Go to Part 1

FINDING THE RIGHT STRUCTURE

You can get a lot of mileage out of node-based scenario design, but it’s not a cure-all. The goal here is not “make everything node-based.” The goal is identifying (or, in some cases, creating) the structure best-suited to running the scenario.

This also means that the more scenario structures you have in your toolkit, the more often you’ll see ways to crack needless railroads into meaningful gameplay or give purpose to meandering scenarios that are desperately trying to empower the players, but don’t know how.

In addition to mysteries and node-based scenario design, structures we’ve discussed here on the Alexandrian include:

• Raids
• Heists
• Hexcrawls
• McGuffin Keep-Away
• Party-Planning
• Sector Crawls
• Festivals
• Urbancrawls
• Tactical Hacking
• Between the Stars (Interstellar Trade)

That might seem like a big reading list. But here’s the thing: Any one of these game structures is the key to unlocking an infinite number of new adventures for you and your players. That’s exciting!

XANDERING DUNGEONS: Another place where I’ll look for opportunities to break up linear design are dungeon maps.

As with other forms of linear design, there are situations in which linear dungeons make sense. But, when it comes to published scenarios, most linear dungeons are just the result of lazy design (or designers who don’t know any better) and I will seize the opportunity to fix them.

I’m not going to belabor the techniques for doing this here, because I wrote a whole article diving deep into this specific topic called Xandering the Dungeon. The short version is that linear dungeons strip strategic play and meaningful decision-making out of dungeon scenarios, resulting in flat, simplistic play that frequently deprotagonizes the PCs.

Here’s one specific tip, though, when doing a dungeon remix: Add windows!

Obviously this doesn’t apply to subterranean labyrinths (usually), but there are plenty of “dungeons” which are just warehouses, slavers’ enclaves, or the mansions of nefarious nobles. A surprising number of these lack windows in published adventures, forcing the PCs to enter them through the one-and-only-door.

In addition to being terrible fire hazards, these buildings don’t really make a lot of sense when we think about how buildings actually work. You can usually fix them pretty quickly by jotting in a few lines indicating the locations of windows, instantly adding a ton of dynamic interest to the scenario by allowing the PCs to choose how they’re going to infiltrate or assault the building. (Or run away when things go poorly at the Fortress of Black Night.)

In addition to the examples given in Xandering the Dungeon, there’s also a practical example of this in the Dungeon of the Dead Three in Remixing Avernus.

PLAY- AND PLAYER-FOCUSED MATERIAL

The other thing I’ll look for in published adventures is material that has no mechanism for bringing it to the table.

This is surprisingly common in published adventures. You’ll read all kinds of nifty stuff, only to realize that there’s no way for the players to ever learn about it. This usually takes the form of cool background material, but sometimes you’ll find vast, Machiavellian struggles being carried out between NPCs without the PCs ever knowing any of it is happening.

It’s top secret, right? So, logically, the PCs shouldn’t know about it!

But if an amazing secret falls in the forest and there’s no one around to see it, does it make a sound anyone care?

Once you identify an element like this you can figure out how to bring that lore into the game:

• Make it a revelation and seed clues that allow the PCs to learn about it.
• Work it into conversations with NPCs. (If it’s an NPC’s dark secret, this conversation may or may not be with them.)
• Have the “secret” actions reverberate throughout the campaign world, creating ripples in the form of rumors, jobs, and other opportunities for the PCs.

There are exceptions to this, and it’s not unusual for an adventure to have a “hidden” elements that exist only to provide context for your rulings.

But if it’s awesome, let the players see it.

On a similar note, make sure that the PCs are the protagonists of the scenario.

Any place where the adventure says, “And then an NPC does something awesome!” think long and hard about how you could redesign that moment so that the PCs are doing the awesome thing.

Any place where the adventure has an NPC tell the PCs what they need to do next, figure out if there’s a way to let the PCs figure that out for themselves.

Failing that, try to make sure that quest-givers are observing the Czege Principle:

When one person is the author of both the character’s adversity and its resolution, play isn’t fun.

Specifically, in this case, structure things so that the quest-giver tells the PCs what the adversity is, but then the players need to figure out how to solve that. This is the difference between, “We need you to get Lord Cleverpants’ technobabble widget!” and “You will sneak into Lord Cleverpants’ castle through the sewers and steal his technobabble widget.” By posing a problem and then letting the players figure out how to solve it, you are giving them agency and the space to actively engage with the scenario.

(The quest-giver who gives them a list of everything they need, instead of parceling them out as sequential quests, is another example of how to do this.)

CONCLUSION

To briefly sum up the heart of a successful remix: Identify cool stuff. Add more cool stuff. Figure out how to reveal that cool stuff to your players and give them the space and structure to play with it.

And have fun!

When using a location-crawl structure (of which a dungeoncrawl is obviously the most well known example), the PCs explore an area room by room. You can see this clearly in the default action of the ‘crawl: If a PC is standing in a room and there’s nothing interesting for them to do in that room, then they should pick an exit and go to the next room.

This structure works well when designing an area that has a high density of interesting stuff in it. Not every room in a dungeon, for example, needs to be filled with interesting stuff, but probably at least half of them do, otherwise the pacing of the scenario collapses as the players robotically churn through empty rooms.

So what can you do when the scenario calls for a crawl-type exploration of a large area with only a few points of interest?

To some extent, of course, we’re talking about a fictional world and you can simply choose to design it differently: Shrink the scope of the area to sync up with the desired points of interest. (Or, alternatively, increase the number of points of interest to match the scope of the area.)

But this is not always desirable or even possible. For example, if the PCs are heading to an abandoned skyscraper in the post-apocalyptic urban wastelands, changing the skyscraper into a duplex is to fundamentally alter the nature of the scenario. And simply filling the skyscraper to the brim with various encounters would be inconsistent with the general premise that Downtown is a thinly populated desert. (It could also quite easily bloat the skyscraper exploration out of proportion to its importance in the scenario.)

Which, of course, brings us to the sector crawl. Like other crawls — location-crawls, hexcrawls, urbancrawls, etc. — it features keyed locations, geographic movement, and an exploration-based default goal. But while superficially similar to a location-crawl, the sector crawl is designed to handle low density areas. It consists of sectors, connections, and encounters.

SECTORS

You’ll start prepping your sector crawl by identifying your sectors. Basically, you’re breaking your scenario’s area into large chunks. Ideally you want these sectors to:

1. Broadly map to the characters’ understanding of the environment.
2. Generally have one point of interest per sector.

In fact, one way of designing a sector crawl is to list your points of interest — hive of mutants, cache of medical supplies, sarcophagus of the seer, etc. — and then figure out what “sector” each point of interest is in. It’s okay to have a few sectors with multiple points of interest, but if you end up with LOTS of sectors like that, you’ll probably want to think about breaking that part of your scenario down into smaller sectors. On the other hand, sectors without a point of interest should generally be very rare (or not used at all).

For example, each floor in our post-apocalyptic skyscraper might be a separate sector. Or perhaps the skyscraper is broken into three sectors – the Lower Floors (1-10), the Middle Floors (11-46), and the Upper Floors (47-69). Or maybe it’s the Ground Floor, Lower Floors, Upper Floors, and Penthouse.

The scale of these sectors, as you can see, will vary depending on the scenario, but you want them to flow naturally from the logic of the campaign world (as opposed to being arbitrary divisions) so that the players can choose how to navigate through the area. (Ideally the players won’t even know you’re running a sector crawl because their navigation decisions will flow naturally from a completely in-character point of view.)

CONNECTIONS

That brings us to the connections between sectors, which is the second thing you’ll want to prepare for your sector crawl.

By default, however, a sector crawl isn’t about the paths between sectors. You can generally access any sector from any other sector, although that navigation might be chokepointed through, for example, a central hub or elevator.

This open access might, in some cases, be only conceptual in nature. If you actually followed every step the PCs take, for example, you might discover that getting from the Lower Floors to the Penthouse is, technically speaking, only possibly by passing through the Middle and Upper Floors. But conceptually, players on the 5^th Floor can say, “Let’s take the elevator up to the Penthouse,” and they can just do that.

This open-access doesn’t mean that the PCs aren’t exploring the location. It just means that the meaningful choices will be ones of sequence and priority rather than geographical navigation.

(If this is making you uneasy, think about how node-based scenario design routinely gives PCs a slate of clues that effectively give them a menu of places to go next. An open-access sector crawl basically does the same thing.)

To support this, as you’re thinking about how your sectors are related to each other, you’ll usually want to avoid sector structures that only make sense after the characters have already explored them. Looking at a post-apocalyptic skyscraper, for example, the players can immediately intuit that it’s made up of various floor and they can go to various floors by going up or down through the skyscraper.

Conversely, if they arrive at the Great Gate of an abandoned dwarven city, they might have no idea what other sectors exist, which will make it difficult for them to make sector selections. If the PCs can’t get an immediate overview of a location when they arrive there, you may need to provide that information in some other form — maps, local guides, skill checks, divine visions, etc.

Tip: The overview might be provided as the point of interest for the Arrival sector. In other words, the PCs arrive at the Great Gate, and when they explore that sector you can tell them about the Grand Promenade, the Lower Galleries, and the seven Dwarven Minarets.

Sometimes, though, you’ll want a mysterious sector for the PCs to discover. These are sectors that are connected to (or hidden within) a specific sector, and can only be discovered or unlocked when the PCs explore that sector.

In most cases, once a mysterious sector has been revealed or accessed, it will become part of the sector crawl’s open access. (Meaning that it can be freely selected like any other sector in the future.) But it’s also possible for some sectors to act as chokepoints: In order to get from a sector on one side of the chokepoint to a sector on the other side of the chokepoint, you do, in fact, have to pass through the chokepoint sector.

Chokepoints can be used to create isolated sectors, but in other configurations you can actually think of them as the connection point between two different sector crawls.

A similar type of sector is the hub sector: If you’re in the hub, you can access any of the sectors connected to the hub. Conversely, you’ll need to pass through the hub to reach any of the sectors connected to it.

Hubs can be a useful way of conceptualizing particularly large sector crawls. Instead of needing to provide an overview of every single sector in the entire complex, you only need to provide an overview of the sectors connected to the hub and any “neighboring” hubs. It’s only when the PCs go to a new hub sector that you’ll need to overview the sectors connected to that hub.

The Great Gate of the dwarven city, for example, can serve as the hub for nearby features within the city (e.g., the Lower Galleries and the Dwarven Minarets), while also providing access to the Grand Promenade which is another hub.

A final key thing to understand is that the more chokepoints, hubs, and mysterious sectors you add to your sector, the more specific you are making the connections between your sectors. The more you increase that specificity, the more your sector crawl will begin blurring into a pointcrawl. This will become even more true if your sectors begin shrinking towards single points of interest.

Which is not necessarily a bad thing, of course, but it’s useful to be aware of what you’re doing.

ENCOUNTERS

A sector crawl doesn’t need encounters, but they’ll usually enhance the scenario and make the area feel like a dynamic, living environment. The specific encounter methods generally resemble those in other ‘crawl structures (like hexcrawls and dungeoncrawls) — they might be random or programmed; keyed to specific sectors or the entire scenario; and so forth.

RUNNING THE SECTOR CRAWL

Running a sector crawl largely consists of going to a new sector or exploring the current sector. Each time the PCs do one of these things, trigger or check for an encounter.

Exploring the sector will turn up:

• Its “identity” (if this was not already known);
• Its point of interest (i.e., the content keyed to the sector); and
• Any secret connections to other sectors (if you’re using mysterious sectors).

If a sector has multiple things to discover, the PCs might find all of them through a single exploration action or find them one at a time (requiring more exploration actions in order to find additional stuff).

And that’s it! Running a sector crawl is actually quite straightforward.

A SECTOR OF ONE

I’ve found it occasionally useful to conceptually think of some locations as a “sector of one” when running them.

For example, in a mystery scenario there might be a house which contains exactly one clue. The players, perhaps conditioned by the ubiquitous location-crawls found in RPG scenario design, might decide to start searching the house one room at a time – essentially treating it like a dungeoncrawl.

Knowing that it’s a rather large house and they’ll spend most of their time not finding stuff that isn’t there, I might push them into a sector crawl and frame their actions appropriately: They say they’re going to search the kitchen, but I simply handle the resolution of the action as searching the whole house (i.e., the house is a single “sector” and they’re performing an exploration action there).

Note: I’m not saying this is the one-true-way of handling this situation. I can think of a half dozen different reasons why you might want to handle the house search as a location-crawl even though there’s only one clue to be found.

This can also be a good thought experiment for how you run a sector crawl, particularly if you’re finding yourself defaulting back into treating the area as a location-crawl. Unless the players are aware of the sector crawl structure (and it’s a structure that doesn’t always lend itself to that), you’ll often be figuring out how to interpret the players’ declared actions within the context of the structure.

For example, they may not explicitly say, “I’m exploring this sector.” But they might say that they’re looking for a place to sleep. Or are searching one section of the sector. Or are looking for medical supplies. All of those can be treated as exploring the sector (and trigger the sector’s discoveries and encounter).

The same thing will be true of navigation, where you’ll need to figure out if certain actions are just moving around inside their current sector or if it’s actually moving to a different sector (and, if so, which one).

UNTESTED: MEGADUNGEON SECTORS

This is something I haven’t had a chance to actually test yet: Converting sections of a megadungeon into sectors.

I’m not talking about designing a megadungeon-like scenario as a sector crawl. (Although you can easily do that, as demonstrated with the example of an abandoned dwarven city above.) What I mean is running a megadungeon as a dungeoncrawl, but when the PCs have “cleared” a particular section of the megadungeon (akin to clearing hexes in a hexcrawl) you convert that section of the megadungeon into a sector.

This would hypothetically allow the players to quickly move through cleared sections of the megadungeon. If you’ve run large dungeons or megadungeons before, you’ve probably already done something akin to this in a purely informal way. (Skipping the boring stuff and getting to the interesting bits, right?) What I find interesting about formalizing this into a sector crawl is that it would still provide a structure for triggering encounters there or even exploring to discover hidden secrets the PCs had previously missed.

It would also give a convenient structure for handling restocking the dungeon – i.e., unclearing the area as new monsters migrate in. For more details on that sort of thing, check out (Re-)Running the Megadungeon.

If you’d like to see a sector crawl in action, there’s one in the upcoming Apeworld on Fire! adventure for the Feng Shui roleplaying game. Designed by Paul Stefko, we used a small sector crawl for a section of the adventure in which the PCs are exploring an abandoned arcanowave laboratory while being hunted by a nanostock demon!

Go to Part 1

As with our previous installment of these capsule reviews, my goal is to just give a very high overview of my thoughts/impressions of each book. These reviews were written as part of my survey of Descent Into Avernus-related material on the Dungeon Masters Guild while working on the Alexandrian Remix of the campaign. Unless otherwise noted, the material has not been playtested.

You may also want to review this Guide to Grades at the Alexandrian. The short version: My general philosophy is that 90% of everything is crap, and crap gets an F. I’m primarily interested in grading the 10% of the pile that’s potentially worth your time. Anything from A+ to C- is, honestly, worth checking out if the material sounds interesting to you. If I give something a D it’s pretty shaky. F, in my opinion, should be avoided entirely.

WARLORDS OF AVERNUS: This supplement caught my attention particularly because I’m hoping to beef up the warlords of Avernus (we have title!), and it delivers quite nicely with four new warlords with very cool concepts supported by a full suite of stat blocks. I would have perhaps liked just a touch more flavor (more fully drawn personalities for the gang members in addition to those given for the warlords themselves), but Rodrigo Kuerten has presented a really great, tight package with high utility. Warlords of Avernus is very much worth $2.

• Grade: B-

BALDUR’S GATE – CITY ENCOUNTERS: When I was running Dragon Heist, I got a huge amount of quality play from Waterdeep: City Encounters (lead design by Will Doyle). That book contains 75 different encounter types, most of which have 3-6 variations, and a random table that splits them up across the different neighborhoods of the city. Borrowing a technique I brainstormed while writing Thinking About Urbancrawls, whenever the PCs went somewhere in the city I would just roll a random encounter for the neighborhood they were going to. It filled the city with life.

So when I saw that there was a Baldur’s Gate: City Encounters book, I snapped it up right quick. Unfortunately, this book (lead design by Justice Arman and Anthony Joyce) is considerably less useful than the Waterdeep version. It includes two sets of encounters: Neighborhood Encounters and Tension Encounters.

The Neighborhood Encounters consist of one encounter for each neighborhood in the city, which is just enough, in my opinion, to not be particularly useful. If I sort of squint at it sideways I can sort of see how you could theoretically have a one-encounter-per-neighborhood structure where the first time PCs enter or pass through a neighborhood you’d use the encounter, which would establish the tone/environment of that neighborhood for the group. (But the encounters here don’t really do that.)

The Tension Encounters are potentially more interesting: They present a five step scale modeling the current level of “tension” in the city and then support this scale with different encounters that can be had at each tension level. How the PCs choose to resolve the encounters can then affect whether the city tips more towards chaos or order!

Conceptually this sounds great, and could provide a great contrapuntal development as the PCs are pursuing their investigation and getting tangled up in Portyr politics. But there are significant problems in practice: First, the scale is supposedly between Order and Chaos, but the actual scale has Pandemonium on one side (with the Cult of the Dead Three performing blood sacrifices in the streets) and Martial Law on the other side (with a corrupted Flaming Fist declaring martial law and instituting pogroms while politicians are assassinated in the streets). It’s thematically incoherent, largely negating the whole point of the exercise.

Second, while promising a system by which the tension meter would change over time, the effort to provide such a system apparently ran aground, with the designers ultimately just throwing their hands up and saying “the DM decides what impact, if any, the encounters in aggregate had on the level of tension in Baldur’s Gate.”

Third, a lot of the tension encounters are kind of nonsensical. Like, there’s one where the PCs are walking down the street when Liara suddenly draws up next to them in a chariot, gives a speech declaring herself Grand Duke of Baldur’s Gate (not how that works), and then offers a ludicrously paltry 250 gp bounty to anybody in the crowd who assassinates any remaining dukes in town.

On that note, the biggest problem I have with the book is that many of the encounters aren’t encounters: They’re scenario hooks to much larger scenarios that the GM would then need to design. (Random encounters spawning unintended scenarios and digressions is a thing that can happen, but they shouldn’t be half-baked into the design.)

The book also includes a neighborhood map of Baldur’s Gate which, for reasons I don’t really understand, doesn’t match any other extant maps of Baldur’s Gate.

• Grade: D

MONSTER LOOT – DESCENT INTO AVERNUS: I snagged Monster Loot: Descent Into Avernus because it seemed to directly address something that I feel is, in fact, generally lacking in the 5^th Edition adventures I’ve seen: Loot. In short, Anne Gregersen supplies a loot listing for every encounter in the campaign.

The book includes two major new mechanics for equipment: First, the option to harvest body parts from foes. Second, broken items that don’t work until you repair them. Unfortunately, it’s largely on the shoals of these two mechanics that the book runs aground.

The problem with the broken mechanic, primarily, is that it’s just massively overused. Virtually every single weapon and piece of armor listed has been broken. On the one hand, this is relatively easy to just ignore. On the other hand, it feels indicative of a certain skittishness in letting the PCs get “good loot” that’s kind of antithetical to what I wanted the book for.

With a book specifically dedicated to customizing loot lists for every NPC, I was really hoping to see some unusual, eclectic, and flavorful stuff. Instead, in almost every case, it’s just “the weapons they’re carrying, the armor they’re wearing, and it’s all broken.” Which, frankly, I don’t really need. That stuff is already in the stat block.

Where Monster Loot: Descent Into Avernus really unleashes, though, are those harvesting rules: You can skin flesh, yank teeth, and cut off tails that do all kinds of crazy stuff. I was actually really interested in this because I find hunter-based play interesting in my open tables and I’m always wishing I had better support for it. But in this specific instance I found the result slightly… distasteful.

The book says that “harvesting body parts, such as hide and flesh, from humanoid creatures is not something this document covers because we don’t encourage adventurers to tear into the bodies of people.” But it means that in the most literal sense of the humanoid monster type. The book happily provides you the details on skinning angels and all kinds of intelligent creatures (including bipedal intelligent creatures).

At just $2.95 I flirted with giving this one a D, but ultimately I think I’m not going to bother having this at the table when I run the campaign. So, unfortunately…

• Grade: F

THE HELLRIDERS’ KEEP: This supplement adds a new location to Elturel. Conceptually it’s great. Not only does making Elturel a richer location for the PCs to explore make a lot of sense, but Carter VanHuss very astutely notes that the published adventure doesn’t cleanly clue the PCs into the true history of the Hellriders and designs this scenario to remedy that. The descriptions of the environment are really good, with lots of little details that are not only specific, but also packed full of lore. Exploring this space will immersively draw players into the world.

Unfortunately, the book does get a little hamstrung by a couple of structural issues. First, the hook is just another, “NPC tells the PCs to go some place, the PCs go there” affair. To some extent, I can see how his hands were tied by the published campaign itself, but it feels like with a little extra effort several hooks could have been more organically woven into the campaign to make PCs aware of the Hellriders’ Keep.

The more significant problem is the lack of a map: The entire structure of the adventure is exploring the castle, but the two maps in the product are instead battlemaps. Individual areas are keyed and an effort is made to describe how they relate to each other, but without a map it’s all needlessly confusing.

Despite this, I think it’s worth grabbing a copy of this if you’re going to run Descent Into Avernus (even if you will end up needing to draw a map).

• Grade: C

MONSTER HUNTS – AVERNUS: This book promised to be a bunch of plug ‘n play side quests for use with Descent Into Avernus. I thought this would be a slam dunk in terms of usefulness, providing all kinds of awesome content for fleshing out a hexcrawl of Avernus.

Unfortunately, not one of the one-page scenarios is actually set in Avernus. In this case, “for use with Descent Into Avernus” means that it uses the stat blocks from the appendices of Descent Into Avernus.

Ignoring the disappointing bait ‘n switch (which renders the book completely unusable for what I wanted it for), the scenarios themselves are also very poorly designed (so that I wouldn’t want to use them for anything): For example, most of the dungeon maps, instead of being keyed, are described in rambling, unfocused paragraphs. The text is frequently filled with prima facie nonsense (like a claim in the first scenario that it will take PCs forty minutes to walk two city blocks). And it’s almost impressive how many times they try to force a railroad on PCs even when they’re just exploring a simple dungeon.

The book also promises an “easy to use hunting system,” but I can find nothing of the sort. Instead, the majority of the scenarios lead off with some form of “make this skill check to find tracks or skip the rest of this adventure.”

• Grade: F

HELLTUREL: James Introcaso has really hit the nail on the head with Hellturel. This 32-page supplement presents four new locations for Elturel, nicely fleshing out the city for PCs who want to explore it. Not only are the locations well-designed, they are connected using node-based scenario design so that exploring one location will provide leads pointing to the others.

The only thing I would have liked to have seen would be some guidance for how clues could be added to the locations described in Descent Into Avernus in order to also link them to the locations in Hellturel. That creates a little bit of extra lifting. There are also some minor continuity glitches (for example, the first location says the Order of the Gauntlet has moved to the second location, but at the second location there’s only one member of the Order of the Gauntlet and, as far as I can tell, no indication of what happened to the rest of them) that probably needs to be cleaned up.

But, as I say, really good stuff. Recommended.

• Grade: B-

More DMs Guild Capsule Reviews – Go to the Avernus Remix

“Most of the campaigns I’ve really enjoyed have been in systems I didn’t like.”

“A great GM can take any RPG and run a good game.”

“I just want a system that gets out of the way when I’m playing.”

What I think these players are discovering is that most RPG systems don’t actually carry a lot of weight, and are largely indistinguishable from each other in terms of the type of weight they carry.

In theory, as we’ve discussed, there’s really nothing an RPG system can do for you that you can’t do without it. There’s no reason that we can’t all sit around a table, talk about what our characters do, and, without any mechanics at all, produce the sort of improvised radio drama which any RPG basically boils down to.

The function of any RPG, therefore, is to provide mechanical structures that will support and enhance specific types of play. (Support takes the form of neutral resolution, efficiency, replicability, consistency, etc.) If you look at the earliest RPGs this can be really clear, because those games were more modular. Since the early ’80s, however, RPGs pretty much all feature some form of universal resolution mechanic, which gives the illusion that all activities are mechanically supported. But in reality, that “support” only provides the most basic function of neutral resolution, while leaving all the meaningful heavy lifting to the GM and the players.

To understand what I mean by that, consider a game which says: “Here are a half dozen fighting-related skills (Melee Weapons, Brawling, Shooting, Dodging, Parrying, Armor Use) and here are some rules for making skill checks.”

If you got into a fight in that game, how would you resolve it?

We’ve all been conditioned to expect a combat system in our RPGs. But what if your RPG didn’t have a combat system? It would be up to the GM and the players to figure out how to use those skills to resolve the fight. They’d be left with the heavy lifting.

And when it comes to the vast majority of RPGs, that’s largely what you have: Skill resolution and a combat system. (Science fiction games tend to pick up a couple of additional systems for hacking, starship combat, and the like. Horror games often have some form of Sanity/Terror mechanic derived from Call of Cthulhu.)

So when it comes to anything other than combat — heists, mercantile trading, exploration, investigation, con artistry, etc. — most RPGs leave you to do the heavy lifting again: Here are some skills. Figure it out.

Furthermore, from a utilitarian point of view, these resolution+combat systems are all largely interchangeable in terms of the gameplay they’re supporting. They’re all carrying the same weight, and they’re leaving the same things (everything else) on your shoulders. Which is not to say that there aren’t meaningful differences, it’s just that they’re the equivalent of changing the decor in your house, not rearranging the floorplan: What dice do you like using? What skill list do you prefer for a particular type of game? How much detail do you like in your skill resolution and/or combat? And so forth.

SYSTEM MATTERS

What I’m saying is that system matters. But when it comes to mainstream RPGs, this truth is obfuscated because their systems all matter in exactly the same way. And this is problematic because it has created a blindspot; and that blindspot is resulting in bad game design. It’s making RPGs less accessible to new players and more difficult for existing players.

I’ve asked you to ponder the hypothetical scenario of taking your favorite RPG and removing the combat system from it. Now let’s consider the example of a structure which actually HAS been ripped out of game… although you may not have noticed that it happened.

From page 8 to page 12 of The Underworld & Wilderness Adventures, Arneson & Gygax spelled out a very specific procedure for running dungeons in the original edition of Dungeons & Dragons. It boils down to:

1. You can move a distance based on your speed and encumbrance per turn.

2. Non-movement activities also take up a turn or some fraction of a turn. For example:

• ESPing takes 1/4 turn.
• Searching a 10′ section of wall takes 1 turn. (Secret passages found 2 in 6 by men, dwarves, or hobbits; 4 in 6 by elves.

3. 1 turn in 6 must be spent resting. If a flight/pursuit has taken place, you must rest for 2 turns.

4. Wandering Monsters: 1 in 6 chance each turn. (Tables provided.)

5. Monsters: When encountered, roll 2d6 to determine reaction (2-5 negative, 6-8 uncertain, 9-12 positive).

• Sighted: 2d4 x 10 feet.
• Surprise: 2 in 6 chance. 25% chance that character drops a held item. Sighted at 1d3 x 10 feet instead.
• Avoiding: If lead of 90 feet established, monster will stop pursuing. If PCs turn a corner, 2 in 6 chance they keep pursuing. If PCs go through secret door, 1 in 6 chance they keep pursuing. Burning oil deters many monsters from pursuing. Dropping edible items has a chance of distracting intelligent (10%), semi-intelligent (50%), or non-intelligent (90%) monsters so they stop pursuing. Dropping treasure also has a chance of distracting intelligent pursuers (90%), semi-intelligent (50%), or non-intelligent (10%) monsters.

6. Other activities:

• Doors must be forced open (2 in 6 chance; 1 in 6 for lighter characters). Up to three characters can force a door simultaneously, but forcing a door means you can’t immediately react to what’s on the other side. Doors automatically shut. You can wedge doors open with spikes, but there’s a 2 in 6 chance the wedge will slip while you’re gone.
• Traps are sprung 2 in 6.
• Listening at doors gives you a 1 in 6 (humans) or 2 in 6 (elves, dwarves, hobbits) of detecting sound. Undead do not make sound.

I’ve said this before, but if you’ve never actually run a classic megadungeon using this procedure — and I mean strictly observing this procedure — then I strongly encourage you to do so for a couple of sessions. I’m not saying you’ll necessarily love it (everyone has different tastes), but it’s a mind-opening experience that will teach you a lot about the importance of game structures and why system matters.

The other interesting thing here is that Arneson & Gygax pair this very specific procedure with very specific guidance on exactly what the DM is supposed to prep when creating a dungeon on pages 3 thru 8 of the same pamphlet. (These two things are conjoined: They can tell you exactly what to prep because they’re also telling you exactly how to use it.) Take these two things plus a combat system for dealing with hostile monster and, if you’re a first time GM, you can follow these instructions and run a successful game. It’s a simple, step-by-step guide.

“Now wait a minute,” you might be saying. “You said this procedure had been ripped out of D&D. What are you talking about? There’s still dungeon crawling in D&D!”

… but is there?

THE SLOW LOSS OF STRUCTURE

Many of the rules I describe above have passed down from one edition to the next and can still be found, in one form or another, in the game as it exists today. But if you actually sit down and look at the progression of Dungeon Master’s Guides, you’ll discover that starting with 2nd Edition the actual procedure began to wither away and eventually vanished entirely with 4th Edition.

The guidance on how to prep a dungeon has proven to have a little more endurance, but it, too, has atrophied. The 5th Edition core rulebooks, for just one example, don’t actually tell you how to key a dungeon map. (And although they have several example maps, none of them actually feature a key.)

One of the nifty things about a strong, robust scenario structure like dungeon crawling is that with a fairly mild amount of fiddling you can move it from one system to another. This is partly because most RPGs are built on the model of D&D, but it’s also because scenario structures in RPGs tend to be closely rooted to the fictional state of the game world.

This is, in fact, why you probably didn’t notice that 5th Edition D&D doesn’t actually have dungeon crawling in it any more: You’re familiar with the structure of dungeon crawling, and you unconsciously transferred it to the new edition the same way that you’ve most likely transferred it to other games lacking a dungeon crawling structure in the past. In fact, I’m willing to guess that removing dungeon crawling from 5th Edition was not, in fact, a conscious decision on the part of the designers: They learned how to run a dungeoncrawl decades ago and, like you, have been unconsciously transferring that structure from one game to another ever since.

Where this becomes a problem, however, are all the new players who don’t know how to run a dungeoncrawl.

Most people enter the hobby through D&D. And D&D used to reliably teach every new DM two very important procedures:

1. How to run a dungeon crawl

2. How to run combat

And using just those two procedures (easily genericizing the dungeon crawling procedure to handle any form of location-crawl), a GM can get a lot of mileage. In fact, I would argue that most of the RPG industry is built on just these two structures, and that most GMs really only know how to use these two structures plus railroading.

So what happens when D&D stops teaching new DMs how to run a dungeoncrawl?

It means that GMs are now reliant entirely on railroading and combat.

And that’s not good for the hobby.

THE BLINDSPOT

If you need another example of what this looks like in practice, check out The Lost Mine of Phandelver, the scenario that comes with the D&D 5th Edition Starter Set.  It’s a fascinating look at how this really is a blindspot for the 5th Edition designers, because The Lost Mine of Phandelver includes a lot of GM advice. They tell you that the GM needs to:

• Referee
• Narrate
• Play the monsters

They give lots of solid, basic advice like:

• When in doubt, make it up
• It’s not a competition
• It’s a shared story
• Be consistent
• Make sure everyone is involved
• Be fair
• Pay attention

There’s a detailed guide on how to make rulings. They tell you how to set up an adventure hook.

Then the adventure starts and they tell you:

• This is boxed text, you should read it.
• Here is a list of specific things you should do; including getting a marching order so that you know where they’re positioned when the goblins ambush them.
• When the goblins ambush them, they give the DM a step-by-step guide for how combat should start and what they should be doing while running the combat.
• They lay out several specific ways that the PCs can track the goblins back to their lair, and walk the DM through resolving each of them.

And then you get to the goblins’ lair and…  nothing.

I mean, they do an absolutely fantastic job presenting the dungeon:

• General Features
• What the Goblins Know (always love this)
• Keyed map
• And, of course, the key entries themselves describing each room

But the step-by-step instructions for how you’re actually supposed to use this material? It simply… stops. The designers clearly expect, almost certainly without actually consciously thinking about it, that how you run a dungeon is so obvious that even people who need to be explicitly told that they should read the boxed text out loud don’t need to be told how to run a dungeon.

And because they believe it’s obvious, they don’t include it in the game.  And because they don’t include it in the game, new DMs don’t learn it. And, as a result, it stops being obvious.

(To be perfectly clear here: I’m not saying that you need the exact structure for dungeon crawling found in OD&D. That would be silly. But the core, fundamental structure of a location-crawl is not only an essential component for D&D; it’s really fundamental to virtually ALL roleplaying games.)

THE BLINDSPOT PARADOX

Paradoxically, this blindspot not only strips structure from RPGs by removing those structures; it also strips structure from RPGs by blindly forcing structures.

It is very common for a table of RPG players to have a sort of preconceived concept of what functions an RPG is supposed to be fulfilling, and when they encounter a new system they frequently just default back to the sort of “meta-RPG” they never really stop playing. This is encouraged by the fact that the RPG hobby is permeated by the same meme that rules are disposable, with statements like:

• “You should just fudge the results!”
• “Ignore the rules if you need to!”

A widespread culture of kitbashing, of course, is not inherently problematic. It’s a rich and important tradition in the RPG hobby. But it does get a little weird when people start radically houseruling a system before they’ve even played it… often to make it look just like every other RPG they’ve played. (For example, I had a discussion with a guy who said he didn’t enjoy playing Numenera: Before play he’d decided he didn’t like the point spend mechanic for resolving skill checks; didn’t like XP spends for effect; and didn’t like GM intrusions so he didn’t use any of those mechanics. He also radically revamped how the central Effort mechanic works in the game. Nothing inherently wrong with doing any of that, but he never actually played Numenera.)

As a game designer, I actually find it incredibly difficult to get meaningful playtest feedback from RPG players because, by and large, none of them are actually playing the game.

And these memes get even weirder when you encounter them in game designers themselves: People who are ostensibly designing robust rules for other people to use, but in whom the response to “just fudge around it” has become so ingrained that they do it while playtesting their own games instead of recognizing mechanical failures and structural shortcomings and figuring out how to fix them.

EXCEPTIONS TO THE BLINDSPOT

To circle all the way back around here: System Matters. But due to the longstanding blindspot when it comes to game structures and scenario structures in RPGs, we’ve stunted the growth of RPGs. Most of our RPGs are basically the same game, and they shouldn’t be. The RPG medium should be as rich and varied in the games it supports as, for example, the board game industry is. Instead, we have the equivalent of an industry where every board game plays just like Settlers of Catan.

Now there are exceptions to this blindspot when it comes to scenario structures. Partly influenced by storytelling games (which often feature very rigid scenario structures), we’ve been seeing an increasing number of RPGs beginning to incorporate at least partial scenario structures.

Blades in the Dark, for example, has a crew system that supports developing a criminal gang over time. Ars Magica does something similar for a covenant of wizards. Reign provides a generic cap system for managing player-run organizations in competition with other organizations.

Technoir features a plot-mapping scenario structure that’s tied into character creation and noir-driven mechanics.

For Infinity I designed the Psywar system to provide support for complex social challenges (con artistry, social investigation, etc.).

You can also, of course, visit some of the game structures I’ve explored here on the Alexandrian: Party Planning, Tactical Hacking, Urbancrawls. The ongoing Scenario Structure Challenge series will continue to explore these ideas.

Go to Game Structures

Not to be confused with hexcrawling, hex-clearing is the process by which monsters and other hostile forces were cleared out of a hex in preparation for a stronghold to be constructed. Clearing a hex was the first step towards bringing civilization to an uncivilized portion of the world. It was also the transitional point between the low-level activities of monster slaying and the high-level activities of realms management. It is one of the oldest game structures in D&D, yet I feel comfortable saying that probably 99% of all current D&D players have never done it.

In pursuit of a tangentially-related project, I decided to do a brief survey of the extant hex-clearing procedures in old school D&D. I offer them here in the thought that they might be of use to a wider audience.

OD&D HEX-CLEARING

Hex Scale: 5 miles

1. Referee rolls a die to determine if there is a monster encountered.
2. If encountered monster is defeated or if no monster is encountered, the hex is cleared.
3. Territory up to 20 miles distant from an inhabited stronghold may be kept clear of monsters once cleared.

AD&D HEX-CLEARING

Hex Scale: 1 mile / 30 miles

CLEARING HEXES

1. Make wandering monster check.
2. If encountered monster is defeated or if no monster is encountered, the hex is cleared.

Once cleared, hexes will remain cleared, except:

1. Once per day, check to see if a monster has wandered into an uncleared border hex.
2. Once per week, check to see if one of these monsters has wandered into the cleared territory.

Patrols: If regular (1/week) patrols from a stronghold are made through a cleared territory, the check to see if a monster has wandered into a border hex is made only once per week.

CONSTRUCTING THE STRONGHOLD

1. Must map and clear the central hex (location of stronghold) and six surrounding hexes.
2. Unless 7 hexes are actively patrolled, there is a 1 in 20 chance per day that a monster will enter the area.

GYGAXIAN VAGUERY – PATROLS

Because Gygax was objectively terrible at writing rulebooks, the rules above are actually incomplete. They overlap with a different set of incomplete rules which directly contradict the first set of rules. If you use this second set of rules, a cleared hex that is being patrolled should be handled in this way:

1. Once per week, check on the Uinhabited/Wilderness encounter table to see if a monster enters the cleared territory.
2. Once per week, also check on the Inhabited table. Or, if there is a road, check three times on the Inhabited encounter table.

Zone of Civilization: If a territory is cleared to a 30 mile radius [should probably be 30 mile diameter, filling the large hex that the stronghold is at the center of], make ONLY the second type of checks, but ignore all unfavorable checks except once per month.

Reversion to Wilderness: If patrols are not kept up, the territory automatically reverts to wilderness status. “Unless the lands around it are all inhabited and patrolled” in which case “all of the unsavory monsters from the surrounding territory will come to make it a haven for themselves.” [So it won’t revert to wilderness, it will just really revert to wilderness.]

RULES CYCLOPEDIA – HEX-CLEARING

Hex Scale: 8 miles / 24 miles

Clearing the Hex: You just… do it. “An area is considered clear when all significant monsters in the area have been killed, driven out, or persuaded (through bribery, threats, persuasion, or mutual-defense agreements) to leave the PC’s subjects alone.” There are no further guidelines.

Constructing the Stronghold: Clear the 8-mile hex in which the stronghold is being built.

Patrols: Cleared areas automatically remain free of monsters as long as they are patrolled.

• Patrols can range 24 miles from a stronghold in clear terrain.
• Jungles, swamps, and mountains require a garrison every 8 miles.

There are more detailed rules for dominion management, but they don’t really pertain to hex clearing.

EXPERT SET VARIATIONS

• Hex scale is not clearly defined. (Isle of Dread, the sample adventure included in the set, uses 24 mile and 6 mile hexes.)
• Patrol ranges are limited to 18 miles and 6 miles (instead of 24 miles and 8 miles).
• The 18 mile limit of patrols matches the 18 miles an encumbered character can travel on foot in a day. The Rules Cyclopedia oddly maintains the same rule for determining overland movement rates (divide by 5 to determine the number of miles a character can travel over clear terrain per day, and therefore 90’ divided by 5 = 18 miles per day), but the Traveling Rates By Terrain table doesn’t follow that rule and instead uses values calculated to divide evenly into hexes (so an encumbered character only travels 12 miles per day in clear terrain).

JUDGES GUILD – HEX-CLEARING

Hex Scale: 5 miles

As I’ve mentioned in the past, Judges Guilds’ hexcrawl procedures and management had a major impact on the game. Virtually all of OD&D’s hexcrawling procedures, for example, were abandoned by AD&D in favor of systems clearly drawing from Judges Guild material. This was somewhat less true when it comes to hex-clearing, but I thought reviewing the material from the Ready Ref sheets might be useful. In this case, it largely was not:

Constructing the Stronghold: Clear 4 hexes radiating from the stronghold’s hex.

Patrols: Automatically keep hexes clear of monsters, except for mountains, swamps, and dense woods.

Back to Reactions to OD&D