There’s been Discourse™ of late about the use of GenAI/LLMs in creating RPGs. Not the artwork in an RPG book (that’s a whole ‘nother kettle of fish), but the actual design and development of the game itself: Feeding game text into ChatGPT, Claude, or similar chatbots and asking it to critique, analyze, revise, or otherwise provide feedback.

If you know anything about how LLMs work, it will likely be immediately obvious why this is a terrible idea. But the truth is that a lot of people DON’T know how LLMs work, and that’s increasingly dangerous in a world where we’re drowning in their output.

Michael Crichton described the Gell-Mann amnesia effect: “You open the newspaper to an article on some subject you know well. In Murray’s case, physics. In mine, show business. You read an article and see the journalist has absolutely no understanding of either the facts or the issues. Often the article is so wrong it actually presents the story backwards—reversing cause and effect. (…) In any case, you read with exasperation or amusement the multiple errors in a story—and then turn the page to national or international affairs, and read with renewed interest as if the rest of the newspaper was somehow more accurate about far-off Palestine than it was about the story you just read. You turn the page… and forget what you know.”

Flipping that around, I think analyzing stuff like LLMs in arenas we’re familiar with is valuable because we can more easily see the failures and absurdities. My particular arena of expertise and familiarity — and one I think is likely shared by most of you reading this — is RPGs. So let’s use that familiarity as a lens for looking at LLMs.

Before we start, let’s set a couple baselines.

First, I don’t think AI is completely worthless. I also don’t think it’s the devil. Whether we’re talking about LLMs or some of the other recent technology that’s all getting lumped together as “AI” or “GenAI,” there’s clearly specific ways of using those tools (and also building those tools) which can be ethical and valuable. I don’t think pretending otherwise is particularly useful in trying to prevent the abuse, theft, propaganda, systemic incompetence, and other misuse that’s currently happening.

Second, I am not an expert in LLMs. If you want a truly deep dive into how they work, check out the videos from Welch Labs. (For example, The Moment We Stopped Understanding AI.)

I think the key thing to understand about LLMs, however, is that they are, at their core, word-guessers: They are trained on massive amounts of data to learn, based on a particular pattern of words, what the next most likely word would be. When presented with new input, they can then use the patterns they’ve “learned” to “guess” what the next word or set of words will be.

This is why, for example, LLMs were quite bad at solving math problems: Unless they’d “seen” a specific equation many times in their training data (2 + 2 = 4), the only pattern they could really pick out was X + Y = [some random number].

LLMs are actually still incredibly bad at math, but the “models” we interact with have been tuned to detect when a math problem is being asked (directly or indirectly) and use a separate calculator program to provide the answer. So they look significantly more competent than they used to.

DESIGNING WITH CHATGPT

It’s truly remarkable how far what are fundamentally babble generators can take us. With nothing more than word-guessing, LLMs can create incredible simulacrums of thought. Every generation interprets human intelligence through the lens of modern technology — our brains were full of gears and then they were (steam) engines of thought before becoming computers — but it’s hard not to stare into the abyss of the LLM and wonder how much of our own daily discourse (and even our internal monologue?) is driven by nothing more than pattern-guessing and autonomic response. We see it in the simple stuff:

Ticket Taker: Enjoy the show!

Bob: Thanks! You, too!

But does that sort of thing go deeper than we’ve suspected?

Regardless, there’s one thing missing from LLMs: The ability to form mental models. They can’t read a text, form a mental model of what that text means, and then use that mental model. They can’t observe the world, think about it abstractly, and then describe their conclusions. All they can do is produce a stream of babbled text.

This is why the term “hallucinate” is deceptive when used to describe LLMs’ propensity for spreading misinformation. A “hallucination” would imply that the LLM has formed a false mental model of the world and is now describing that false understanding. But this is not, in fact, what’s happening. What’s happened is that it guessed a word and that word, while matching the patterns found in the model’s training data, did not conform to reality. It’s just words. There is no underlying mental model behind them.

It’s also why asking LLMs to critique anything more complex than the grammar of individual sentences is a waste of time. In order to meaningfully critique something, you have to be able to form a mental model of that thing, have deep and original thoughts about it, and then figure out how to express the conclusions you’ve drawn. An LLM can’t do any of that. At best, it can produce a simulacrum of criticism — a babble that you could perhaps use like a Rorschach blot to free associate your way to a useful insight.

Which brings us to the current movement convinced that they can meaningfully “collaborate” with ChatGPT when designing RPGs. They want to feed in the text of a rulebook and ask the LLM things like:

• Can you build an encounter for 7^th level characters?
• Is this new class I’ve designed balanced with existing classes?
• Are there any problems with the probability of my core mechanic?
• What attribute scores and skills should I have?

In light of what we know, of course, we can immediately see the problem: ChatGPT can’t reliably do basic arithmetic, but it can definitely do complex analysis of mathematical game balance in a system of rules it fundamentally can’t understand?

Someone recently challenged me to “see for myself”: Take an existing RPG, feed the rules into ChatGPT, and ask it to playtest the rules.

This was an interesting challenge, and I’m certainly always open to being proven wrong.

Let’s take a peek at what happened.

CHATGPT DOES A PLAYTEST

I pointed ChatGPT to a copy of Lasers & Feelings, a one-page space opera RPG by John Harper. Then I asked it to generate a playtest transcript of the game.

The choice of game was deliberate: Using a one-page RPG should make it much easier for ChatGPT to analyze and use the entire text. The game is also limited to two mechanics: The action roll and Helping (which adds +1d6 to another character’s action roll). This was the challenge in its easiest form possible.

ChatGPT begins by saying this:

Lasers & Feelings — Playtest Session Transcript

GM: You’re the crew of the interstellar scout ship Raptor, alone while Captain Darcy recovers in a medical pod.

ChatGPT is combining two different sentences from the PDF. Then:

Your mission: find the source of a distress signal coming from an uncharted system.

This is incorrect. In Lasers & Feelings, the GM is supposed to roll on a mission generator. ChatGPT’s mission cannot be generated by the game. But let’s give it a pass here, because there are plenty of human GMs who would also ignore the mission generator and just riff.

ChatGPT now moves on to character creation. It does a decent job of identifying the four steps of character creation, but almost immediately screws up basic math:

Number: 3 (balanced but slightly leaning Lasers)

In Lasers & Feelings, the player picks a number for their character from 2-5. “A high number means you’re better at LASERS (technology; science; cold rationality; calm, precise action). A low number means you’re better at FEELINGS (intuition; diplomacy; seduction; wild, passionate action).”

The number 3, you’ll note, does not “lean Lasers.” It does the exact opposite.

Furthermore, there’s no such thing as “balanced” in Lasers & Feelings. The game specifically doesn’t give you the choice of a midpoint. The whole point is that there’s a tradeoff between Lasers & Feelings. ChatGPT has fundamentally misunderstood the core design principles and theme of the game.

After character creation, ChatGPT proceeds with a transcript of play, and it almost immediately makes a skill check:

Lee: I want to pilot us carefully in. That’s Lasers because it’s technical precision.

Lee rolls 1d6 + 1d6 (prepared) since they have experience navigating rock fields.

This is incorrect. Lee is a Pilot, which means he’s an expert in piloting. If he’s also prepared (as ChatGPT asserts), he should be rolling 1d6 + 1d6 (expert) + 1d6 (prepared).

GM: Your target number is 3 (your stat). Under 3 counts as a success for Lasers.

This is incorrect. A 3 would also count as a success (in addition to generating a Laser Feeling).

ChatGPT cites this blog post as a source for this, but the blog post summarizes the mechanic correctly. ChatGPT just screwed up.

Lee rolls: 2, 5. ✔️ One die under 3 — Success!

GM: You thread us through the asteroids. The distress beacon pings again. What do you do?

According to the rulebooks: “If one die succeeds, you barely manage it. The GM inflicts a complication, harm or cost.”

The GM did not inflict a complication, harm, or cost. ChatGPT has screwed up again.

The “transcript” continues in this vein. Sometimes ChatGPT gets the rules right. It frequently doesn’t, in a wide variety of ways.

You can see the pattern and understand the root cause: ChatGPT can’t actually understand the rules of Lasers & Feelings (in the sense of having the words of the rulebook create a mental model that it can then use independent of the words) and, therefore, cannot truly use them. It can only generate a sophisticated pattern of babble, guessing what the next word of a transcript of Lasers & Feeling game session would look like based on the predictive patterns generated from its training data.

And if it can’t understand the rules well enough to accurately call for a simple action roll, what possible insight could it have into the actual design of the game?

None, of course. Which is why, when I asked it what changes it would make to the game to reinforce the themes, it replied with stuff like:

• The GM should only be allowed to inflict consequences that affect relationships. (Making the game functionally unplayable.)
• Encourage players to switch modes between Feelings and Lasers by inflicting a -1d penalty to the next Feelings roll each time a characters uses Lasers. (This rule would obviously have the exact opposite Plus, it doesn’t recognize that many rolls only use 1d, so how would this rule even work?)

Maybe one of these nonsense ideas it generated will spark an idea for you, but it’s inspiration from babble. Mistaking it for actual critical insight would be a disastrous mistake.

AI GAME MASTERS

Reading ChatGPT’s “transcript” of play, however, it’s nevertheless impressive that it can produce these elements and distinct moments: The distress call isn’t from the rulebook. It’s plucked that out of the ether of its training data. When I mentioned earlier that it’s remarkable how much can be achieved with an ultra-sophisticated babble engine, this is the type of thing I was talking about.

Examples like this have led many to speculate that in the not-too-distant future we’ll see AI game masters redefine what it means to play an RPG. It’s easy to understand the allure: When you want to play your favorite game, you wouldn’t have to find a group or try to get everyone’s schedules to line up. You’d just boot up your virtual GM and start playing instantly. It’s the same appeal that playing a board game solo has.

Plus, most publishers know that the biggest hurdle for a new RPG is that, before anyone can play it, you first have to convince someone to GM it — a role which almost invariably requires greater investment of time, effort, and expertise. If there was a virtual alternative, then more people would be able to start playing. (And that might even end up creating more human GMs for your game.)

There will almost certainly come a day when this dream becomes a reality.

But it’s not likely going to come from simply improving LLM models.

This Lasers & Feelings “transcript” is a good example of why:

• The PCs are following a distress signal.
• It turns out that the distress signal is actually a trap set by bloodythirsty pirates. Two ships attack!
• ChatGPT momentarily forgets that everyone is onboard ships.
• We’re back in ships, but now there’s only one pirate ship.
• And now they’re no longer pirates. They’re lost travelers who are hoping the PCs can help them chart a course home.

It turns out that the GM’s primary responsibility is to create and hold a mental model of the game world in their mind’s eye, which they then describe to the players. This mental model is the canonical reality of the game, and it’s continuously updated — and redescribed by the GM — as a result of the players’ actions.

And what is ChatGPT incapable of doing?

Creating/updating a mental model and using language to describe it.

LLMs can’t handle the fictional continuity of an RPG adventure for the same reason they “hallucinate.” They are not describing their perception of reality. They are guessing words.

The individual moments — maneuvering through an asteroid belt to find the distress signal; performing evasive maneuvers to buy time for negotiations; helping lost travelers find their way home — are all pretty good simulacra. But they are, in fact, an illusion, and the totality of the experience is nothing more than random babble.

And this is fundamental to LLMs as a technology.

Some day this problem will be solved. There are a lot of reasons to believe it will likely happen within our lifetimes. It may even incorporate LLMs as part of a large AI meta-model. But it won’t be the result of throwing ever greater amounts of computer at LLM models. It will require a fundamentally different — and, as yet, unknown — approach to AI.

Much like the word “immersion,” the term “trad RPG” is one that I’ve lately seen confusing conversations more than illuminating them.

The key thing to understand is that “trad RPG” comes from at least three different places, all of them using it to mean different things.

First, storytelling games use it to mean roleplaying games. This arose because STGs – games primarily focused on narrative control mechanics instead of character-associated mechanics – referred to themselves as RPGs, but also knew that they were something different, and wanted (or needed) a term to describe all of the existing RPGs.

Then the Old School Renaissance used “trad RPG” to mean old school play — i.e., the traditional form of play which existed before various “new schools” of play. (With the OSR generally defining the first of these “new schools” as the linear-narrative play that came to domination RPG adventure design post-Dragonlance.)

Most recently, an essay called the Six Cultures of Play defined “trad RPG” as the linear-narrative play championed by Tracy and Laura Hickman in, among other things, Dragonlance. This article was notable for looking at the RPG hobby in terms of cultural behavior rather than mechanical or adventure design. It was filled with severe historical inaccuracies, but, unfortunately, this has not prevented it from being quite popular in online discussions.

And now you can probably spot the problem: You’ve got one group using “trad RPG” to mean pre-Dragonlance play and another group using it to mean post-Dragonlance play. So now “trad RPG” means literally the opposite of itself.

But it gets even more confusing! In response to the use of trad RPG to mean “Dragonlance-style gaming” some segments of the OSR now use it to mean “not the OSR,” which means they also include STGs in “trad” play.

Which means trad RPG now means:

• old school, pre-Dragonlance styles of play
• post-Dragonlance linear-narrative styles of play
• storytelling games
• not storytelling games

So, ultimately, what does “trad RPG” or “traditional RPG” mean?

Nothing.

It’s a term that I’ve used in the past, but one that I generally try to avoid using now. It’s unfortunate because it was a useful term and I haven’t found a reliable replacement, but the reality is that using the term now — no matter how you choose to use it — means that a pretty good chunk of people will end up thinking you meant the exact opposite of whatever you meant. The result, of course, is confusion and needless arguments.

Although, to be fair, what could be more traditional about RPG discussions than that?

In the past few days I’ve been asked several times why I still play Dungeons & Dragons.

After all, I’ve criticized the rules, the adventures, the design, and even the company. So why am I still running and playing this game?

There are a bunch of answers to this question. In fact, part of the answer to the question is that there ARE so many answers to the question.

The most basic answer is the Arneson & Gygax designed one hell of a game,  and with the exception of the most mangled edition of the game, D&D has never drifted so far that its core gameplay has become unrecognizable. There was a time, back in the ’90s, when, so frustrated that my house rules for AD&D had become longer than the core rulebooks themselves, I did give up on D&D for a while, wandering off in pursuit of the One True Way of Gaming. I’ve since learned that there is no One True Way of Gaming, which makes it easier to appreciate D&D for what it is instead of dwelling on everything it isn’t.

Another answer is, in the words of Ryan Dancey, network externality: There are a lot more people who want to play D&D than there are people who want to play any other RPG you’d care to name. It’s just easier to recruit players for D&D and, if I’m looking for a group to play in, I’m statistically more likely to find a group playing D&D. This matters less to me than it used to: After years of running open tables, I’ve recruited a large network of players I can tap into regardless of what system I’m running. (For example, I’ve had little or no difficulty getting players for my new Mothership-based open table.) But you can never completely escape D&D’s gravity well: I’m currently setting up a Heroes of the Borderlands table for my daughter and some of her friends who are desperately interested in playing D&D for the first time. I could try to run something else for them, but I’m not going to tell a bunch of kids to NOT be excited about the game they’re already excited about.

My own familiarity with D&D and its milieu has its advantages. Not only do I like D&D-esque fantasy in general, but my immense experience prepping and running D&D adventures means that I have a very large “bag of stuff” (as Robert Conley puts it) that makes it a lot easier for me to create and improvise stuff while playing D&D.

(This can actually be a double-edged sword if you’re not careful: Yes, I have a finely honed sense of what D&D-eseque fantasy means to me. But that can also be a creative rut that makes it difficult to break out of those tropes. Even with conscious effort, it can be easy to fall back into that rut, specifically because of the comfort and confidence it affords.)

This, of course, would apply to any of the many D&D-adjacent RPGs that are out there. But as a professional RPG designer, it’s also frequently important for me to stay tuned into the industry leader. I’ve had people ask me, for example, why So You Want to Be a Game Master assumes that the reader is most likely interested in running D&D and only pivots to discussing other systems in earnest after the newbie has gotten their feet under them. And the answer, of course, is that this is just the reality of things. D&D remains not only the primary gateway to the RPG hobby, but can still be relied on as a universal touchstone. (To the degree that, when I discuss other RPGs or RPGs in general in my writing or my videos, I not infrequently get comments from people confused because they think D&D is the only RPG in existence.)

(This need to stay in-tune with the current edition of D&D can be another double-edged sword for me. I’d probably have enjoyed several of the D&D 5th Edition games I’ve run over the years more if I’d house rules them more aggressively.)

D&D’s dominance, particularly in the 3E/5E OGL era, also means that D&D is blessed with bountiful source material. For example, when I was working on The Vladaam Affair last week and needed stat blocks for the archmages of the Red Company of Magi, all I needed to do was hit up Volo’s Guide to Monsters. I’ve talked in the past about looting bestiaries and trawling published modules when doing adventure and campaign prep. It’s something that very few other games even come close to achieving.

(This is something that I wish more RPGs would try to provide in a targeted fashion. Not only do I think published adventures are essential for an RPG, but modular components designed to be plugged into prep are vital. For example, Eclipse Phase produced a supplement called NPC File 1 which was just a collection of supporting cast stat blocks. It was literally a game changer for running Eclipse Phase games, because it was so effective at speeding up prep and enabling improvisation when the players jagged off in an unexpected direction. But, unfortunately, they never released a sequel and have never updated the NPC File for 2nd Edition.)

To make a long story short (too late), I really don’t see a time in my future when D&D won’t be part of my life.

Recently, though, I have been giving some long, hard thought to my relationship as a consumer (and reviewer) of Wizard of the Coast’s official D&D books. During the November meeting of the Alexandrian Hangout Club, I was asked if I would be reviewing Forgotten Realms: Adventures in Faerun. This was a particularly interesting release because it included 50+ one-page adventures, which definitely sounds like my jam. The only problem? I flipped through the book at my local game store and saw that, like other recent releases, many of the dungeon maps aren’t properly keyed. And that was only the most obvious failure in basic adventure design. (For a deeper dive, take a peek at Questing Beast’s recent video.)

Although I came somewhat late to D&D 5th Edition (finally getting sucked in by the intriguing promise of Dragon Heist), a combination of reasonably paced releases, financial security, and professional interest made it the first time that I lived out my childhood dream of buying every official D&D book as it came out. But I think that time is coming to an end. I may still be playing and running D&D 2024, but I think this revised edition is also a great jumping off point for me.

Will I be back some day? Probably. Last time I jumped off official D&D as a consumer was back in 2008, and I was gone for a decade. I have high hopes I’ll be back sooner this time. There’s new leadership at Wizards of the Coast, and perhaps we’ll see them right the ship over there.

Or perhaps not.

But either way: Good gaming!

Human teeth wear out. Even if you avoid the deleterious effects of cavities and gum disease, sheer wear and tear will eventually grind your teeth down. So by the end of a few decades of life, it’s not at all unusual for people to start losing some or all of their teeth.

Elves, on the other hand, live for centuries. Perhaps even longer.

So one of three things must be true.

First, elves have human-like teeth, which means that elder elves must surely be toothless. We would expect elven dentures to be quite common and well-crafted, perhaps even becoming enchanted in various ways. We might also expect elven cuisines to feature a great deal of very soft foods, both to preserve their teeth for as long as possible, and then to support their toothless elder class.

Second, elven teeth are quite different from human teeth, possessed of almost adamantine endurance to resist the grind of centuries. It would follow that this “elven ivory” would be considered quite valuable among the unethical. We might also imagine that whatever material gives elven ivory its unique properties might have some unique appearance. Perhaps elven teeth glow silver in the dark?

Third, elves are not limited to a single set of adult teeth like humans, but instead have multiple sets over the course of their lives. Elves might correlate these to various stages of “post-adult” life that are completely alien to human understanding. Perhaps they might refer to their elders as “ninth-toothed” and, to an elf, the adjective “toothy” means wise.

Alternatively, perhaps they are constantly shedding teeth like a shark. Their ephemeral companions on the road might see them carelessly spitting out loose teeth into the fire. Trackers following the trail of an elven hunting party might find the occasional tooth they’ve shed and left in their wake.

UPDATE

Delightful discussion here, on social media, and on the Alexandrian Discord pointed out a fourth option: Elven teeth continually grow from the root, like a badgers. An ironic oversight because these worldbuilding fancies were tangentially prompted by my new elven Pathfinder character filing his incisors into “Gorum fangs” in honor of his now-dead god.

Ever-growing teeth, however, require constant wear. This elven cuisine would likely be filled with unusually hard delicacies, perhaps even flavored sands. Other options would also be pursued: While wood elves might just gnaw on any old stick they find in the forest, more civilized elves would have elegant gnaw-tapers filled with aromatic flavors and placed within elegant holders. With ever-renewing teeth, elven cultures would likely also feature decorative and highly elaborate living scrimshaw. There’d be at least some elven cultures where the elven upper class shows off their wealth by letting their teeth grow into twisted (they’d say “fluted”) spires, and others where having ungnawed or irregularly gnawed teeth would make you a slack-jawed yokel. The phrase “raw-toothed fool” would enter the elven lexicon, and likely also be applied as a slur to humans and their weak little mouth-bones.

But, of course, this is not all:

Thanks particularly to Loris, Kyo618, Arno, and, of course, Mr. Stross. Also check the great ideas in the comments below.

To understand the concept of “Mother, May I?” in RPG design, start by looking at the D&D combat system: This system is based on a complete game structure with predefined actions sorted into an initiative system. The players don’t have to ask the DM if they can attack with their sword or even how the attack with the sword will be resolved — that’s all baked into the system.

Compare that to, say, a PC running a tavern. The player says something like, “I want to get some new types of ale on tap to attract new customers.” Everything about that interaction requires the DM to make rulings: How do they find new types of ale? What types? Does this actually let them attract new customers? How many? What effect do these customers have on their business? The player is effectively going to the DM, hat in hand, and asking them to let them do the thing they want to do.

If we wanted to eliminate Mother, May I?, there are a couple ways we could do that:

1. Try to bake everything into the rules so that every interaction works like combat (with predefined actions, etc.).
2. Create a storytelling game instead, codifying a distribution of narrative control so that a player can, for example, declare the existence of certain brands of ale.

Once broken down like this, we can really begin interrogating the idea that Mother, May I? is an inherently bad thing. Storytelling games are great fun, but they’re not the totality of narrative tabletop games. And the first option is actually impossible without stringently limiting the scope of the game. What makes an RPG special (and distinct from board games like Descent or Gloomhaven) is, in fact, the player’s ability to have their character do anything they can imagine, and it’s the GM’s ability to make flexible and responsive rulings that make that possible.

On the other hand, it can still be valuable to think about the effect that Mother, May I? can have on play. Because, in my experience, players do have a predilection towards structure.

For example, consider running away from a fight. One of the reasons players tend not to do that is because it means exiting a structure of play (combat) where they feel like they have control over their actions and, therefore, the outcome, instead entering a Mother, May I? mode of play in which they’re basically just asking the GM to make a ruling that they won’t be killed.

By contrast, the original 1974 edition of D&D had an explicit Escape/Pursuit structure you can use to resolve fleeing from combat. When I tell players this system exists (and how it works), suddenly they start running away from fights. I put these same players back into a D&D 3E or 5E game and the running away disappears again.

So even though Mother, May I? is the secret sauce that makes it possible for a player to do literally anything they can imagine, it turns out that, paradoxically, selectively adding structure in the right places can actually expand the scope of play.

Even more importantly, it turns out, in my experience, that flexible structures designed to empower GM rulings rather than trying to box the GM completely out of the process are usually the best, reducing unnecessary crunch while simultaneously creating richer play driven by player creativity.

THE GM’S ROLE

If you’re a GM, though, what can you learn from the principles of Mother, May I?

First, it’s always useful to remember that when players propose an action, they are almost always doing so because, if the action is successful, they think the outcome will be fun. So it’s almost always a good idea to Default to Yes:

With that being said, the players are not always right about this. And players also want the thrill of risk and the sweet taste of victory. Plus, the consequences of failure are interesting and vital to a well-rounded and entertaining experience. So don’t fall into the trap of always saying yes. Your judgment is of vital importance at the gaming table.

More advanced GMs can also keep an eye out for complex actions, particularly those that have become or might become a common part of play. When these situations arise, rather thank just making a one-off ruling (e.g., “make a check at Challenging difficulty”), think about how you could instead create a structure that could consistently handle these situations. Even better if you can make the structure player-facing, so that they can make meaningful decisions within the structure.

Remember that these structures don’t have to be terribly complex, and it’s more than all right if they’re a little loose and flexible. For example, consider our earlier example of the PCs running a tavern. A simple structure might look something like this:

• Rate the business in terms of its weekly income.
• Create additional tiers of income (both above and below the current income) – e.g., 10 gp, 25 gp, 50 gp, 100 gp, 250 gp, 500 gp, etc.
• Players who make an investment or improvement to the business can make a skill check to advance the income tier.

When trying to figure out a structure like this, there are a couple of useful rules of thumb:

• Can the players use this structure to proactively take action? (e.g., creating a cool new feature of the tavern to spend investment cash or trying to track down new types of ale to feature on tap)
• Can you hang scenario hooks off of it? (e.g., the PCs learn that the lost recipe for dwarven moon mead might be found within the ruins of Khunbaral).

If one or both of these are true, then your structure will have the capacity to spark creativity and integrate itself into the wider experience of play. (As opposed to mindless dice-rolling in a disconnected minigame.)

Not all such structures need to be player-known, but, as noted above, it’s often the case that making a structure player-known can be the quickest way to get players to engage with it and begin exploring its possibilities.

Over time, you may find one of these structures becoming an increasingly central or frequent part of play. If so, you’ll likely want to add additional details or features in response to what’s happening at the table. Or you might find flaws or shortcomings that need to be fixed. For example, maybe each tier becomes a progress clock instead of a single skill check. Or we could add the concept of a crisis (competition, larceny, natural disaster, recession, supplier shortages, etc.) that could either impose a one-time cost or even reduce the income tier of the business.