Sandboxing tips: The tentpole dungeon

I recently changed my mind on the big tentpole dungeon.

A dominant (if not necessarily the dominant) playstyle in the OSR is the wide-open sandbox. Players usually start as nobodies exploring a frontier beyond which is an unknown, depopulated wasteland, usually brimming with adventure sites and treasure. Taking a cue from Keep on the Borderlands, a lot of these homebrew campaigns seem to feature a dungeon that stands out from the rest. It's usually bigger than other adventure sites, and accordingly features hallmarks of megadungeon play, such as faction interaction (although it doesn't strictly have to be a megadungeon). The idea is that, while the party might discover other adventure sites, this is the one they keep coming back to and chipping away at little by little, as a mainstay of the campaign. This is the tentpole dungeon.

I've been aware of the concept of the tentpole dungeon for a while, but I never really got on board. It felt more organic to me to have lots of adventure sites of varying sizes scattered around, rather than one big dungeon the players were "supposed" to explore, and a host of little ones to break it up. Generally speaking, I'm fairly allergic to anything that has a whiff of the GM telling the players they're "supposed" to do anything. (My players make fun of me for this - every time they find a map or a helpful NPC they start saying things like "Looks like we've found the main quest lads" and "All aboard the railroad!") But two things changed my mind on the tentpole dungeon.

First, I've been thinking a lot about expectations recently. All games run on expectations of play; you expect the players to do certain things, and they expect certain things to happen if they do. Those things can be more broadly or narrowly defined depending on the type of game, but they're always there, even in a wide-open, sandbox style. More than that, it's vitally important to get clear on the expectations you have, and the expectations the players should have going into the game, in order for the game to run smoothly.

With that understanding, it makes a lot of sense to offer the players the tentpole dungeon as a sort of alternative game mode to overworld sandbox play. In one mode, they get to explore and expand their knowledge of a dungeon they know for definite is there, and where they've already established something of a foothold. In another, they're more on their own, exploring the wilderness looking for adventure, travelling from place to place to fulfil their own goals, and facing challenges accordingly. The advantage of having both (and making it clear to the players that both are options) is that they can choose what to do depending on how they feel - essentially, choosing which game they want to play that session. For instance, my players have largely involved themselves in political wrangling and grand strategy in the overworld, and retreat to the dungeon when they get bored of playing with armies and cities, and want some more personal tomb-robbing adventure. As I understand it, this mirrors the original Blackmoor style of play, where players would switch between managing armies in the ongoing war(s), and delving into the dungeons under Castle Blackmoor. The larger-scale stuff is often seen as endgame content, but with a tentpole dungeon, there's no reason why it can't exist in parallel with low- to mid-level dungeon diving.

(As an aside, I'm becoming more convinced that the retainer limit - the limit on how many people you can hire to actually follow you into a dungeon, as opposed to how many generic mercenaries you can hire for jobs in the overworld - is there to maintain this distinction. I.e.: You may have a small army willing to do as you command, for whatever reason, in the overworld, but that doesn't extend to having them file into the dungeon to simply overwhelm the denizens and traps there - you need special retainers willing to do that. All of which casts the retainers themselves in a new light, either as people who are completely desperate, or specialists in a niche industry, or both. But that's perhaps a discussion for another time.)

To sum up, I've rolled back on being reticent to set these sorts of hard distinctions in expectations; I think it's fine - important even - to set explicit expectations for your players, so they can see how the game works. Everything runs more smoothly if you do this. Shying away from setting these boundaries clearly, in fact, is what creates a lot of the toxicity down the line that you see in games. And having made my peace with that, I'm less averse to the idea of singling out one dungeon as being of particular significance.

The second reason I've come around to the tentpole dungeon is because it gives the players a backstop gameplan. This isn't quite the same as what Justin Alexander calls a "default action", but they're similar. Sandbox games run on the creativity of the players - you're very reliant on them forming their own goals and plans, because you're not giving them a thread you want them to follow. The tentpole dungeon gives players something to do, that's guaranteed to yield fun results, when they can't (or aren't in the mood to) think of something original. And, as a bonus, it still leans on the players making goals and strategising about how to accomplish them, in that way that dungeons do compared to quests.

There's an interesting thing there, actually, that I find helpful to note from a design perspective: Players can put the tentpole dungeon on hold while they go and do other things. It's not very easy to engineer a "tentpole quest" that allows you to do that.

People sometimes chuckle at the fact that, in Skyrim, you can get a particularly important and urgent mission as part of the main quest, and completely ignore it for months on end. Quest items will sit in your inventory (often weighing nothing), and the NPC you're supposed to give them to will wait patiently for their extremely urgent message or ingredient or whatever, while you divert to that glade across the map that looks kind of interesting.

Weirdly, I think this is actually good design. It's unrealistic, sure, but it sits in line with the game's ethos of letting you explore on a whim. Skyrim achieves a sense of exploration that a lot of other games (including subsequent video games, and to be honest, a lot of tabletop games) struggle to deliver, and this is in large part down to its allowing you to explore opportunistically. I.e.: On the way from A to B, you notice interesting thing C, and go to check it out, forgetting about B for a while. If B had a tight deadline - or even a forgiving deadline - you'd feel rushed, and like you didn't have time to divert (earlier games in the Elder Scrolls series made this error, sometimes quite egregiously). So the fact that NPCs don't really care about your being punctual isn't an oversight - it's actually an important, functioning part of the game's overall feel.

This is where a tentpole dungeon has an edge over a tentpole quest or other structure: You can leave a dungeon for a while, and (aside from the usual repopulating) nothing is pushing you to go back if you don't feel like it yet. Of course, it's fine if events in the world push players to be somewhere or face consequences (just thinking of my allergy above), but it's a real bonus to have the default game mode be something you can ignore if you feel like there's something more interesting to be doing.

As a bonus, starting off the campaign with a few sessions in the tentpole dungeon is also a really good way to train new players in the OSR style, and, weirdly, to de-programme players with abused-gamer syndrome. It's difficult to explain - probably a topic for another post - but being in an environment where you can choose path A over path B or C, and then being able to go back and verify that B and C were really genuine options (or even looping back on them from following A) does a lot to talk players down from looking for the one true, intended path. Of course, this only really works properly if you're designing your dungeon along Jaquaysian lines (RIP).

On top of all that, uncovering a big dungeon bit-by-bit is just cool, but it's something players can get weary of without a palate-cleanser in the overworld. I'm becoming more and more convinced that, everything else aside, the best way to run a megadungeon campaign is actually to situate it in a - largely optional - overworld wilderness, for players to take a break and have more diverse overall gameplay.

I guess I really am a convert then, if I think the best megadungeon campaign has a wilderness sandbox attached, and the best sandbox campaign comes with a tentpole megadungeon. Maybe I'm just carried away with my last campaign, just recently wrapped up (on which I'll write a post soon), but it seems more and more like Arneson and Gygax had it all figured out.