safety procedures doesn’t force a corresponding edit into Loner just to keep a combined document internally consistent.

The Bridge Already Exists. It’s Just Not a Merger

This isn’t a case of refusing to acknowledge the overlap. The overlap is documented, deliberately and in detail. Ensemble points readers to Loner by name for solo play. Loner carries a full sidebar on importing Ensemble content, dice conversion included, down to how NPC Luck should be rescaled. That’s the right amount of unification: shared DNA, explicit conversion notes between the two expressions of it, and procedures that stay separate because the situations they’re built for are genuinely different.

What I’m not willing to do is sand down that difference to make the books look more similar on a shelf than the games are at a table. Loner and Ensemble are the same idea, built twice, on purpose, for two different things you might want to do with it. That’s not an oversight. It’s the design.

A few days ago, published a scene-by-scene comparison of RISUS and Sparkbound using the same Avatar: The Last Airbender setup: a Fire Nation wagon, three prisoners, one captain, a narrow bridge. If you haven’t read it, go read it first. It’s one of the cleaner demonstrations I’ve seen of how two systems can produce almost identical fictional outcomes while feeling fundamentally different in play.

Wilonas Cave

I Played Avatar: The Last Airbender With Two Different Narrative RPGs — RISUS and Sparkbound

Read more

4 months ago · 3 likes · Wilonas Cave

What struck me reading it was what both systems share: a GM, a group of player characters, and mechanics that engage with the confrontation as a confrontation. RISUS resolves it through competing Cliché dice. Sparkbound resolves it through VP and Sparks. Both are measuring something about the fight.

I wanted to add a third angle.

Loner 4e is a solo RPG. There is no GM, no group, and no combat system in the traditional sense. The Oracle answers questions about the fiction. The Challenge Track measures progress toward a goal. The protagonist’s Luck is what’s at stake, not an enemy’s health.

That changes what you ask, and it changes what the scene is about.

Below is the same prisoner wagon scene, run through Loner 4e. Same characters, same location, same outcome. Different premise entirely.

Here’s the Loner 4e section, built on actual 4e mechanics throughout.

Avatar: The Last Airbender — Loner 4e

On Form

Loner is a solo RPG. One Protagonist, one player, and the Oracle. There is no GM, and everyone else in the story (Katara, Sokka, the captain) exists as fiction.

RISUS and Sparkbound assume a group and a game master. Loner and Sparkbound do not. In Loner, the Oracle is the only other voice in the room, and it only answers questions that matter to the story. That changes the framing before a single die hits the table.

The Protagonist

Aang, Last Airbender

• Concept: Airbending Avatar-in-Training

• Skills: Airbending Mastery; Acrobatic Evasion

• Frailty: Will not use force beyond what is strictly necessary

• Gear: Glider Staff; Air Nomad Robes

• Goal: Free the Earth Kingdom prisoners from Fire Nation custody

• Motive: No one should be imprisoned for where they were born

• Nemesis: The Fire Nation

• Luck: 6

Katara (Protective Waterbender) and Sokka (Strategic Boomerang Fighter) are present in the fiction. Their contributions surface as Oracle questions rather than separate rolls.

The Fire Nation Captain has Tags (Ruthless Firebender Officer, Fire Nation Authority) that tell the Oracle what he would plausibly do. He has no hit points. His defeat is a fictional outcome.

Challenge Track: Free the Prisoners □ □ □ □

Scene: The Prisoner Wagon

Before asking the Oracle anything, frame the scene.

Where: A packed-earth road through scrubland. A Fire Nation wagon moving toward a narrow bridge fifty meters ahead.
Who: The captain on horseback, six soldiers on foot, three prisoners inside the wagon.
What: Aang, Katara, and Sokka intend to intervene before the wagon crosses.

A framing question before the action begins:

“Is there time to act before the wagon reaches the bridge?”

Chance: 5 / Risk: 3 → Chance > Risk → Yes

Time exists. Barely.

1. Aang Scouts Ahead

Before committing to anything, Aang needs a clear picture. He rides a current silently above the road.

Oracle Question: Does Aang’s aerial pass give the group a clear view of the situation without alerting the guards?

Airbending Mastery applies: silent, elevated movement is textbook airbending. Advantage: roll 2 Chance Dice, keep the highest.

Chance: 6 or 3 → keeps 6 / Risk: 4 → Both > 3 → Yes, and...

Aang returns with positions, numbers, and one detail that changes the plan: a drainage ditch running the full length of the approach road, filled with standing water. Enough for Katara to work with.

The “and...” is banked. The scouting did not just inform the plan — it created a resource.

Challenge Track not marked. Scouting is preparation.

2. Sokka Opens the Lock

Sokka identifies the angle. The boomerang leaves his hand.

Oracle Question: Does Sokka’s throw reach the lock before a guard can intercept it?

No Skill or Frailty applies to Aang here. This is Sokka’s action, resolved as an Oracle question about what happens in the world.

Chance: 5 / Risk: 2 → Chance > Risk → Yes

The boomerang curves past the raised spear and strikes the lock. The door swings open. The prisoners are out and running.

Challenge Track: Free the Prisoners ■ □ □ □

3. The Captain Moves

The wagon door is open. The prisoners are running.

Oracle Question: Does the commotion give the prisoners time to reach the bridge before the captain intervenes?

No Advantage or Disadvantage. The captain’s Tag, Ruthless Firebender Officer, means he does not hesitate.

Chance: 2 / Risk: 3 → Risk > Chance, both < 4 → No, but...

He is already pivoting. His fire hand is up. The prisoners will not make the bridge on their own. But: Katara is already moving, pulling water from the drainage ditch Aang located. The “but” is her opening.

Challenge Track unchanged. The situation just became more complex.

4. Katara Holds the Line

A sheet of fire. Katara raises a water wall drawn from the ditch.

Oracle Question: Does Katara’s water wall stop the captain’s fire and force him back?

The scouting result put that water in reach. The “and...” from beat 1 cashes in here: Advantage.

Chance: 6 or 3 → keeps 6 / Risk: 4 → Both > 3 → Yes, and...

The wall stops the fire. Steam fills twenty meters of road in both directions. The captain cannot see through it, cannot aim, cannot project authority he cannot enforce. His grip on the scene breaks.

Challenge Track: Free the Prisoners ■ ■ □ □

5. Aang Intercepts the Soldiers

Two soldiers have broken from formation toward the prisoners. Aang drops into the road between them.

Oracle Question: Does Aang’s air burst redirect the soldiers without injuring them?

Airbending Mastery and Acrobatic Evasion both apply. Advantage.

Chance: 3 or 2 → keeps 3 / Risk: 1 → Both < 4 → Yes, but...

The burst catches both soldiers and drops them into the ditch. They stay down. But: the displaced air scatters the steam just enough for the captain to pick out Aang’s position. He has an angle.

Challenge Track: Free the Prisoners ■ ■ ■ □

6. The Final Moment

One box left. The captain has a line on Aang. The prisoners are not yet over the bridge.

Oracle Question: Does Aang hold his ground long enough for the prisoners to clear before the captain’s fire lands?

Aang’s Frailty — will not use force beyond what is strictly necessary, is directly in play. He will not escalate to protect himself. No Advantage.

Chance: 4 / Risk: 4 → Equal → Yes, but... +1 Twist Counter (Counter: 1)

Katara channels the last of the steam into a spiraling arc. Sokka’s boomerang clips the captain’s wrist a half second before the fire releases. The shot goes wide. The captain stumbles into the ditch.

The soldiers do not stand up.

Challenge Track: Free the Prisoners ■ ■ ■ ■ — Complete.

The Twist Counter sits at 1. The scene resolved cleanly, but something remains unsettled, a soldier who saw too much, a signal fire that may have gone up before the steam reached it. The next scene inherits that pressure.

Result

The prisoners cross the bridge. Sokka takes the wagon. Katara broke the captain’s authority. Aang kept the confrontation nonlethal throughout.

Luck: 6. Aang came through without taking Harm. The Twist Counter at 1 says the world noticed.

Aang makes the captain promise to walk back to the nearest outpost. He agrees. He has no other options.

What Loner Demonstrated

The Oracle asks about fictional outcomes. The questions were never “does Aang hit a soldier” or “does Katara win the duel.” They were “does the water wall stop his fire?” and “does the commotion give the prisoners time to run?” That is a question about what happens in the world, not about character competence measured against a stat.

Skills shift probability, not outcome. Airbending Mastery made the scouting and the soldier intercept Likely to succeed. The dice still rolled. Beat 5 succeeded by the narrowest margin, and the “but” cost Aang his cover.

The captain has no HP. His defeat emerged from accumulated fictional weight: fire stopped, visibility lost, soldiers neutralized, position exposed, angle cut off. The Challenge Track filled. That is his defeat. Nothing subtracted from a number on his sheet.

Companions live in the fiction. Katara and Sokka did not roll their own dice. They appeared inside the Oracle questions: does Sokka’s boomerang reach the lock, does Katara’s water wall hold. The Oracle answered on their behalf. Loner does not pretend the Protagonist is alone. It makes you the sole narrator of a story that is still full of other people.

Luck pressure lands on the Protagonist. The captain never threatened a stat block. He threatened Aang. If that final roll had gone differently, Aang’s Luck would have taken the hit, and Luck at 4 going into the next scene is a different story than Luck at 6. That asymmetry is deliberate. The stakes are personal.

The Twist Counter is the story’s memory. That equal roll at the end did not create a problem in the moment. It registered one. Something is not finished. The Oracle will remember.

The conflict was never the subject. The subject was whether three people could be freed before a wagon crossed a bridge. Loner only ever asked about that.

This piece adapts concepts from the Fiction Engine framework and companion essays. Originally developed for TTRPG designers, adapted here for players and GMs navigating campaign structure, consequence, and the hidden architecture of story memory.

If you’ve played enough tabletop RPGs, you’ve probably experienced two very different kinds of campaigns.

In one, every session feels like it’s building on the last. A burned bridge stays burned. A broken promise shapes how NPCs treat you three months later. When a character dies, it hurts because the table has lived through the history that led there. The game feels alive, heavy with consequence.

In the other, sessions are fun, but they feel interchangeable. Prior events are referenced as flavor, but they don’t actually change what’s possible next week. The table resets to a neutral starting line. A character death is a mechanical loss, not a fictional wound. The story moves forward, but it never really accumulates.

Neither experience is broken. But they’re architecturally different. And understanding the difference will change how you run your games, how you interpret rules, and why certain design choices land with weight while others just feel like noise.

Let’s talk about episodic fiction vs. accumulated fiction. And more importantly, let’s talk about what your rules are secretly doing to protect one or the other.

Two Kinds of Story Memory

Think of RPG fiction like TV storytelling.

📺 Episodic Fiction is your classic monster-of-the-week or anthology show. The premise resets every episode. Last week’s trauma might get a passing reference, but it doesn’t structurally constrain what happens next. The world is a stage, and each session is a self-contained performance. This is how one-shots, dungeon-crawls, and procedural adventures often work.

📖 Accumulated Fiction is a novel or a serialized drama. Chapter 10 depends on Chapter 1. Choices compound. Relationships evolve. When you make a decision, you’re not just picking an option; you’re narrowing the possibility space for everything that follows. The burned bridge cannot be un-burned, not because a rule forbids it, but because the shared fiction now contains a burned bridge, and acting like it’s still there would break the story’s internal logic.

Most games sit somewhere on a spectrum between these two. But the structural difference matters more than we usually admit.

Why Accumulation Creates Real Stakes

When accumulated fiction works, it doesn’t just feel more “serious.” It actually changes how consequence, choice, and surprise function at the table.

🔹 Consequence needs history. A character’s death only carries weight if the table has invested time in who they were, what they were trying to do, and what they’ll leave unfinished. Strip away that accumulated history, and the death is just a stat sheet adjustment. The fiction hasn’t lost anything it built. The weight comes from the past, not the dice.

🔹 Choices feel real when they’re unequal. In an accumulated fiction, prior events make some paths harder and others easier. Trusting a guild isn’t the same choice after they’ve already betrayed you once. Taking the mountain pass isn’t equivalent after the fiction established that one route is watched and the other is unknown. Those asymmetries don’t come from your character sheet. They come from what you’ve already lived through.

🔹 True surprise grows from inside the fiction. The best twists don’t arrive out of nowhere. They contradict something the table has already established as reliable. The trusted ally betrays you. The safe route is ambushed. That kind of surprise requires the fiction to have stored expectations to violate. Reset games can drop in random twists, but they’ll always feel imported from outside. Accumulated games make them feel earned.

What the Rules Are Secretly Protecting

If you’ve read the Fiction Engine framework, you know it insists on a few core guidelines:

• Outcomes should advance or complicate the fiction, not stall it.

• Use fictional descriptors (“exhausted,” “injured,” “on the run”) instead of abstract numbers.

• Cut rules that generate administrative tracking without producing fictional change.

• Make outcomes traceable to prior events.

On the surface, these sound like advice for keeping play fast and efficient. But they’re actually doing something deeper: they’re protecting the game’s memory.

Here’s why:

📌 Fictional facts persist naturally. If the table agrees your character is exhausted, that fact carries into the next scene automatically. Players will roleplay it, NPCs will react to it, and future actions will be filtered through it. A stamina score of 2? That only persists if someone actively tracks it. Fictional states self-maintain. Administrative states decay.

📌 Dead overhead breaks the chain. Every time a rule forces you to pause the fiction to fill out a form, check a table, or calculate a modifier, you insert a gap between cause and effect. Do it enough times, and the invisible chain of “what happened → what it means → what comes next” frays. The story keeps moving, but it stops accumulating.

📌 Traceability is the fingerprint of memory. The framework’s “Traceable” criterion (can you point to earlier events that led here?) isn’t just a referee checklist. It’s a structural requirement. If an outcome can’t be traced back to established fiction, it didn’t emerge from the story. It was dropped in from outside. The difference between a contrivance and a consequence is whether the fiction remembers how you got there.

The Fiction Engine rules aren’t just about efficiency. They’re about preserving the causal tissue that makes a campaign feel alive. When that tissue stays intact, prior events constrain future play. When it frays, you’re just playing a sequence of disconnected episodes.

When Reset Is Actually the Point

None of this means episodic play is inferior. It’s just optimized for different things.

One-shots, anthology formats, and heavily randomized dungeons lean into the reset. They sacrifice long-term consequence for intensity, variety, and clean starting points. You don’t need three months of history to run a tight, self-contained heist. Some tables actively prefer the episodic rhythm.

The key is intentionality. If you’re building a campaign meant to accumulate weight, your rules and rulings should protect memory. If you’re designing a monster-of-the-week format, your rules should facilitate clean resets without guilt.

Knowing which architecture you’re building is half the battle.

The Takeaway: Protect the Chain

Next time you sit down to play or prep a session, ask yourself:

1. Are prior events shaping what’s possible now, or just coloring the dialogue? If they’re only flavor, you’re running episodic. If they’re constraints, you’re running accumulated.

2. Do your outcomes connect back to what came before? Or do they feel like they arrived from a random table? Traceability is your quick-check for structural coherence.

3. Are you building memory, or managing administration? Rules that force tracking over fictional states tend to reset. Rules that embed consequences directly into the fiction tend to accumulate.

The best campaigns don’t happen by accident. They happen when the table treats prior events as structural, not decorative. When the fiction remembers what you did, and refuses to let you pretend otherwise.

Your rules can help that happen. Or they can quietly wipe the slate clean. Choose deliberately.