When I think of my most successful sandbox campaign and go down the bullet points of what I think made it work, I realized almost none of those points are shared in common with any of my other successful sandbox campaigns I've run. Only two points hold constant. They were dungeon-crawl heavy. And each individual session was fun in itself.

And really I think the reason the campaigns were dungeon crawl heavy is because I find dungeon crawls lend themselves to easily make for sessions that are fun on their own. So it's really just that one point. Of course, each session being fun in itself will probably help any style of campaign be more successful.

I tried to think of something more specific that separates sandboxes from working versus not. And what I came up with is smashing symmetry. This, too, I find helpful in any sort of campaign and junctures in which players must make a choice. But since players are making choices non-stop in a sandbox campaign, it's going to come up all the time in sandbox play. It's also something that is particularly noteworthy in sandbox play because I think there's going to be a tendency to do the exact opposite if you don't have this one on your checklist.


Here's what I mean about smashing symmetry. When presenting choices, they can't be functionally identical or equal. I think there is a tendency in running a sandbox to specifically try to force choices to be equal as possible. In play what happens if choices are too balanced or too indistinct is players take a lot longer deciding, and that can grind the game to a halt if it gets really bad.

Here's an example of introducing asymmetries.

When using Appendix A (random solo dungeon), in empty rooms with multiple exits, or in corridor intersections, a lot of times there's just no good reason to choose one way to go over another. Especially when Out-of-Character you know it's just a random dungeon. And what I've found is this can be solved by including a chance for dropping in one or more things randomly generated from Appendix I (dungeon dressing) located in a random direction.

"You reach a four-way intersection--do you go left, right, or straight ahead?" becomes "You reach a four-way intersection. On the ground straight ahead you notice a broken arrow. You hear screams coming from the left. What do you do?"

Huge difference. And this will probably also be a successful pattern interrupt to players who go through dungeons with algorithmic SOPs. Like stay right.

Quote from: Lunamancer on October 21, 2023, 01:27:18 AM
Here's what I mean about smashing symmetry. When presenting choices, they can't be functionally identical or equal. I think there is a tendency in running a sandbox to specifically try to force choices to be equal as possible. In play what happens if choices are too balanced or too indistinct is players take a lot longer deciding, and that can grind the game to a halt if it gets really bad.

Here's an example of introducing asymmetries.

When using Appendix A (random solo dungeon), in empty rooms with multiple exits, or in corridor intersections, a lot of times there's just no good reason to choose one way to go over another. Especially when Out-of-Character you know it's just a random dungeon. And what I've found is this can be solved by including a chance for dropping in one or more things randomly generated from Appendix I (dungeon dressing) located in a random direction.

I think of this in terms of player information. When I'm trying to have a proactive campaign, I give the players a lot of information about what is going on - so they can make informed choices. My rule of thumb is that the PCs should be better informed overall than most of the important NPCs they encounter.

This sort of sandbox is different than a lot of hexcrawl, but it is still distinguished by the open structure rather than a given mission.

I'll typically have several factions ongoing, and try to keep track of what each is doing. A common useful tool is a relationship map -- which is a diagram of NPCs with lines between them showing how they connect to each other. I first remember seeing these in White Wolf modules.

Quote from: jhkim on October 21, 2023, 06:54:19 PM
I think of this in terms of player information. When I'm trying to have a proactive campaign, I give the players a lot of information about what is going on - so they can make informed choices. My rule of thumb is that the PCs should be better informed overall than most of the important NPCs they encounter.

I don't disagree with this. The tricky part is we start at zero with a brand new campaign, and I'm not a big-fan of front-loading a bunch of lore. I want to get to playing as quickly as possible. Once play starts, PCs will accumulate information over time, and so we eventually get to a really good place there. But it initially helps a lot to do something to stir the pot. And even in the long run, I think the occasional stir is helpful.

QuoteI'll typically have several factions ongoing, and try to keep track of what each is doing. A common useful tool is a relationship map -- which is a diagram of NPCs with lines between them showing how they connect to each other. I first remember seeing these in White Wolf modules.

I like using the 1E loyalty system for individual relationships. I'm not as enthusiastic about factions, but I can say last time I ran the basic D&D module The Veiled Society, I had the key choices would earn a point with one of the three factions, and the running scores would determine at the end of the adventure which faction the PCs are allied with. The "net deeds" in that direction would affect the PCs "Repute" score (this taken from the Lejendary Adventure RPG).

My most successful. Well I started it on Harn which is pretty well detailed already and used RuneQuest (without magic) which was pretty gritty and low powered.
So it started with the characters working for an Earl and learning the world and factions so it was a bit railroady at first as they did missions for the Earl. 

Then I planned out a civil war. Sieges, troop movements, intervention by a neighbor. I put that on a timeline with the idea each event would occur unless the PC did something I felt might change things. Improvising within that context was pretty easy. Nearly every adventure was political or some direct mission to thwart something.

Once the PCs were cut off from the Earl they started to make their own plans and seek alliances and try to stop the bad guys. They stopped the invasion by the foreign power but eventually the bad guys won and they fled the area and worked out plans to return for revenge. The campaign fell apart then as we all went to university. It was very memorable and fairly easy to GM.