How do I run a good sandbox game?

How do I run a good sandbox game?

Most important things thing is players that don't have to be handheldall the time, other than have a lot of general loose plans and be prepared to adapt to your players actions.

You don't. A good sandbox game has a player/PC driven plot or plots. That, by the way, is why they often fail, since most players are absolute dogshit at actually driving a plot.

But assuming you have a player group that is capable of doing this, all you need to do is make a detailed setting with lots of opportunity for conflict but none of it moving very quickly, get someone in the group to commit to a goal or goals, and watch them go.

I've never dungeon mastered, but I would recommend making encounter tables and quick dungeons for your world's regions.

A sandbox is only as interesting as the toys you fill it with.

Litter the landscape with tombs, ruins, and monster lairs worth exploring. Make sure there is at least one mega-dungeon (less than a day's travel of a smallish town) that you've detailed down to the 5th or 6th level but could go deeper if needed. Have tables ready for re-stocking this dungeon from time to time: whereas small tombs and ruins can be cleared, the mega-dungeon should be fundamentally inexhaustible.

Make sure the towns are full of rival factions, people with problems in need of solving (side-quest givers), and guilds that would be willing to let the PCs join. Make sure that the guilds can provide basic gear and services the PCs need, but also make sure that they can't solve problems *for* PCs (rather, they too have problems that need for the PCs to solve them, if the PCs are willing). Assuming that you're using some version of D&D as your rules-set, character classes and multiple players ensure that your game probably won't fall into the "Elder Scrolls" trap of one special snowflake character eventually becoming guildmaster of every society in the campaign.

Finally, come up with what can only be described as a "Hypothetical Future Timeline" that describes the next year or two of major campaign events, as they would happen if the PCs never existed or choose never to intervene. Then make sure that you religiously update this timeline according to what the PCs do or don't do. The game only has meaning as a sandbox rather than a railroad if the PCs' actions actually change this timeline. But the world should still be living and breathing, which means that Time Marches On™ and Stuff Happens™, regardless of whether the PCs are active agents or self-absorbed dwadlers.

One other thing, fairly important: clues and mysteries. Every good sandbox (and this is also true of every good mega-dungeon) has at least one major mystery in need of solving. Decide what it is, litter the place with small clues that might help the PCs tie it together, but don't sweat it if the players never catch on or never figure it out. That's why it's still a mystery.

You don't.

It's like writing a good mary sue or being a great that guy. By its existence it becomes bad

But why are sandbox games bad?

Present the illusion of the game being a sandbox while running a normal campaign.

There is no such thing as a sandbox game. All games are linearly driven by the GM. Players are usually incapable to do sandbox even if they know the setting, because they don't have any set goals. It never happens they have a plan to do something, all they want is a game. Sure they like "freedom" but are dummies that can't go without your lead, you have to flock your sheep. For you are supposed to be a river to your people.

You know, I kind of feel bad for you. Not that I've done it often, but I have run sandbox games, with a set of PC goals and plans to pursue them. And the funny thing was, I wasn't planning on that, they just derailed my proto-campaign into something sandboxy.

A good sandbox game requires both the DM and the players to have the right mindset for it.

Also, don't drop the players in a world and go "Ok, do whatever" and then wait for them to make the first move. Drop a few plot hooks in, see which one they bite for, and modify the game accordingly.

>t. someone who bought into Bethesda's sandbox meme and got burned because it turned out they were theme parks all along.

By it's nature, it's directionless and a waste of time.

A good GM can expand and be flexible with the world, but sandbox is just GMing with no direction or purpose.

>By it's nature, it's directionless and a waste of time.

If you have shitty players, yeah.

Usually the players want the sandbox campaign, but then they sit around and don't do shit. They need to pick a direction, or the GM needs to give them something, then they need to move from there.

However it is best if there is an overarching conflict to the world that drives the action. Apocalypse, especially zombie apocalypse campaigns, are good for sandbox because there is a "plot" but it up to the characters how they respond to the in-world events, and their choices can wildly influence the entire direction of the campaign.

But that's the point, in a sandbox, the players/pcs provide direction and purpose, and the GM facilitates this.

There's a reason that old modules had tables of random rumors, some true and some false, that would be given to newly-rolled up 1st level characters.

You most certainly can just drop the players into the world and tell them, "Do whatever", as long as they're all aware that you need to find 1 gp to earn 1 XP, and Naegl the fighter thinks that the nearby Caves of Pain have buried pirate treasure (true), Largo the cleric thinks that somewhere in the Murky Forest there's a collapsed silver mine (false), and Ellismere the Elf has heard that an abandoned castle five leagues north of town is full of magical artifacts (partially true, a collapsed wizard's tower has a cache of scrolls and potions and a half-charged wand).

They'll take the hint.

Whups, meant to be replying to this post:

Not him, but for this statement,

>However it is best if there is an overarching conflict to the world that drives the action. Apocalypse, especially zombie apocalypse campaigns, are good for sandbox because there is a "plot" but it up to the characters how they respond to the in-world events, and their choices can wildly influence the entire direction of the campaign.
I would fundamentally disagree; not just that it's a good idea, but that it's a sandbox at all. A sandbox, at least to me, has a wholly player emergent plot, and is not simply "give them lots of options to deal with whatever the problem is". As soon as you have an overarching conflict that drives things, you've left the sandbox.

What I try to do is to work with a player or players to actually have a proactive goal, and usually one that will take some time to do.
>Rebuild/take control of a magic tower that they purchased on the cheap.
>Spread your character's minor religion until it becomes a major faith.
>Build up this magical underground exile hellhole into an actual respectable political unit.
But they're all things where the players are actively trying to do something, instead of reacting to an external threat.

You give them the illusion of a sandbox.

You have stories and plots you want to run, and wherever they go, that's what they find.

You want to run a scenario where they rescue a kidnapped princess. So the next town, that's what they get presented with. They don't want to do it and fuck off somewhere else? Mmkay. shelve that. Wherever they go next you drop the next plot in your shelf of ideas until something sticks.

Then build on that and try to develop player agency. If they start getting goals and plots of their own, run with it. If they don't, keep throwing stuff at them until it sticks.

I ran a VtM game for about 2 years in the 90's that was a "sandbox".

Think on your feet. Adapt material from from the game and relocate as needed. Steal ideas from anywhere shamelessly.

Don't. Give the PCs a clear goal ahead of time or they will fuck up. They will always fuck up.

If given infinite options and complete choice but no direction, most people will either do nothing of value, or dick around pointlessly for a while then declare that there's nothing fun to do.

The two ways of solving this are to either have so many setpieces and loose plotthreads that the players are tripping over things to do whenever they want, or by having a well-developed setting that reacts believably to the players' actions (in more complex ways than sending unending waves of annoying mooks until the players are dead) and managing to convince one of them to kick the metaphorical beehive

Alternatively, start out with the skeleton of a plot and a handful of "clues" for them to find, then just go along with whatever insane ideas they come up with about what's going on. They'll write the entire campaign for you if you're capable of winging it.

Hi, Oldbeard here.

You can run a good sandbox game. Yes, it can happen.

While a sandbox is player-driven, you can't just dump all the toys in, lean back in your chair, and put your brain on autopilot. You have to react to your player's decisions, and that helps generate fodder. Yes, the player initiates. Yes, you don't have to hold their hand. But it's more of a dance, really. The best sandbox DM will have an entertaining reaction to the player's decisions.

"But what if the players are unoriginal?"

Then provide an original reaction to it. See, the sandbox sells itself to the DM by making them think, "Wew, I don't have to write backstory anymore!" But the catch is now that you have to trade it in for pure improvisation.

But what if you're not very good at improv? What if you can't provide an original reaction on the fly?

Then do the un-original one. Go ahead and rip-off a movie or TV show if there's a high probability your players haven't seen it. If it's not completely hackey, it will have your personality fingerprint all over it and they won't even notice. Especially if you take it from another genre completely.

> You give them the illusion of a sandbox.

Yes. Exactly. But don't overdo it. You need an aforementioned "shelf" that's big enough to recycle ideas until something sticks. Experienced players can hear the faintest clickety-clack of a railroad from miles away, "Yes, but this princess is wearing a green dress, it's a completely different princess." Better to turn the princess into a valuable document, a Roman battle standard, or General Lee's horse. Big cosmetic changes like that will get you more mileage.

>Don't. Give the PCs a clear goal ahead of time or they will fuck up. They will always fuck up.

And they will learn from it. You sound like a nanny-stater railroad tycoon. Why not put your PCs in a blanket burrito and pin a medal of valor to each one? We're trying to give them a real sense of accomplishment here.

I disagree with your definition of sandbox. Imagine this start to a campaign: All the PCs, for one reason or another, have sought to join an underground Thieves Guild. Their initiation sneaks up on them, quite literally, as they're drugged in their sleep following what they were lead to believe was a solid no. Drugged, they're carried off to some other town, and let loose practically naked but with instructions: Within three days, they must have gathered 200gp, obtained a weapon and forged a false identity that at least one trusted townsperson will vouch for.

This kind of cold start sets clear goals for the players, but unless the GM has concretely planned out a way or ways the players could obtain these things, they've definitely been dropped in a sandbox.

Why would you want to run a false sandbox tho
Half the appeal of a sandbox is that you, the dm dont know whats gonna happen.

If you have hidden rails then its just as boring as a non sandbox campaign.

>Why would you want to run a false sandbox tho

Because it's only "real" when it's thrown over the DM screen. Before that, it's nothing more than a very flexible concept/plot device with maybe some skin on it. If it's not already been established in-play, the DM can "re-dress the set" behind the scenes any way they want.

"Hm. What about that idea I had with the talking furniture? They ran right past it. I know, I'll do the same thing with the talking statue on DL 4, she's a real bint, they'll love it."

>Step 1:
Acquire a sandbox.

>Step 2:
Play a game in it.

I don't know why this questions keeps coming up. It's so simple...

Sorry, I'm a father, it's a compulsion. In all seriousness though, I'd recommend making things vague at first, and taking advantage of a slower schedule (and possibly enlisting player help) to refine things as the players decide what to focus on.

Start with a setting, established or homebrew, figure out your starting location in a bit of detail, and then map out a selection of locations, factions, and conflicts. Really feel out for what they're interested in, ask them what they think of X or Y aspect of it, and then call the session to start planning out the next bit in more detail. (And be sure to put a slight bit of effort into their second or third choices, just in case)

If it's an established setting, give them a book or two and ask them to pick out what they like about it. Sand-boxing is collaborative as shit.

Don't tell your players anything that makes sense.
Everything you say must be laden with such dire meaning that if they scrutinise and analyse your words they become paranoid, and if they don't then their ignorance effectively removes them from the game.

Again that sounds boring to dm since it doesnt matter what the players do you already know what plot is going to happen

>Again that sounds boring to dm since it doesnt matter what the players do you already know what plot is going to happen

Only if you're assuming the DM has one plot card. They should have several in-reserve. Your job in a sandbox is to decorate the trope factory.

Not to mention plots that organically emerge from the session itself:

- Player makes an excited speculation out-loud. Other players dismiss it, but you silently note it down. This would make a great surprise encounter a month or two from now; long after they forgot the discussion entirely. Especially after you mess with the details.

- Improv can literally create new plots on the fly.

Pic related.

>run sandbox
>Players resolve issues before expected
>Run out of cool ideas
>Need help and inspiration once again.

>Sorry, I'm a father, it's a compulsion

Relevant how?

>Gives example of completely impotent one-shot "sandbox" without following through.


Anywhere you can find a copy of that book online?

You don't. You cheat and fool the players into thinking it's sandbox with quantum ogres.

Has it right. Make the world or use one, and make sure it's full of things to see and do.

>These two countries are on the verge of war. Pick a side?
>There is a whole new continent that was just discovered. Go to the frontier to stake your claim for yourself or a faction of the old world
>This country is researching ancient weapons and magic, stop them or join them
> Mercenary bands are fighting for control of a valuable region, what's one more?

Take these broad hooks and let the players go where it interests them. If they ignore all of them, have an antagonist draw them to one. From there, give sub hooks they can look into to further the main hook. Personally, I hold unused hooks in a sort of stasis until a player action affects them. If the players are faffing about in the new continent or with the mercs, then the war is still cold. You decide to take your new Merc band on tour or you join under the passionate of one of those factions and suddenly it might start heating up.

Let the players choose where in the sandbox they want to play, and let their actions build it along with you.

Hi Relevant
I'm Dad

Just talk to the nearest hipster improv nerd (they're everywhere) and right away they'll give you a used copy for free.

Just don't tell them what you're using it for, or they'll want it back.




I use those fuckers as much as dice. And follow them about as strictly....

Who is creating that initial goal though? There's more to a sandbox than "we have a lot of options as to how to proceed". Imagine a campaign where you're agents of a weak king in an ASOIFish sort of setting. You need to get enough of the major noble houses on board with a campaign the king is planning to field 6,000 soldiers between them. He doesn't care which houses you get, as long as you can deliver the troops within the year; players have total discretion over which houses to woo and what methods used to get them.

Would you really call that a sandbox? What makes it sandboxy, other than operational freedom? At what point, since said operational freedom is a continuum and not a binary, do you say you've left the sandbox? Suppose the instructions are to align these 5 great houses to get the troops and not others? Suppose you have restrictions given on the PCs methods but not which houses to align? Suppose you have both?

>I've never DMed
>let me offer some dogshit tier advice anyway
You're a credit to the thread user

Start them off with a quest, players need a push and once the ball is rolling they'll take it from there.

Same. I even roll them out in the open sometimes.

So yeah, pure sandbox really is possible.

Good point. A lot of DMs are confused that they have to spoonfeed goals to the players. That's not necessary. If you create a detailed environment with enough flaws, and encourage the players' sense of personal initiative, they'll want to start fixing things entirely on their own.

Unfortunately, the majority of DMs out there are control-freaks who can't handle that. So they have to micromanage everything.