D&D General Worldbuilding considerations for a West Marches sandbox

[Yora]

For the sake of this discussion, a West Marches campaign is one that has more players that play at any given time, who go on many site based adventures in constantly changing parties, based on who's playing on a given day.
There are lots of different ways to approach this, like enforcing that each adventure is a one-shot and each game must end back at the starting town so every character is back in the stable next time the GM runs an adventure, or players having multiple characters so that several ongoing adventures can be active at the same time without the players having to wait for all the players in the party to find a time to continue before they can play again. You can make it a strict rule that all adventures have to be in dungeons outside of the starting town, or you can havr a campaign were some adventures deal with the locals of the town. Amd you can set up all kinds of systems do handle scheduling.
What matters in the end is that you have a larger number of players whose characters don't have fixed parties, and who tell the GM which places in the environment they want to go visit next time, instead of the GM telling them what adventure they'll be having today.

Running such a campaign comes with its own limitation of what works well as setups for adventures, but also opportunities you don't have in the same way in other campaigns. And I believe that it worls much better in an environment that is set up with these dynamics in mind. And of the world that the GM describes feels like it makes sense for the PCs to go on adventures in the way they do. I see this primarily as an issue of how you arange your pieces in the sandbox environment, but I think being selctive about what pieces you want to put into it in the first place could also help with making the campaign feel more engaging and believable.

Something that I thinkmis kind of an unstated default assumption with these campaigns is that they are taking place in a lawless borderland and that the main activities of PCs is checking out nearby caves and ruins.
Setting the campaign in an area with a small population, little existing infrastructure, and barely any official authority creates a good environment in which adventurers have free rein to roam around like a small army and can stumble on numerous places that have not been cleared of any potential valuables centuries ago.
Having just one major town, which might be fairly new, can be a great compromise between having the wilderness right out the door, and having access to a great number of services. And it also doubles as the point where the many PCs keep coming together to form parties for their next adventure.
When you have only one real town where all PCs always return to, I think it could be a really good investment of setup work to give that town some detail. That's one of the things I'm really interested about with this topic.

If the campaign is strict about each adventure being only a one-shot and not being left in the air to be continued at another time, then all the places players can go to would need to be pretty close together. Say you have 3 to 4 hours to play a whole adventure, then you don't want to spend more than 30 minutes getting to the dungeon and back each.
But that range can be extended to open up new areas for adventures by having other settlements that can serve as forward base camps, like keeps or highway inns, that are connected to the starting town by a road that allows PCs to directly move between them without spending any play time on it. The characters are assumed to have formed a group in the starting town and having travelled to the base camp without incident, where the adventure begins, and then later the adventure also ends there, with the journey home to the starting town also assumed to have been uneventful.

An important thing about the dungeons is that an adventure that needs to e wrapped up in one go doesn't need to explore the entire dungeon.You can also have larger dungeons that will easily take 3, 6, or 10 adventures to explore fully. But I think such dungeons in particular should be close to the starting town or one of the base camps. Even when you handle the journeys to and from the dungeon in a way that you can do three or four days of travel in 30 minutes. It would feel a bit strange that PCs would travel for a whole week two spend two or three hours exploring and then returning to civilization to do the whole thing again. If they can set out from a place where they can rest and recover in the morning snd get back before nightfall, then constantly going back and forth makes for a much more believable story.

One thing from the original West Marches campaign, that I think might actually have been the player's idea, which seems really useful is to have some kind of master map for the campaign that is accessible to all PCs (and respectively players), to see what has been discovered about the environment so far. This should be an object that exists as an object somewhere in the game world, probably the starting town. This can be the original approach of a big tavern table in which adventurers scratched doodles to plan their next adventures. For my campaign, I am planning on having a small cartographer's society that has a big public map of the border area where travellers can submit their own notes which are added by the custodian scribe after having compared notes from different people. There's probably other cool ways this can be done.

 

[Shiroiken]

Something that you should consider for such a campaign is to ignore travel that isn't part of the adventure. If the party is going to investigate some bandits, they won't face anything until they get into the bandit's territory. This keeps them from wasting session time with unnecessary distractions.

Designing the town can be crucial, depending on how much of an element you want it to be. If you plan any adventures in it, or want to use it for gathering information before they head out, then the more you put into it, the better off you are. I often use Gary Gygax's World Builder for town details, and Gary Gygax's Living Fantasy for cultural aspects (his Extraordinary Book of Names is also valuable to give consistent names to NPCs).

Practically, you don't want to just let the players pick the adventure at the start of the session, as this requires you to have every adventure prepped out in advance. I know it's cool, but you'd spend weeks in prep before you'd ever have the first session. A better approach is to drop clues about the different adventure options during play, then during the signup method put the introduction of the adventure. Depending on your group, this allows players to pick the adventures they're most interested in (and avoid ones that don't). This allows you time to prep between each session.

 

[WinkyDinkus]

I recently started a West Marches game and I intentionally added starting rumors to the board that were so far out it wouldn't be feasible to go to them anytime soon. The goal of their west marches is to reestablish a foothold for an empire, so they are trying to amass wealth to build holdings in the wild after pacifying it. This basically makes it so when they settle down and make fresh level 1 characters, they'll have a launchpad to go further and further out, while their old characters get to do really high level activity and domain play(not sure this would work for every table, my players liked it when we did it awhile back though so I figured I could bring it back.)

It's also possible to help travel by using things like portals at the end of dungeons or hidden away in ruins to facilitate travel long distances, so players only have to travel a couple days to reach the portals and be sent further up to speed travel long distance. I'm sure there is other ways to facilitate long travel in short spans of time but thats just what I could come up with while planning our game.

The biggest problem we've had running it now for around 5 sessions is keeping an appearance of a formerly lived in world alive to the players with every location. It becomes hard to believe after the 3rd or so guard tower looted that any group of people would need this much defense in the middle of nowhere, so keeping locations interesting and also believable has been pretty hard. Recently, I started inserting some unique one page dungeons into hexs where i wanted something, and since the creation process was already done, it let me take the time to consider why it was even there in the first place more seriously. It goes a long way when the players reach an old mansion and can immediately recognize the name as belonging to the bandit king's old long disgraced family or something along those lines.

 

[DeviousQuail]

A different system than D&D but I played in a west marches-esque game a long time ago that had a very Stargate feel to it. We were exploring the bones of an old empire and the number one priority at all times was to find the teleportation gates and getting the "address" to allow travel deeper into the continent. Overland travel was an absolute pain due to the vastness of the land, unforgiving environment, and plethora of magical dead zones.

Each adventure (1 to 2 sessions) would start with us teleporting to a region with some kind of mission. The DM, an absolute madman and inspiration, would sprinkle each mission with hidden lore and objects that tied into other missions/regions in really fun ways. They were usually bits of maps, logs and journals dropping hints about research and experiments, or "keys" that would open up new adventures. Everything was compiled upon our return for use in future sessions. I remember the map was less to scale and more of a general guideline as to what each region had going on.

My takeaways for world building:

• Having a method of quickly returning home is nice, but make sure that method has a real cost. Each teleportation required us to power up the gates and those power stones required a lot of money. While it doesn't have to be money, it sure does help incentivize bringing back things of value. For 5e I'd suggest an increased cost to the teleportation circle spell based on distance.
• Maps are cool, make one. Just don't try to fill it all in ahead of time. You never know what you or your players might want to add down the road. A few points of interest spread out with a bunch of empty space to be filled in later works.
• The home town/base/ship/etc should feel alive and grow in a way to reflect the players adventures. Not much more than a passing fancy or your players might never leave!
• As a DM you should know exactly why this area is so devoid of civilization. The players can have an idea but not the nitty-gritty details. It gives a bit of mystery that you can choose to unfold in bits and pieces over time. False flags and multiple versions of events can muddy the waters a bit and make for interesting notes for other groups.
• The DM had a list of potential missions for us that grew over time. Every time we finished one we'd have at least two new ones. I don't know what his system was but after a while he'd declare certain missions were completed by NPCs and were no longer available in order to thin things out. Looking back, I'm 99% sure he just recycled them while giving the impression of change.

 

[Yora]

Something that you should consider for such a campaign is to ignore travel that isn't part of the adventure. If the party is going to investigate some bandits, they won't face anything until they get into the bandit's territory. This keeps them from wasting session time with unnecessary distractions.

In a sandbox campaign, discovering new things along the way is a major aspect of the core concept. Especially when it is focused on exploring an unknown wilderness, which I think is essential for West Marches campaigns. Finding hints for new places to check out, and always having multiple choices of where you can go next is what the game is all about.

I recently started a West Marches game and I intentionally added starting rumors to the board that were so far out it wouldn't be feasible to go to them anytime soon.

That' a good element to mention. In a campaign where PCs are meant to return back to a main base often and relatively quickly, having pointers to where they could go to find something interesting might be a better approach than sweeping trough lot of hexes and hoping to find something.
Those rumors can be both generated by the GM, but also be things that other players noticed on their adventures but didn't get to check out before they returned home. Either because they were currently focused on something else, or because it didn't seem like something they were not interested in themselves. I think it could really help if there is some kind of institution within the game world through which adventurers exchange such information. That could be some specific place where information is collected, but also a kind of custom that exits among adventurers.
For my campaign, I am setting up a system where players can increase the reputation scores of their characters by writing some kind of report of what they've seen and done on their last adventure, which get collected and are part of the game world as tavern chatter that they have spread. Players can agree to not share what they did on an adventure and keep their discovery secret and simply not get any reputation for it. But they could also agree to not mention certain details which will be left out of any reports they write, so they can go back to check things out further later.

The biggest problem we've had running it now for around 5 sessions is keeping an appearance of a formerly lived in world alive to the players with every location. It becomes hard to believe after the 3rd or so guard tower looted that any group of people would need this much defense in the middle of nowhere, so keeping locations interesting and also believable has been pretty hard. Recently, I started inserting some unique one page dungeons into hexs where i wanted something, and since the creation process was already done, it let me take the time to consider why it was even there in the first place more seriously. It goes a long way when the players reach an old mansion and can immediately recognize the name as belonging to the bandit king's old long disgraced family or something along those lines.

A fun idea is to think about what civilizations existed in the sandbox area in the past. For my campaign I have six, though that's probably overkill. Three should totally be enough. The idea is to consider during the creation of the environment who had inhabited the area before and at what times, and what typical traits are common to the ruins they left behind. In my setting, the last civilization disappeared from the area a thousand years ago, but in the centuries since then, various barbarian tribes have come through the area. Barbarian sites are either in caves or the uppermost sections of older ruins. They have relatively little treasure in them and almost none of it is coins. If they left any traps behind, they are made of wood and ropes and often already nonfunctional when PCs get to them. And there might be plenty of warning markers to not continue into cursed areas at the very edges of where constructions by the barbarians end.
The idea is that when player have been to several barbarian or naga ruins before, they are able to recognize one when they discover it, and have some knowledge about what challenges they could expect ahead, and what contingencies they don't need to worry about. Floor traps might be a serious danger in naga ruins, but nobody has ever seen one in a giant ruin, for example.

 

[NotAYakk]

I'm thinking of adding a hex crawl to one phase of my next campaign (tm).

In this world, Doors (capital D) lead to Dungeons. They are left over pocket-planes built during the last apocalyptic war for various purposes, from Fallout Vault-like survival, to storage vaults, military supply depots, labs of wizards (aka demiplane), to long-term storage of magical "toxic waste". You access them via Doors, and current the economic phase-change in the world is fed by the fact that one empire has found and opened a Door to some ancient magi-tech (not that they are talking about this fact). (Think Indiana Jones, but where the Nazis managed to get ahold of one of those magic artifacts, and it wasn't useless to them)

So exploring the wilderness to find Doors, and being able to use Corridors (Dungeons with two or more Doors) as short cuts, should fit. There is going to be a scramble to find and search for Doors & search Dungeons for more of that juicy, juicy previous-era magic.

I'm not sure if I can fit a real hex crawl in. I have plans for T1 that didn't include a hex crawl. It would have to fit it somewhere in T2 I suspect; by T3, a full on magic-tech revolution and a global magitec war should be kicking off, and airships/teleportation/plane shifting starts showing up. A hex crawl won't work as well.

I suppose I could rotate my T1 plans to be a hex crawl. And T1 might be the best levels for it.

 

[Lanefan]

Re the idea of rumours and other adventure hooks:

In a game like this you could go a step further and literally have a notice board at the local mercenaries' guild, on which anyone who needs adventurers or knows of an adventure that needs doing posts it; and any adventurers looking for work can either take action on one or post their own "seeking employment" notices. (this also makes recruiting new PCs hella easy!)

Worldbuilding concerns:

If there's to be lots of characters that probably means your players will explore the variety of playable races, species and cultures available in your setting; meaning you need to have a homeland or point of origin for each one noted somewhere on the map, or note that a given species or culture tends to co-exist with (an)other(s).

Unless your home base town is impervious to attack, you'll need to find some way of keeping the adventure sites well away from it. One possibility here is a permanent gate or teleporter that takes the user outbound from town to the wild lands but (to prevent invasion) will only bring someone back inbound if that person has previously used it for outbound travel. (yes this presents issues if-when the party meets or rescues someone in the field, but so be it)

Either that, or you'll have to somehow rationalize why all those monsters in the adventures haven't overrun the town ages ago.

 

[Yora]

The monsters just might not have any interest in the town.

I'm somewhat surprised how many people are jumping on teleportation networks here. Though that is of course an option with this campaign structure.

 

[DeviousQuail]

You certainly don't need a teleportation network but it is one solution to the issue of venturing further and further out from home. Obviously, this is only an issue if you worry about the timeframes of travel. If you don't mind applying some handwavium you can turn a day, a week, or a month into a minute.

 

[Lanefan]

You certainly don't need a teleportation network but it is one solution to the issue of venturing further and further out from home. Obviously, this is only an issue if you worry about the timeframes of travel. If you don't mind applying some handwavium you can turn a day, a week, or a month into a minute.

Except if the party's travelling for a month just to get to the adventure (and thus, perforce, another month to get back) either they're going to have to take pack-mule-loads of rations with them or there'll need to be villages and waypoints along the way where they can stock up.