Sandbox RPG: help me understand

Best Selling RPGs - Available Now @ DriveThruRPG.com

lategamer

Writer, Sailor, Filmmaker, Irishman,
Joined
Dec 7, 2021
Messages
1,768
Reaction score
3,875
I had a conversation about this on reddit, but it left me unsatisfied. The debater was more interested in telling me THAT I was wrong rather than HOW I was wrong in my perception of RPGs and Sandboxes. He did say it was more environment-driven than plot-driven which, because I write fiction for part of my living, doesn't make a huge amount of sense (for me ....they're interlinked)

I plainly do not understand what people mean when they talk about sandbox TTRPGs. Because from my point of view, all TTRPGs are sandbox. You can be playing a hard-boiled detective in 1920a Chicago and the GM can throw superheroes at you. Or you can (especially in a narrative game), make up any possible outcome.

So what am I missing?
 
These terms I think a best understood as a spectrum rather than absolutes. So if you were to take a common roleplaying premise of occult Investigators, you could have:

Railroad: In the adventure we playing today you are investigating a haunted house. You will go through an exact, preplanned sequence of events and the only valid solutions to the adventure are the ones written in my notes.

Mission based: In the adventure we playing today you are investigating a haunted house. How you deal with the investigation and attempt to solve it is entirely up to you.

Directed sandbox: You are occult investigators. There are a whole bunch of haunted houses out there, which ones you choose to investigate and in what order is up to you.

Pure sandbox: There some haunted house out there, but if you want to start a rock band or open a car dealership instead, that's cool.
 
These terms I think a best understood as a spectrum rather than absolutes. So if you were to take a common roleplaying premise of occult Investigators, you could have:

OK, I dig that. Very good explanation.

I think I've always played with the premise of pure sandbox but with presenting multiple possible missions if the players care to look.

My T2K4 Xmas one shot is most likely directed. My Dune campaign is Pure Sandbox - though the introduction is a Railroad to introduce the player characters to each other.

Nice.
 
PolarBlue did a pretty good job. No arguments from me there.

Here's another example, from a slightly different perspective.

Imagine an archetypal premise: the PCs are from Nice Kingdom, there's a looming threat rising over in Evil Empire. In a non-sandbox game, there's probably an assumption the PCs will learn of the threat, undertake a series of quests, and eventually defeat Emperor Evil.

In theory, running that as a sandbox is as simple as deciding, "The PCs don't have to face off against or care about Emperor Evil, but if they ignore him they have to deal with the consequences." However, if Emperor Evil's unopposed activities dominate everything that happens, it's probably not really a sandbox anymore, because the PC's are seeing the world collapsing around them, and the consequences of ignoring it may make the game untenable.

So, instead of a single, overwhelming threat, a good sandbox with a similar grand scope will typically have an array of different factions, all with their own agendas, that the PCs can interact with, influence, work for, fight against, play one against the other, etc ...The game that then emerges is likely to be one that neither players nor GM envisaged when the game started. And the important point is, if the players aren't interested in one aspect of the world, there should be other things for them to sink their teeth into.

Typically (but certainly not always) I'd say a good sandbox also has a world that carries on being a world, without caring what the PCs are doing, unless the PCs are able and willing force their will upon the world. This, I think, is what the other person meant by, "environment-driven". Things are happening around the PCs, and they can choose to get involved or not. Frequently, these events are situations for which the GM doesn't necessarily have a solution (ie, plot) in mind.
 
Last edited:
Typically (but certainly not always) I'd say a good sandbox also has a world that carries on being a world, without caring what the PCs are doing, unless the PCs are able and willing force their will upon the world. This, I think, is what the other person meant by, "environment-driven". Things are happening around the PCs, and they can choose to get involved or not. Frequently, these events are situations for which the GM doesn't necessarily have a solution (ie, plot) in mind.

OK, another good perspective. I'm not being adversarial here but this bit seems to assume some sort of status quo or isolation from the world?

I mean, Ars Magica has the sandbox of the papacy and the heretics and the plague....in the quasi-historical world and then the more player-driven stuff with infernalia, the fee, and the other magical trappings which arguably have a more plot driven bent for the players?
 
OK, another good perspective. I'm not being adversarial here but this bit seems to assume some sort of status quo or isolation from the world?
Not really or at all I would say. It's just saying the world isn't built/doesn't function around the players. So a king might invade a town the players are in because he was personally planning to do so anyway. Events will play out as they would based on NPCs motivations, geography and so on.
 
We've had some extensive (and contentious) threads on this subject in the past. This one remains the primary resource:


The definitions we came up with were as follows:

SANDBOX
An approach to playing RPGs, defined by how players and referee behave instead the details of a given system, wherein the players are granted complete freedom to do anything within the setting as their character, limited only by common sense restrictions. The players have the ability to "trash the campaign", their actions not limited to staying within the bounds of a preconcieved premise, while the GM acts as a neutral arbiter of events in the gameworld and the manner in which the gameworld responds to the players, without an objective or steering the action in any particular direction.

QUALIFIED SANDBOX
Anything that a person in that setting can do, PCs can do, with several pre-agreed upon exceptions.

WORLD IN MOTION
A specific form of Sandbox, coined by Vreeg in 2010, wherein the gameworld "lives and breathes" outside the PCs' scope; events occur which they may not even be aware of, or became aware some time after they actually occured. The inhabitants of the world have a will, motivations, and goals of their own which they will act upon, regardless of the PCs' own motivations, unless they are in direct relation to each other. "World in Motion" applies both to this approach to gamemastering a Sandbox, and the various Tools used to support and enable that playstyle.

SCHRODINGER'S SANDBOX
The opposite of a World-in-Motion campaign, a Sandbox where the content in the setting is developed randomly or by improvisation in response to the player's choices.

ARBITRARY RESTRICTION
A restriction on player character choices that would not exist if the gameworld was a real place and the characters were real people living in that gameworld. In other words, it cannot be rationalized "in game" or "in character"

BAG OF STUFF
Term coined by @robertsconely meaning material that the human referee has internalized and can draw on to create elements of the campaign on the fly such as locales, characters, and plans.

INITIAL CONTEXT
Term coined by @robertsconely meaning the situation at the beginning of the campaign that the players are aware of as their characters, intended to give players the information needed to make informed choices from the start.
 
OK, another good perspective. I'm not being adversarial here but this bit seems to assume some sort of status quo or isolation from the world?
I'm not exactly sure which status quo you're referring to. (And I'm not seeing anything adversarial. Seems like a good faith discussion, to me.)

In theory, the world is evolving and changing according to the interactions of the powers-that-be, independently of the PCs. However, if the PCs start getting involved, this can effect the course events take (or not, depending on how powerful/effective they are).

If the PCs are a bunch of gutter-dwelling ne'er-do-wells, they most likely won't influence international politics, but the scope of the game is most likely focused on the politics of street gangs, hucksters, the local watch etc.... If they're epic heroes, their decisions might determine the fate of nations. A game might move from one political/power level to the next, or start and finish around the same tier of power.

Edit: Also, what Séadna Séadna said.
 
This is still a good, no BS-guide to running one. From it comes the quite wonderful aphorism "Plot is a four-letter word"


Heh, some more good ones…
”If you’re engaging in Edition Wars, you’re already fucking up.” :grin:
Bounties - good one, not only is it a perfect PC-chosen adventure, but makes things look lived in.
”Where they can have a lot of agency and even the GM can be surprised.”

He pretty much nails the essentials and gets everything right. B2 is the standard go-to. I’d add Frandor’s Keep as an upgrade, but they never finished the Mines of Chaos.
 
And really, a good World In Motion is one that can take the GM AND the players by surprise.

Case in point: so the party's based out of a large capital city of a maritime nation; more often than otherwise, I've been running groups out of this city since the 1980s, and its fate has a disproportionate influence on gameplay.

And so it's Storm Season, prevailing weather coming out of the northwest. And the party's planning to sail west out of the harbor, which is a tricky thing at the best of times unless you've got a wizard who can manipulate or negate wind. So I roll for daily weather. The table I'm using at the time is a Rolemaster-based one. And I roll high open-ended, which for those of you unfamiliar with RM means some really freaking extreme result.

A few more rolls ensue. Without belaboring the process further, the end result was an epic hurricane that overwhelmed the city's weather wizards, and not only flattened the city's shanteytown -- unprotected, and right in the teeth of the storm track -- but one of the spinoff tornadoes flattened the city's hall of records, which included the land and tax records for the entire kingdom. Sodden confetti galore.

This entirely random event, which took place several years ago, STILL bedevils the kingdom, and was a material cause of the subsequent civil war.
 
ARBITRARY RESTRICTION
A restriction on player character choices that would not exist if the gameworld was a real place and the characters were real people living in that gameworld. In other words, it cannot be rationalized "in game" or "in character"

What would be an example of arbitrary restriction?
 
I think the two essential characteristic of a 'Sandbox' style rpg session are: 1) true, absolute player agency, and 2) a DM who has their act together well enough to keep play moving if the characters go somewhere or do something they hadn't predicted. In practice, these are closely related things because one of the main reasons sort-of-but-not-really skilled DM's structure sessions around pre-determined events or places is that they don't understand how to organize and run a setting where players are truly free to make their own decisions about what they do and where they go. So they get uncomfortable or offended when players decide to walk away from a dungeon they mapped or shrug at a court intrigue they planned in gory detail. This sort of situation often gets presented as players being d!@$, but really the problem is the DM doesn't have the resources or confidence to react at the table when something unplanned happens.
 
And really, a good World In Motion is one that can take the GM AND the players by surprise.

Case in point: so the party's based out of a large capital city of a maritime nation; more often than otherwise, I've been running groups out of this city since the 1980s, and its fate has a disproportionate influence on gameplay.

And so it's Storm Season, prevailing weather coming out of the northwest. And the party's planning to sail west out of the harbor, which is a tricky thing at the best of times unless you've got a wizard who can manipulate or negate wind. So I roll for daily weather. The table I'm using at the time is a Rolemaster-based one. And I roll high open-ended, which for those of you unfamiliar with RM means some really freaking extreme result.

A few more rolls ensue. Without belaboring the process further, the end result was an epic hurricane that overwhelmed the city's weather wizards, and not only flattened the city's shanteytown -- unprotected, and right in the teeth of the storm track -- but one of the spinoff tornadoes flattened the city's hall of records, which included the land and tax records for the entire kingdom. Sodden confetti galore.

This entirely random event, which took place several years ago, STILL bedevils the kingdom, and was a material cause of the subsequent civil war.

Honestly, that's the sort of stuff I live for in a campaign, and where the majority of my enjoyment comes as a GM, together with being surprised by my players. It really makes the world feel "real" in a way I can't quite put into words.
 
I think the two essential characteristic of a 'Sandbox' style rpg session are: 1) true, absolute player agency, and 2) a DM who has their act together well enough to keep play moving if the characters go somewhere or do something they hadn't predicted. In practice, these are closely related things because one of the main reasons sort-of-but-not-really skilled DM's structure sessions around pre-determined events or places is that they don't understand how to organize and run a setting where players are truly free to make their own decisions about what they do and where they go. So they get uncomfortable or offended when players decide to walk away from a dungeon they mapped or shrug at a court intrigue they planned in gory detail. This sort of situation often gets presented as players being d!@$, but really the problem is the DM doesn't have the resources or confidence to react at the table when something unplanned happens.
It's also about system and time. T&T with the wonderful flexibility of monster rating makes it easy to stat up an on the fly encounter. BRP based systems mean you have to have pre-assembled possible encounters to be able to do the same, 4e D&D to get the best out of the tactical play, needed you to have the stats and playboards available (though the amazing encounter generator, that wasn't supported, made that a lot easier).

Hence the attachment of time limited GMs to the stuff they prepared :smile:
 
We've had some extensive (and contentious) threads on this subject in the past. This one remains the primary resource:


The definitions we came up with were as follows:

SANDBOX
An approach to playing RPGs, defined by how players and referee behave instead the details of a given system, wherein the players are granted complete freedom to do anything within the setting as their character, limited only by common sense restrictions. The players have the ability to "trash the campaign", their actions not limited to staying within the bounds of a preconcieved premise, while the GM acts as a neutral arbiter of events in the gameworld and the manner in which the gameworld responds to the players, without an objective or steering the action in any particular direction.

QUALIFIED SANDBOX
Anything that a person in that setting can do, PCs can do, with several pre-agreed upon exceptions.

WORLD IN MOTION
A specific form of Sandbox, coined by Vreeg in 2010, wherein the gameworld "lives and breathes" outside the PCs' scope; events occur which they may not even be aware of, or became aware some time after they actually occured. The inhabitants of the world have a will, motivations, and goals of their own which they will act upon, regardless of the PCs' own motivations, unless they are in direct relation to each other. "World in Motion" applies both to this approach to gamemastering a Sandbox, and the various Tools used to support and enable that playstyle.

SCHRODINGER'S SANDBOX
The opposite of a World-in-Motion campaign, a Sandbox where the content in the setting is developed randomly or by improvisation in response to the player's choices.

ARBITRARY RESTRICTION
A restriction on player character choices that would not exist if the gameworld was a real place and the characters were real people living in that gameworld. In other words, it cannot be rationalized "in game" or "in character"

BAG OF STUFF
Term coined by @robertsconely meaning material that the human referee has internalized and can draw on to create elements of the campaign on the fly such as locales, characters, and plans.

INITIAL CONTEXT
Term coined by @robertsconely meaning the situation at the beginning of the campaign that the players are aware of as their characters, intended to give players the information needed to make informed choices from the start.
This certainly gives me new terms with which to describe how a game I run is likely to go. "Session 0 to make characters and establish Initial Context and discuss any Arbitrary Restrictions. It will begin as a Schrodinger's Sandbox, but as things progress I'll have a larger Bag of Stuff, and it will likely end up somewhere between Schrodinger's and a World in Motion."
 
I think the two essential characteristic of a 'Sandbox' style rpg session are: 1) true, absolute player agency, and 2) a DM who has their act together well enough to keep play moving if the characters go somewhere or do something they hadn't predicted.

There's a third: the GM's willingness to let go. If the players solve everything three hours early and have a walkover, then they do. If the direction the players go in invalidates your game prep, then it does.

What would be an example of arbitrary restriction?

"Wizards can't use swords, so you cannot physically pick that sword up and you can't physically swing it at anyone or anything."

Honestly, that's the sort of stuff I live for in a campaign, and where the majority of my enjoyment comes as a GM, together with being surprised by my players. It really makes the world feel "real" in a way I can't quite put into words.

And, after all, it leads to plot all its own. Take that hurricane. The players decided to dive into the relief effort. The party wizard had some repair spells that could (a) gather scattered pieces of a discrete object into a neat pile, from which if there was enough gathered he could (b) reassemble the object. This endeared him to the land records bureaucracy, which he leveraged into giving him a free hand in rebuilding Shanteytown, which with the help of the party Multimillionaire turned into him buying large tracts of the area on the cheap -- the loss of the records put paid to the multi-decade lawsuit among heirs of the former landowners, which had paralyzed any notion of development. The wizard then moved agemates from wizarding academy onto the property, with him as the landowner, and now he's got genuine political power.

Which stood the group in good stead when the civil war kicked off, about 15 minutes later. (grins)
 
I had a conversation about this on reddit, but it left me unsatisfied. The debater was more interested in telling me THAT I was wrong rather than HOW I was wrong in my perception of RPGs and Sandboxes. He did say it was more environment-driven than plot-driven which, because I write fiction for part of my living, doesn't make a huge amount of sense (for me ....they're interlinked)

I plainly do not understand what people mean when they talk about sandbox TTRPGs. Because from my point of view, all TTRPGs are sandbox. You can be playing a hard-boiled detective in 1920a Chicago and the GM can throw superheroes at you. Or you can (especially in a narrative game), make up any possible outcome.

So what am I missing?

Can I ask what your definition of sandbox is (if you have one)?

I do think among most sandbox gamers, we tend not to frame it as about plot or fiction, because it is a response to things like plot driven railroads. That doesn't mean stuff doesn't emerge that could be called story or called plot (my character went into the tomb, lost his arm and came out with a radically different personality, could be described as plot and could happen in a sandbox). I think the main thing about how stuff like story is handled in a sandbox is: it isn't the purpose or goal, and storylines and plots are not considered terribly precious or in need of reaching a fulfilling conclusion. Most of the priority is on not knowing what is going to happen, seeing where things go, and players having a sense that they are making choices that could lead anywhere. But every sandbox is a bit different. Every sandbox has slightly different priorities.

To me at its basic level sandbox is about creating a large setting, with locations, NPCs, groups, etc, that the players are free to explore however they wish. There is no planned adventure for the evening. Any adventure that does arise is a response to what the players do.

We have had tons of discussions here about what sandbox is, and debates about whether it exists, whether it is what is claims to be. My view is sandbox is a structure and approach that works for some people, but if you are playing something that is more like a situational adveture style campaign (I am just trying to guess based off your post) it is close enough to the level of agency you get in a sandbox, a sandbox might not be an especially useful model for you.

Also I think people who place great emphasis on story in RPGs that also maintain player agency, and people that play sandboxes are kind of offering two different solutions to the same fundamental problem: avoiding railroads---especially scripted ones.

Just to give you an idea of how I plan and run a sandbox to see if this means anything to you. And this assumes a wuxia fantasy setting, which is just what I have mostly run the past several years (I would tend to take a different approach to the modern campaigns I run). I begin with a map and some rough sketches of a setting and its history. To keep things simple, I might start in a particular region on a single piece of graph paper. Before I put anything on the map I develop a rough sense of the groups, the conflicts, etc. Then I start marking down stuff like the HQ of different sects, important settlements, ruins, caverns, tombs, places where manuals might be hidden, roads, trails, caravan routes, inns, teahouses, brothels, resources, key manors, temples, notable supernatural or weird spots, etc. A lot of the work after that is developing the different sects, making sure there is stuff 'dungeony' to explore, but mostly focusing on building a martial world. Then I start putting together things like encounter tables (which is how I populate the threats and wildlife). By the time I am done, I have enough stuff on the map, and enough material for any of those things, that if the players decide to go there and interact, it just sort of feels like chemistry and a kind of adventure reaction easily unfolds. I try to make all the locations gameable (there is stuff there that will be fun in a session) but I try not to make it feel like a video game where you get to spot A and there is a mission waiting for you (it is more like the players figure out how they want to interact with the stuff at this place and they drive where that goes). A lot of what ends up fueling my campaigns is conflicts and alliances between groups and NPCs that the players involve themselves in (this is a setting with a martial world, so that lends itself pretty well to adventures of that sort). I also treat my NPCs and groups as active forces in the world. I try to have them react and move around how player characters might. They formulate plans, respond organically when players do things. Another important element here is not having any pre-planned notion of what is going to happen, not wanting as a GM for the campaign or individual adventures to reach any particular destination.

I think a lot of this stuff can happen in a number of adventure structures. What sandbox has done for me is fold a lot of best practices that work for what I like, into one canopy (that I find helpful for guiding prep). Before I even knew what a sandbox does I was using a lot of these elements (especially the treating NPCs as live, with going with the flow and allowing things to unfold naturally through player choice, NPC reaction, dice). Again a lot of this is best practices, but I think for a lot of people sandbox play was a reaction to both the scripted railroad style of adventure that was common in the 90s and the encounter driven railroad that was common in the early 2000s. I know for me it was the latter that particularly drove me back to things like hex crawls and back to the original AD&D dungeon masters guide and other early RPG books to get a sense of what might have been thrown out with the bathwater for me.

When I run modern settings I tend not to think of them as much as sandbox (though I use many of the same techniques). When I run modern games I think more in terms of clash Bowley's idea of situational adventures and of power group based adventures. It does depend on the genre though. For a modern horror game it is pretty easy for me to apply that sandbox map approach but just with various things to investigate on a modern day map.
 
Can I ask what your definition of sandbox is (if you have one)?

Well, for me ... without thinking about it too much....I considered every RPG to be a "sandbox" - the best thing about a RPG as opposed to a computer game board game or choose-your-own-adventure was that the players could take any action. I've never railroaded anyone into anything other than maybe starting In Media Res. But that's just my style. I've enjoyed reading others interpretations and learned a lot.
 
There's a third: the GM's willingness to let go. If the players solve everything three hours early and have a walkover, then they do. If the direction the players go in invalidates your game prep, then it does.

This is a big thing for many GMs. Not only allowing players to bypass or flip your preparation on its head, but also allowing them to fall between a rock and a hard place, TPKs, and whatever problems they create on their own. Letting them run free can be hard at times but it makes for much more enjoyable games IMO.
 
Well, for me ... without thinking about it too much....I considered every RPG to be a "sandbox" - the best thing about a RPG as opposed to a computer game board game or choose-your-own-adventure was that the players could take any action. I've never railroaded anyone into anything other than maybe starting In Media Res. But that's just my style. I've enjoyed reading others interpretations and learned a lot.

One way to look at it, is sandbox is simply a style of play that leans heavily into the bolded, and avoids things that thwart the bolded. Also I think most people who run sandboxes aren't thinking in terms of this RPG being sandbox and this one not. It is more a playstyle than a game type (though some systems clearly could make running a sandbox more difficult or even impossible depending on the mechanics). Theoretically every RPG can be a sandbox. And like you the thing that leapt out at me when I first played an RPG was the sense that I could take any action. That was really powerful. I think people who run sandboxes, they are often trying to get back to that sense as best they can, and the sandbox structure is just one way to achieve that.
 
One way to look at it, is sandbox is simply a style of play that leans heavily into the bolded, and avoids things that thwart the bolded. Also I think most people who run sandboxes aren't thinking in terms of this RPG being sandbox and this one not. It is more a playstyle than a game type (though some systems clearly could make running a sandbox more difficult or even impossible depending on the mechanics). Theoretically every RPG can be a sandbox. And like you the thing that leapt out at me when I first played an RPG was the sense that I could take any action. That was really powerful. I think people who run sandboxes, they are often trying to get back to that sense as best they can, and the sandbox structure is just one way to achieve that.
Yeah, I understand that.

I mean, within the confines of the game world. So, we have some expectations on what can happen. But I remember playing with toys and I had similar rules. Like if my toy was a Batman figurine, I knew he had money and technology and I didn't think he could literally fly or lift a whale while swimming.

It's an interesting juxtaposition. My worlds have rules ....established in the narrative. Not arbitrary (like the wizards can't even pick up swords rule). So...this is why I think all RPGs are sandboxes. All of mine are.
 
It's an interesting juxtaposition. My worlds have rules ....established in the narrative. Not arbitrary (like the wizards can't even pick up swords rule). So...this is why I think all RPGs are sandboxes. All of mine are.

All can be. But many aren't. Many a scenario has foundered on the players being impelled to do things THAT way and no other -- that the scenario will hang up if the party can't get past that one door, and furthermore they need to lockpick it. (No, it can't be broken down, because, well, hrm, hum, it's made of stone. No, they can't bypass it, because it's the Only Way Through. No, they can't sweet talk/seduce/cajole/kill the keeper of the keys, because, well, hrm, hum, there isn't one. No, they can't go through the thin wooden ceiling, because, well, because fuck you, that's why ...) Mages can't physically pick up swords. That's the only NPC who knows anything about anything, because, well, reasons. The monster can only be defeated by hitting it in the left testicle with a magic mace, because, well, hrm, the official bestiary says so. The castle is impervious to any other plan of attack but a frontal assault. The Big Bad can only be overcome by killing him ... don't even try making him an offer of a chest of jewels to take an honorable retirement in Fiji. And so on.
 
All can be. But many aren't. Many a scenario has foundered on the players being impelled to do things THAT way and no other -- ...

That all sounds fairly horrible.

I admit, sometimes the players might need to be TOLD the ceiling is false. Or that they remember a person with keys. But I chalk that down to them being told about a world rather than experiencing it directly. And being lawyers, social workers, receptionists and IT guys for their day jobs and not FBI agents with a trained eye. So I'm very forgiving with "clues", maybe too much.

But if they can rationalise access to explosives? Sure. If they find a monster struck with a winch, we might be talking.
 
To me at its basic level sandbox is about creating a large setting, with locations, NPCs, groups, etc, that the players are free to explore however they wish. There is no planned adventure for the evening. Any adventure that does arise is a response to what the players do.

That is how I have always defined it. Within that basic definition, there are various ways of going about things, so there can be different styles of sandboxes, but ultimately a "sandbox" gets down to those fundamentals.
 
I had a conversation about this on reddit, but it left me unsatisfied. The debater was more interested in telling me THAT I was wrong rather than HOW I was wrong in my perception of RPGs and Sandboxes. He did say it was more environment-driven than plot-driven which, because I write fiction for part of my living, doesn't make a huge amount of sense (for me ....they're interlinked)

I plainly do not understand what people mean when they talk about sandbox TTRPGs. Because from my point of view, all TTRPGs are sandbox. You can be playing a hard-boiled detective in 1920a Chicago and the GM can throw superheroes at you. Or you can (especially in a narrative game), make up any possible outcome.

So what am I missing?
These days, I no longer think sandbox is a particularly useful term.

The key distinction for me is between relatively linear games and open games. The latter are approximately what is meant by sandbox in practical terms.

In a linear game the players go through a prepared plot or adenture. In an open game the enviroment is constructued so that the only real plot is one that emerges through interactions within the game. To do this effectively there needs to be enough hooks or moving parts that the possibilities are rather open and this also needs be structured so that it is communicated clearly to the players.

Not all games are sandboxes because not all games are structured in an open way. If an NPC comes to the PCs and offers them a job - it doesn't matter much if they have the freedom to refuse it if the players are under the impression that this is 'the adventure' and they'd be dicks to refuse it.
 
When I ran an X-Com game, the players received mission orders. They might be sent to investigate reported alien activity around a town, recover tech, kill and capture aliens at a UFO crash site or deploy in response to an alien terror attack. Between sessions, they managed the team cash budget and managed the organisation's overall research effort.

Within any given session, they had a reasonable degree of freedom to approach their current objective any way they wanted to, within the guidelines of their formal orders.

The campaign as a whole, however, would not fall under any reasonable definition of a sandbox. If the players wanted to defect to the alien side, spend half their time hiding recovered tech from their superiors and selling it on the black market, or outright refused to deploy on missions etc ... then the campaign would have ground to a halt -- I wasn't prepared to deal with those kinds of completely off-the-wall ideas, and wasn't interested in running a game about X-Com agents gone rogue.

In a true sandbox, players will be aware it's ok to do that kind of stuff if they want to, and that the GM is ready, willing and able to roll with it.

As TJS TJS indicated, one of the distinguishing features of a sandbox (or open game) is that the players aren't waiting for the GM to feed them the adventure for the session; they are expected to go looking for things to do, and when the GM does throw a plot hook them, there is no sense of obligation to latch on.

And, for what it's worth "sandbox" and "open game" just sound like synonyms to me. Sandbox still seems to be a very useful term. Certainly, when someone like Kevin Crawford offers the world some tools for running a sandbox, I have a pretty good idea what I'm going to be getting and what I'm likely to be able to do with them.
 
Last edited:
And, for what it's worth "sandbox" and "open game" just sound like synonyms to me. Sandbox still seems to be a very useful term. Certainly, when someone like Kevin Crawford offers the world some tools for running a sandbox, I have a pretty good idea what I'm going to be getting and what I'm likely to be able to do with them.
Perhaps. It's just I got extremely extremely tired with the defintions for sandbox which were being thrown around here which seemed to really exclude a whole lot of campaigns just because they hold to the expectation the players will keep to a broad approach to the game (ie be pirates in a pirates game). So I just decided "fuck it, I'll make my own term".

But also because so much discussion of sandboxes seems to be more about what the GM should not do.
 
Perhaps. It's just I got extremely extremely tired with the defintions for sandbox which were being thrown around here which seemed to really exclude a whole lot of campaigns just because they hold to the expectation the players will keep to a broad approach to the game (ie be pirates in a pirates game). So I just decided "fuck it, I'll make my own term".

But also because so much discussion of sandboxes seems to be more about what the GM should not do.

Well a term is worthless if it doesn't exclude things that don't fit the definition.
 
(scritches his head) Erm ... a term doesn't automatically become worthless because there are chowderheads who misdefine or misuse it. Heck, do we no longer use the term "role-playing game" because the marketing departments of console games decided to slap the tag on solo shoot-em-ups?

Use your own private neologism, no one knows what you're talking about.
 
(scritches his head) Erm ... a term doesn't automatically become worthless because there are chowderheads who misdefine or misuse it.

No, as long as one corrects the attempts to misdefine or misuse it before the useless version of the definition becomes common parlance.

The term already causes confusion online, and this largely seems to be, for reasons I can't possibly fathom, people insisting that the term has to apply to their games when it doesn't, as if they are missing out on some nominative prize or nonexistent "seal of approval". Not every game I run is a sandbox. Not every game needs to be. A Sandbox isn't good or bad or better or worse than any other playstyle, except by way of personal preference. But you still have folks decrying "how dare you call my game not a sandbox?" like they failed a term paper.

Best to shut that shit down immediately, lest the lowest common denominators co-opt yet another term and make it impossible to use in casual conversation (like "immersion").
 
Of course people should approach these issues in whatever way they most enjoy, and it shouldn't matter to you what other people agree or disagree with your preferences. But i think this issue can be interesting to talk about because a highly sand boxy style of play is so fundamental to the original purpose and early culture of table top roleplaying games, yet many people post 1980 or so (by which time many commercial products from many different companies encouraged a highly programed or structured style of play) don't have direct experience with it being done well and don't understand how to teach themselves how it can be done. And the disconnect between the two end members in approach to this topic have only gotten more extreme as more and more people enter table top gaming with prior experience playing computer fantasy roleplaying games, which almost universally present specific encounters, tasks and plot lines (i.e., even if you have some liberty to choose between a menu of them or walk around doing nothing, you can't easily create your own). That's the only reason i like to follow and weigh in on this subject - maybe some day a person will learn something and have a good experience that is new for them.
 
These terms I think a best understood as a spectrum rather than absolutes. So if you were to take a common roleplaying premise of occult Investigators, you could have:

Railroad: In the adventure we playing today you are investigating a haunted house. You will go through an exact, preplanned sequence of events and the only valid solutions to the adventure are the ones written in my notes.

Mission based: In the adventure we playing today you are investigating a haunted house. How you deal with the investigation and attempt to solve it is entirely up to you.

Directed sandbox: You are occult investigators. There are a whole bunch of haunted houses out there, which ones you choose to investigate and in what order is up to you.

Pure sandbox: There some haunted house out there, but if you want to start a rock band or open a car dealership instead, that's cool.
I use to play mission based with an occasional railroad adventure in between. Now I am doing a - I think - directed sandbox. Maybe the intention is a pure sandbox and the players tell me they want to do whatever they want in theory, but in practice they still want some pointers where to go and what to. I think they do want some clarity after all.
This certainly gives me new terms with which to describe how a game I run is likely to go. "Session 0 to make characters and establish Initial Context and discuss any Arbitrary Restrictions. It will begin as a Schrodinger's Sandbox, but as things progress I'll have a larger Bag of Stuff, and it will likely end up somewhere between Schrodinger's and a World in Motion."
I think I got the same. My aim is to have World in Motion, but for now I am mostly reacting to what the players are doing so I guess it's a Schrodinger's Sandbox.
 
A Sandbox isn't good or bad or better or worse than any other playstyle, except by way of personal preference. But you still have folks decrying "how dare you call my game not a sandbox?" like they failed a term paper.

Gods, yeah. The piece of cake came when in one of these forum debates someone made the assertion that GURPS was "more flexible" than D&D.

Now this, one would think, would be one of the more obvious no-shit-captain-obvious moments in RPG forum blather history. But, of course, right on cue, the fanboys erupted in rage, hotly denying that GURPS could be more anything than D&D.
 
(scritches his head) Erm ... a term doesn't automatically become worthless because there are chowderheads who misdefine or misuse it. Heck, do we no longer use the term "role-playing game" because the marketing departments of console games decided to slap the tag on solo shoot-em-ups?

Use your own private neologism, no one knows what you're talking about.
Did I not basically define the term at the same time as I used it?
 
Gods, yeah. The piece of cake came when in one of these forum debates someone made the assertion that GURPS was "more flexible" than D&D.

Now this, one would think, would be one of the more obvious no-shit-captain-obvious moments in RPG forum blather history. But, of course, right on cue, the fanboys erupted in rage, hotly denying that GURPS could be more anything than D&D.
Fun fact: I think you're referring to a thread I started, back at TBP. It was funny to watch from the side, probably...:devil:

Also, I don't have anything to add, which is a first in a sandbox thread:grin:!
 
The "they'd be dicks to refuse it" is the interesting part, I think.
I think these kinds of soft expectations are important.

First, you have the game where the GM says "I bought this adventure" and everyone agrees to play it and turns up. In such situation you would indeed I think be a dick to refuse the hook.

But you also get situations where the players feel, perhaps from experience, that although the GM may insist the hook can be refused (and may even claim to be running a sandbox) that if they don't accept the hook the GM will struggle; they haven't actually prepared an open structure, so if they refuse the hook, the game will flounder, and it won't actually be much fun for everyone. In which case you may feel you are doing the best thing by your GM and your fellow players to just go with it.

And of course, absent any prior communication (and possibly even if there was - actions speak louder than words), if an NPC comes an offers you a job, most players will probably assume that this is the type of game in which NPCs are going to come and give you the adventures.
 
Last edited:
Banner: The best cosmic horror & Cthulhu Mythos @ DriveThruRPG.com
Back
Top