Go to Part 1

The first really successful “post-crawl” scenario structure in the RPG industry is the mystery scenario. While it’s easy to find antecedents, it’s really with the publication of Call of Cthulhu in 1981 that the mystery scenario’s popularity exploded.

Thirty years later it can be a little difficult to appreciate how radical Call of Cthulhu was and just how fresh a breath of air it provided. It’s been said that the game “busted gaming out of the dungeon” (which is more literally true if we understand “dungeon” to mean crawl-based scenario structures in general), but its revolution has now become ubiquitous. If you look at published scenarios today, mystery-based scenarios are at least as popular as crawl-based scenarios (and even the crawl-based scenarios often include a large dose of mystery-based play). And outside of D&D the victory is even more complete: Virtually every published scenario is mystery-based.

The degree to which this marketplace dominance reflects actual practice at the gaming table can perhaps be quibbled with. But, either way, it’s actually rather curious once you start reflecting on it: Outside of gaming, mysteries are popular, but they aren’t all-consuming. (And even within gaming you don’t see mysteries consuming the board game or video game markets.)

There are probably several reasons for the popularity of mystery scenarios in RPGs. (For example, an unsolved mystery presents a challenge. Overcoming that challenge – by solving the mystery – provides the kind of clear-cut victory condition that can be achieved with only a few things outside of direct competition or combat.) But I suspect a large part of it is that mystery scenarios present straight-forward game structures that are (a) easy to explain to GMs and (b) intuitively grasped by players.

If we start to look at mystery scenarios using the same structural analysis we used for dungeoncrawls, it’s relatively easy to see this from the player’s point of view:

Default Goal? Solve the mystery.

Default Action? Look for clues.

At this point, however, we have to acknowledge that there’s something different about the mystery scenario structure. While it gives some guidance in answering the player’s question, “What do I do next?” it doesn’t provide a complete answer. Instead, once the players have found their clues, they’ll need to draw a conclusion from those clues. If they don’t draw the necessary conclusion, the mystery scenario will stop providing a default action for them to take.

In other words, mystery scenario structures are more fragile than dungeoncrawls. And this fragility becomes more evident as we shift to the GM’s side of the equation:

Easy to Prep? Unlike a dungeoncrawl, it’s very easy for a neophyte GM to screw up the design of a mystery scenario: If they don’t include a necessary clue, for example, or expect the players to make an intuitive leap they’re not capable of making, the scenario structure will break.

Easy to Run? At a macro-level, mystery scenarios are usually still chunked into smaller, more manageable packets (e.g., each location where clues can be found). But, unlike a basic ‘crawl, these chunks are not as firmly “firewalled”. This generally requires the GM to have a greater understanding of the scenario’s totality and makes a mystery scenario a little more difficult for neophytes to run.

At a micro-level, however, mystery scenarios are generally pretty easy to run if you’re using a skill-based system. (Which, perhaps not coincidentally, were becoming ubiquitous right around the same time mystery scenarios were sky-rocketing to dominance.) This is because skill-based systems can generally be broken down into (a) resolving the success or failure of a physical action or (b) the acquisition of information. Clues, of course, are a type of information, so skill-based systems provide a self-evident method of resolution.

By way of contrast, running a mystery-based scenario in a system like OD&D – which lacks a mechanical basis for determining clue acquisition – can be a lot more challenging. Of course, this difficulty can be overcome if you simply introduce such a mechanic. A simple one would be something like, “If the players say they’re looking around, they automatically find any clues in the area.”

The GUMSHOE system uses a similar mechanic. It says, “If the players say they’re using the right skill, they automatically find any matching clues in the area.” As a game structure, of course, this leaves something to be desired (encouraging, as it does, players to treat their skills like a laundry list in every scene).

But I digress.

FRAGILE SCENARIOS

The point here is that mystery scenarios are a relatively fragile scenario structure: More likely to leave the players stymied; more likely to leave new GMs confused; more difficult for new GMs to design.

What we will discover is that this is often true: Many of the RPG scenario structures we use are remarkably fragile. Like delicate sugar crystals, they are often smashed to smithereens during the course of actual play, forcing GMs to stitch them back together on-the-fly. The robust simplicity offered by the geography of a crawl-based scenario structure – “pick a direction and go” – is difficult to replicate.

Obviously, once you’re aware of the fragility in your game structures, you can start thinking about ways to make them more robust. The Three Clue Rule is a simple example of this: It provides a straight-forward methodology which specifically buttresses the fragile breaking point of a basic mystery scenario.

For now, however, our attention must turn back to the robust demesnes of the ‘crawl structure. For that, we’ll need to journey overland to the wild realms of the hexcrawl…

Go to Part 6: Hexcrawls

Go to Part 1

Tied into the success of the dungeoncrawl is the success of traditional combat systems in roleplaying games. Although individual mechanics may vary, virtually all roleplaying games use a basic game structure for combat derived from D&D: Combat is divided into rounds in which everyone gets to take an action (or actions). Usually combat is further defined by an initiative system of some sort, so that you can easily answer the question, “Who goes next?”

If this sounds like the rigid structure of a boardgame or card game, that’s because it is: Derived from tabletop wargames, the average RPG combat system supplies clear-cut answers to the questions of, “What do I do?” and “How do I do it?”

Or, to break it down in a fashion similar to the dungeoncrawl:

Default Goal: Kill (or incapacitate) your opponents.

Default Action: Hit them.

Easy to Prep: Grab a bunch of monsters from the Monster Manual.

Easy to Run: The combat system breaks the action down into a specific sequence and usually provides a fairly comprehensive method of how each action should be resolved.

This is one of the reasons why so many roleplaying games focus so much mechanical attention on combat: No matter how much the players may be floundering, all you have to do is throw a couple of thugs at them and suddenly everyone at the table knows what to do. It’s a comfortable and easy position to default to.

It should also be fairly easy to see the almost perfect mesh between the macro-level structure of the dungeoncrawl and the micro-level structure of combat: When you pick an exit in the dungeoncrawl and find a room filled with monsters, you seamlessly switch into combat. When the combat has been resolved (and the room emptied of interest), you effortlessly return to the dungeoncrawl scenario structure and pick another exit.

Rinse, wash, repeat.

REWARDS

Combat also shows us how D&D links its systems of reward directly to its default game structures: The dungeoncrawl takes you to monsters, combat lets you defeat the monsters, and defeated monsters reward you with XP and treasure.

This is something which I believe other XP reward systems have generally overlooked: Many games have broken XP away from being a combat reward, but they haven’t reattached the XP award to another concrete game structure.

Whether or not that’s a desirable thing is a completely different discussion; a discussion involving Skinner boxes, the psychology of reward-driven pleasure, mechanical-reinforcement techniques, and a whole mess of other stuff.

But I think it’s an unexplored design space that would be interesting to turn some focus on. Particularly when you consider that (A)D&D has featured non-combat rewards for more than 20 years, and yet people still complain about all the XP in D&D coming from killing monsters.

Go to Part 5: Mysteries

Go to Part 1

The most successful scenario structure in the history of roleplaying games is the traditional dungeoncrawl. In fact, I believe that much of D&D’s success rests on the strength of the traditional dungeoncrawl as a scenario structure. (Notably, it is a structure which has proven extremely effective even when translated into other mediums and executed with completely different mechanics.)

What makes it work?

First, for the player, it provides:

(1) A default goal. Specifically, “find all the treasure”, “kill all the monsters”, or some other variant of “clear the dungeon”. In other words, the structure inherently provides a reason for the player to engage the scenario.

(2) A default action. If a player is standing in a room and there’s nothing interesting to do in the room, then they should pick an exit and go to the next room.

Collectively, these mean that the player always has an answer to the question, “What do I do next?”

Second, for the GM, the dungeoncrawl is:

(1) Easy to prep. In fact, it’s virtually impossible for even a neophyte DM to screw up the design of a dungeoncrawl. What’s he going to do? Forget to draw an exit from the room?

(2) Easy to run. This extends beyond the macro-structure of the dungeoncrawl and begins to depend on the D&D ruleset itself, but, in general, any action proposed by the players within the dungeon will usually have a self-evident method of resolution. The dungeoncrawl also “firewalls” the adventure into discrete chunks (the individual rooms) which can generally be run as small, manageable packets.

Collectively, these mean that even first time DMs can reliably design and run a dungeoncrawl without leaving either (a) their players stymied or (b) themselves confused.

This is huge. Thanks to the dungeoncrawl, D&D can reliably create new DMs in a way that most other RPGs can’t and don’t.

But the dungeoncrawl also has a couple of other key features:

(1) It provides structure, but not a straitjacket. When the players ask themselves, “What do I do now?” the dungeoncrawl provides them with a default answer (“go through an exit”), but doesn’t prohibit them from creating all sorts of other answers for themselves: Fight the goblins. Investigate the arcane runes. Set up a fungal garden. Check for traps. Translate the hieroglyphics. Reverse engineer the construction of dwarven golems. Negotiate with the necromancer. And on and on and on. (It doesn’t even prevent you from leaving the structure entirely: The D&D rules include a multitude of options for bypassing the structure of the dungeon itself.)

(2) Flexibility within the form. The DM can put just about anything into the structure. Each dungeon room – each chunk of content delivered by the scenario structure – is a completely blank canvas.

Having this simple-to-understand, simple-to-design, and simple-to-use scenario structure makes D&D universally accessible in a way that, for example, Transhuman Space isn’t.

Next, let’s shift from the macro-level to the micro-level.

Go to Part 4: Combat

Go to Part 1

All of this can be somewhat difficult to discuss because game structures have become invisible partners in our games. They are generally not spoken of. We don’t even really think about them.

In The Elfish Gene, Mark Barrowcliffe writes: “When D&D is played it sounds like a series of questions and answers.” Similarly, in Apocalypse World, D. Vincent Baker says, “Roleplaying is a conversation. You and the other players go back and forth, talking about these fictional characters in their fictional circumstances doing whatever it is that they do. Like any conversation, you take turns, but it’s not like taking turns, right?”

So as we begin looking at game structures in traditional roleplaying games, what we’re really doing is cracking open this conversation at the game table and looking at what makes it tick. And, just like a formal debate, we’re also looking at how we impose rules onto that conversation and what effect those rules have. (Some of those rules are the actual mechanics of the game, but most of them are actually looser and more universally applicable than that.)

With that in mind, let’s start by imagining ourselves sitting at a hypothetical game table. We want to start playing a traditional roleplaying game. What needs to happen?

(1)   A player needs to propose an action for their character.

(2)   The GM needs to adjudicate that action and provide an outcome.

Once that outcome has been determined, this cycle repeats itself. (The player proposes a new action for their character and the GM then adjudicates that action.) This is the fundamental basis for the conversation at the heart of all roleplaying games.

But, as we established at the beginning of this essay, this is not as simple as it looks. Complications arise once we consider two questions:

(1)   What, exactly, is the action being resolved?

(2)   How, exactly, do we resolve that action?

The first question can be largely thought of as conceptually chunking or breaking actions apart. (Is “exploring the dungeon” a single action or is it made up of many different actions?) Sometimes the GM will need to break a proposed action down into smaller actions; sometimes they’ll need to figure out how to get the player to chunk multiple actions together into a meaningful package. Ideally, of course, both player and GM would get on the same page about what type of “chunk” to propose and resolve. (And this is, of course, where shared game structures come into play.)

The second question is often answered by the mechanics of the game, but not always. (And often those mechanics will only provide a partial answer.)

For a relatively simple example at the micro-level, consider a situation in which Elizabeth the Duchess of Canterlocke is providing a formal introduction for her friend, the Duke of Donalberry. But, in reality, the Duke of Donalberry is a fraud and Elizabeth knows this (she’s trying to smuggle him into some formal affair for reasons that aren’t really important). Is this resolved with a Bluff check? If so, is the check made by the duchess or the duke? Or both? Or should the duchess make a Bluff check while the duke makes a Disguise check? If the duchess fails her Bluff check, does that mean the whole fraud has been discovered? Or does it just alert the target that something funny is going on (thus applying a penalty to the duke’s Disguise check)?

There is no “right” answer here. But which structure you choose will have a significant impact on how this scene plays out at the table.

MICRO vs. MACRO

For the sake of convenience, I’m going to conceptually break down my discussion of game structures into two types:

At the micro-level, game structures usually take the form of GM rulings: As in our example of the Duchess of Canterlocke, the PCs propose an action and the GM determines how that action will be resolved. (In some cases this ruling is straightforward – the players want to do X; there is a rule for X; the GM uses the rule – and in other cases it will require greater innovation.)

At the macro-level, game structures become scenario structures. These larger scaffoldings determine how the players move through complex environments (physical, social, conceptual, or otherwise).

For example, imagine a simple scenario concept: “The PCs go to Castle Osterkark and investigate the rumors of cultist activity.”

A GM could choose to use any number of game structures to flesh out this concept: Is it a crawl from room to room? Is there a network of clues that will take the PCs to various locations throughout the castle? Will the PCs seek out particular NPCs and question them? Is there a timeline of events? Some combination of these techniques? Some other technique entirely?

At both the macro- and micro-level, game structures have a significant impact on what happens at the actual game table.

To help us get a better grasp on what scenario structures look like in practice, let’s take a peek at the grand-daddy of them all: The dungeoncrawl.

Go to Part 3: Dungeoncrawl

One of the most overlooked aspects in the design and play of traditional roleplaying games is the underlying game structure. Or, to put it another way, there are two questions which every game designer and GM must ask themselves:

(1) What do the characters do?

(2) How do the players do it?

These questions might seem deceptively simple, but the answers are complex. And getting the right answers is absolutely critical to having a successful gaming session.

Some of you may already be challenging this. “How difficult can it be? The players tell me what their characters are doing and then we resolve it. What could be easier?”

To demonstrate the oversight taking place here, let me give you a quick example of play:

Player: I want to explore the dungeon.

GM: Okay, make a Dungeoneering check.

Player: I succeed.

GM: Okay, you kill a tribe of goblins and emerge with 546 gp in loot.

Is there anything wrong with that? Not necessarily. But it’s certainly a very different game structure than the traditional D&D dungeoncrawl.

And, of course, that example already assumes that the PCs are fantasy heroes who do things like dungeoncrawling. Given the exact same setting and the exact same game system, they could just as easily be monarchs, dragons, farmers, magical researchers, planar travelers, gods, military masterminds, or any of a dozen other things for whom these dungeoncrawling game structures are irrelevant.

BOARD AND CARD GAMES

The reason game structures become an issue for us is because roleplaying games are functionally open-ended: There is an expectation (and a reasonable one) that the players should be able to say “I want my character to do X” and then we’ll be able to figure out if (a) they’re successful and (b) what happens as a result.

Traditional board and card games don’t run into this problem because their game structure is rigidly defined and limited by the rules: Each time you take a turn in Monopoly or Chess or Arkham Horror there is a precisely defined sequence of actions for you to take. The complexity of this structure can vary quite a bit – in Candyland you simply follow the instructions in order; in Twilight Imperium your decisions would require an incredibly complicated flowchart to model – but the structure is invariable and comprehensive.

Or, to put it another way, boardgames and card games always have an answer to the questions of “What should I be doing now?” and “What happens next?”

ROLEPLAYING GAMES

Consider a hypothetical scenario in which you drop a group of PCs onto a random street corner. Now, take the same group of PCs and drop them into a random room in a dungeon. Why does one group say “I head through the north door” while the other says “I go looking for the local police station”? Why doesn’t the guy in the dungeon say “I go looking for the treasure” and the guy on the street corner say “I take the street on the left”?

Partly, of course, this is a matter of each group of characters having a different set of immediate goals. But it has a lot more to do with habits that have been casually engrained into us through years of playing RPGs.

Another example: Consider the difference between playing D&D and playing the Wrath of Ashardalon boardgame. Both feature similar mechanics, similar settings, and similar character goals. Why is my D&D group likely to spend time examining the walls and investigating arcane circles while my Wrath of Ashardalon group isn’t? Because the game structure is different.

In addition, as our example of dungeon vs. urban scenarios suggests, roleplaying games will often switch game structures. By contrast, computer games usually don’t swap game structures, choosing instead to unify their gameplay: In Elder Scrolls you use the same interface and commands whether you’re exploring a dungeon, traveling through the wilderness, or shopping in town.

On the other hand, games like Final Fantasy VII give you an overland map for travel. And Elder Scrolls V introduced a “fast travel” system that also changes that structure. Meanwhile, at the other extreme, the engine for the original Bard’s Tale was so limited that the town of Skara Brae was a murderville in which citizens attacked like monsters and the gameplay was almost completely indistinguishable from the dungeon at even the micro-level.

In a similar fashion, when I was twelve years old, I tried to run my earliest wilderness adventures as if they were dungeoncrawls: “Okay, you see some trees. What do you do?” “We go north.” “Okay, you go about a hundred feet. There are still trees. What do you do now?”

Use the wrong game structure and you can end up with a really lousy game.

Go to Part 2

GAME STRUCTURES
Part 2: Game Structure Basics
Part 3: Dungeoncrawl
Part 4: Combat
Part 5: Mysteries
Part 6: Hexcrawls
Part 7: Playing With Hexcrawls
Part 8: The Importance of Clean Procedures
Part 9: Archaic Game Structures
Part 10: Incomplete Game Structures
Part 11: Complete Game Structures
Part 12: Using Scenario Structures
Part 13: Custom Structures
Part 14: Scenario Structures for Between the Stars
Part 15: Generic Scenario Structures
Part 16: Player Known and Unknown Scenario Structures

Game Structure: Party Planning
Game Structure: Thinking About Urbancrawls
Game Structure: Tactical Hacking

Addendum: Katanas & Trenchcoats
Addendum: System Matters