- Gnome Stew - https://gnomestew.com -

Nice Myth, Ugly Truth: Sandbox Games Are Better

I hear people brag about their sandbox games. About how the players can have their PCs interact with the world in an unrestricted manner. How the gameworld is not bogged down with a plot that railroads the players, but instead the PCs encounter unique self-contained events that the PCs may investigate further or walk away from at any time.

Every time I have played in a campaign advertised as a sandbox game the game itself was as boring as plain oatmeal. Yeah the PCs could do anything, but it sure would have been nice to have the gameworld offer something for the PCs to do and then apply a little pressure as incentive to do so.

I am not saying that a sandbox game cannot be fun. I am sure that there are such sandbox games. If the definition of a sandbox game is that the PCs may go anywhere within the gameworld then my own games are sandbox games. The problem is not that a sandbox game cannot be an amazing and fun game.

The problem is that some GMs believe that just by running a sandbox game that the sessions are going to be amazing fun.

Just because the PCs can go anywhere in the gameworld, and interact with anything that they encounter does not make the game fun. You still need to provide incentive to the PCs to have a fun game. You have to offer a key component of any game: opposition.

A game just is not a game without some form of a challenge. You need an opposing force for the PCs to encounter and overcome (or even lose to). If my PC can easily board a ship and sail anywhere in the world in your sandbox game, but in GM X’s gameworld the plot requires that I sneak onto a pirate ship in order to infiltrate the Dark Overlord’s kingdom I am ditching the sandbox for the plot. One is a road trip, the other is an adventure!

I like the idea of a completely interactive gameworld. I appreciate not having to move from room to room in order to complete an adventure. But do not label the guy running a dungeon crawl adventure module as being an inferior GM for not running a sandbox game (yes, I witnessed this recently). Having a sandbox game does not guarantee that the game will be fun, and having a linear plot does not mean it cannot be fun.

That is what I think about the matter. What about you? Leave your comments below to share with myself and others. And remember that the GM is a player too, so have fun with it!

48 Comments (Open | Close)

48 Comments To "Nice Myth, Ugly Truth: Sandbox Games Are Better"

#1 Comment By kenmarable On August 25, 2009 @ 10:29 am

Agreed. I’ve never played in an extreme sandbox game but making the players the sole motivators of plot doesn’t sound like a good idea at all unless you have very highly motivated players with very clear goals of their own. In my experience, players often wander around unsure of what to do with a clear incentive from the GM.

However, I wonder if most “sandbox” games aren’t that extreme. For example, I find it works nicely to hybrid it and have more adventure hooks than time for adventuring. Spread those hooks around the world and you are pretty close to a sandbox. Yes, it’s a bunch of largely predefined adventures, but I let the players choose which adventures to actually pursue where (with perhaps adventures that look more fun having hooks and some variations at several locations).

So if they go to Pirate Cove I have adventure hooks for A, B, and C. If they instead travel to the Sultan’s Palace, I have adventure hooks for D, E, and a desert variant of B. If the PCs show interest in a joining a certain guild, then I drop hooks for G, C, and E. As long as I have the first session’s worth of material ready for all of these (or some random encounters strongly flavored to the location), then I can readily handle whatever interests the players.

In a sense it’s Choose Your Own Adventure rather than Write Your Own Novel.

#2 Comment By ChrowX On August 25, 2009 @ 11:20 am

I’ve played a couple games where the Sandbox element turned the game into a total mess. Unfortunately, those were also our longest running games and we spent a great deal of time getting ourselves killed and other players doing nothing.

Now, the biggest problem here is that the game was Changeling, which means that nothing is safe and, if the ST feels like it, there can be true Fae and loyalists anywhere. So, while some players like myself decide to go deal with some of the troubles plaguing the city, like the Hedge trees that strangle people to death, the other half of the group is going to Denny’s (in game) and talking about sex with the player who thought it would be fun to play a character with the mental capacity of a 5 year old.

The biggest issue that comes from sandbox games is that they end up emphasizing too much real-world logic and spit in the face of group cohesion. Why should players have to cooperate and help each other when they can all just sit at a bar and wait for the plot to come to them?

In the end, that game closed with most player characters being killed, captured, or not even showing up to play. The ST was playing a different game than the group was and thought he was being clever by letting the Big Bad gather up absurd power and resources in the few games that we weren’t constantly trying to kill him. Even in the games that we did try to do that, they made the challenge impossible, so I’m not sure if you could blame that on sandbox stupidity or bad storytelling.

#3 Comment By DNAphil On August 25, 2009 @ 11:41 am

@kenmarable: I am with you. I use a hybrid system where I create many potential story hooks and seed them into the game. Then I let the players pick which hook they find interesting and I then write that out and run it for them.

Typically those are 3-4 sessions long. As we get into the final sessions, I poll my players on what they want to do next. Their decision, drives what I will write next.

By doing this, the players are in control of the direction of the story, and at the same time I can provide them a structured story, that they were not railroaded into.

I think that many GM’s still need to do prep (Me included) and cannot improvise a whole game, and that a hybrid approach give the players narrative freedom, and the GM the ability to write and prep a story arc.

#4 Comment By samhaine On August 25, 2009 @ 11:47 am

A good sandbox game requires the GM to be good at improvising, keep a consistent and potentially extensive mental model of the game world in his head at all times, and have a keen empathy for playing up what the players find fun and switching out what they find boring or frustrating.

These are not basic GMing skills, and some of them may not even be achievable without extensive play experience. And even with a practiced GM, conflicting player actions and agendas can derail the experience.

Some sandbox games will be awesome, if that’s what the GM enjoys running. Others, probably most, will be lackluster.

#5 Comment By blalien On August 25, 2009 @ 11:50 am

I’ve been in a couple sandbox games, and they all bored me to tears. Part of it is my own personal preference. If I go too many minutes without killing something, I get twitchy. Incidentally, my favorite campaigns have always been one-shot mission based. Another part is that I’ve spent the last six years on a college campus. The DM who runs a sandbox game is an ARTIST, and my mere philistine mind cannot comprehend his greatness. These games tend to have no combat, and then what’s the point of playing D&D?

But the real problem is that sandbox games tend to consist of following one player’s personal agenda. And the rest of us have to follow along, because, you know, we’re the party. If another player wants to do something, the dominant personality will just refuse to go along with it. So the game essentially boils down to one or two players, with the rest being backup support. A pure sandbox game could work with a group of mature adults who aren’t motivated by pride, but again, college campus.

I could see a sandbox-lite game working. As in, the party has to collect the Twelve Mystical MacGuffins. But the party can track them down in any order. Once they devote themselves to a particular MacGuffin, the game narrows down to a more traditional mission or dungeon crawl. In other words, a sandbox game in which the party has a clearly defined objective, but the path to the goal is nonlinear.

#6 Comment By Zig On August 25, 2009 @ 12:19 pm

I’ve never run a full on sandbox game, but I’ve done a bit of hybridizing as others have mentioned above.

In my campaigns I try to have one major story arc and intersperse shorter arcs. I try to give the PCs some leeway on what arc they will follow at times, but not all the time. Sometimes I have something big planned and the players are somewhat railroaded to do something or other based on the situation they find themselves in.

In D&D I think the times I come closest to sandboxing is when the PC’s wind up in a city. There is lots to do in a city and many, many NPCs to interact with. I tend to let the players branch off in different directions. One may want to check on the local Thieves’ Guild, while another wants to hunt down some lead about something that happened in the past (like the new location of an enemy) and yet another player might want to simply travel about and get a taste of the city.

In Shadowrun I wind up with a more sandbox style of play mainly because travel and communication is so easy. The players can go all over Seattle, but still link up rather rapidly and communication is simply a phone call away. The players tend to go off in all directions to solve the current problem or plan and execute the next run. I have to stay on top of what they might do. However, with my group it’s not too difficult to do that. I know each of their styles of play as I’ve gamed with these friends for 19-30 years.

@kenmarable — I really like the idea of having several story hooks interspersed all over giving the players a lot of freedom in what they do. I think I’m going to try incorporating that idea into my own campaign. Thanks for the insightful post.

#7 Comment By Zig On August 25, 2009 @ 12:21 pm

Oh, and I should have added that I have played in some sandboxed campaigns. One of my friends has run Rifts that way. Typically the campaign begins to flounder because the players don’t have a direction to move in. It also tends to get a bit chaotic as each player starts doing his or her own thing ruining the team dynamics. I have appreciated the lack of railroading, but some signs along the road would have been nice.

#8 Comment By NeonElf On August 25, 2009 @ 12:51 pm

I too have questioned this since I’ve read in several GM blogs about the “sandbox” concept.

I think the point of GMing is to find a balance between the two extremes that both GM and players enjoy. I personally see a purely sandbox game like life: Yes you can go anywhere and do anything and interact with a constantly changing world, but it’s hard to have epic dramatic moments.

Books, Movies, and TV shows are all scripted for a reason. Drama doesn’t “just happen” very often. It’s hard to have recurring villains, and a sense of accomplishment when you could just walk away if you feel you’re going to lose the encounter. Also how can you save the world (a common theme in epic campaigns) in a sandbox game?

I keep thinking if I wanted a sandbox style game, I’d play a Computer RPG. They offer a detailed world where events may be going on around me but not involved with me. I can go anywhere interact with anyone….. Leave the world minding to the computer, I’ll take a plot with drama and action in just the right mix any day over that.

A good GM knows that plot does not equal railroad, and even a bad GM could end up railroading players in a “sandbox” once they hit on an encounter, or story arc…..

#9 Comment By Patrick Benson On August 25, 2009 @ 1:51 pm

Good points across the board.

The issue isn’t that the game is a sandbox game, the issue is that no GM should assume that a sandbox game is the better type of game.

I improvise a lot when running games. I’ve improvised entire campaigns and my group has enjoyed them very much. Yes the PCs can go anywhere in the gameworld, but I can still have an opposing force that threatens the PCs regardless of where they go. I don’t mean it follows them around, although it might, but that is just gathering more power and influence within the gameworld if it is not dealt with. The PCs cannot just run away from it.

Note: A good GM will try to interpret why the plot is being avoided in his or her game. Often it is just a matter of the players not feeling that their PCs are adequate to address the threat being presented. You can address this in many ways. Sometimes the plot is just not interesting. In that case just ditch it and start a new plot around whatever the players find interesting in the gameworld.

The number one reason I am given for why sandbox games are better is that there is no railroading because there is no plot in a pure sandbox. It isn’t railroading to say “Dark Overlord’s army is taking over the continent.” and then see how the players deal with the plot. Maybe they raise their own army, or assasinate Dark Overlord, or even sink the continent to “contain” the threat. That is plot, and it can create drama, but it isn’t railroading.

Good GMs will realize that a sandbox game doesn’t equal instant fun, but bad GMs seems to consider that style of game to be foolproof. It really isn’t, and every GM nees to keep that in mind when running sandbox games.

#10 Comment By valadil On August 25, 2009 @ 1:52 pm

Agreed, wholeheartedly.

The game I aim to run is a sandbox with rollercoasters. Players have the freedom to go where they want whenever they want. But there are rails the players can ride if they so desire. Sometimes a plot will catch them unexpected.

I think one of the dangers in the totally freeform sandbox of a game is that the PCs are the only instigators. It’s all well and good that they have freedom and motivation, but there should be NPCs with just as much influence over the sandbox. The game world has to be alive all the time – it can’t just switch on when the players are looking at it.

#11 Comment By Kurt “Telas” Schneider On August 25, 2009 @ 2:41 pm

I really hate to join a chorus, but I completely agree with y’all. The true sandbox game is not inherently better than the average game, and stands a good chance of being much worse.

The Real Life equivalent of the sandbox game is sitting around and asking each other: “So, what do you want to do tonight?” Bo-ring!

[1] – Exactly. IME, many sandbox games boil down to “What does the most stubborn/loudmouthed/insensitive player want to do tonight?”

The above opinions are not a justification of “Plot-On-Rails”. If we players are just going to play audience, I’d rather watch a movie. The special effects and dialogue are just so much better.

#12 Comment By crowofpyke On August 25, 2009 @ 3:18 pm

I can’t stand the truly open sandbox campaign. Blech. Completely boring. I want a story to interact with not utter randomness of whim.

I weave a story with the players, not in spite of them. I have an overarching campaign plot line (which using the 5×5 method to flush out the outline has been great). We move through the story arc at a steady pace – careful enough that the players don’t feel they’re on a railroad, but strong enough that they feel engaged and are having fun.

I admit that I have incorporated/stolen/appropriated ideas from other modules, books and adventures either published or made by others. When I do use something like that I alter it quite a bit so it fits the campaign world and change things around so they feel integrated into the story as well. It is far from a “form rote” experience, but not an open sandbox either.

The real peril I see in a sandbox is this: theme and flavor. Sandbox chaos has no flavor – it tastes like sand. Give me a story, a theme if you will, an actual plot for gods-sakes. And far from restrictive, plots and themes keep the game flowing, resolving and moving forward… like an actual story. If you don’t like the theme and plot… well, guess what there are TONS of them to explore, try a new one. Just because there is a theme and plot does not mean the players are getting on a railroad – it means the game has some substance more than just random acts of player whim.

Point in the end I suppose that Sandbox games make me want to hurl because they are too open and random. Give me a game with a theme, story, and plot any day.

#13 Comment By Scott Martin On August 25, 2009 @ 3:55 pm

Truly open sandboxes still interest me; [2] is open but works despite lacking an overall plot. A sandbox should also work if the players are motivated and at cross purposes, like a sterotypical Amber Throne War.

Similarly, PBeMs often grant the players power enough to create plots for each other– often as a side effect of their own efforts. [3] games were an excellent example of everyone adjusting to eachother’s pushes and shoves interestingly.

Something, though, has to step up and provide the driving narrative force. I suspect it would take very clearly designed characters to do the lifting on their own– most of the time it’ll take interesting characters on both sides of a conflict to make it interesting. If the GM doesn’t have interesting and motivated characters working to do things that must be stopped, there’ll be a lot of wandering around looking for something to do. That goes double if the players are used to looking for hooks supplied by the GM, not in providing lots of hooks for the GM to work with.

#14 Comment By BryanB On August 25, 2009 @ 5:00 pm

A GM can not start a game and let the players create the adventure on their own. That path will likely be full of fail.

What a GM can do is provide an opening hook, something that sucks the players into a story. The players can even ditch the idea of following the path set in front of them, but the story is in motion one way or another. If the players act, X happens. If the players do nothing, X will happen. But something will happen…

It is nice to have pro-active players though. Many times, those types of players will help shape the story as it unfolds. A sandbox setting should be malleable and ever-changing. It should feel “real.” Fill the world with NPCs that have individual motivations and they will eventually be antagonists or allies to the PCs.

It is the conflict of goals or motivations that fuel the best roleplaying fires. Conflict breeds excitement and excitement is what makes this hobby so much fun. The best games are not linear or open. They are a blend of both styles and they are PC focused, if not PC driven.

#15 Comment By whateley23 On August 25, 2009 @ 5:52 pm

it’s interesting to me, what people are calling “hybrid” is pretty much what i think of when i think of “sandbox”. does anyone really try to create a sandbox without any story hooks? that seems absurd to me.

to my way of thinking, based on my experience of running and playing Traveller, the majority of the world should have a main hook, like Traveller’s default merchant campaign. on top of that fairly mechanical game, there are also Patrons who show up and give the players the opportunity to engage in less directly mercantile pursuits. this is, in my way of thinking, the same as those who say that locations should have various story hooks and there should be an overall story arc, but i would consider that to be a “sandbox” game.

to give an idea of what strikes me as a “sandbox” gaming environment: what i’d like to see more of is location-based “adventures” (really, collections of adventure seeds placed in a larger context), similar to the old city modules from Judges Guild, or Chaosium’s Pavis, where a location is described along with a number of potential story hooks, such as personality npcs with particular goals in mind (Traveller’s Patrons, potential antagonists to possible player schemes – for instance, the security team for a bank – and the like). such an “adventure” would have extensive notes on the politics and relationships of the npcs, the schemes in place and developing of at least the major movers, and so on.

#16 Comment By Patrick Benson On August 25, 2009 @ 9:29 pm

[4] – As you can tell from some of the comments not all sandbox games are equal. I’m sure that there are lots of great sandbox games that follow your description, but I have been sorely disappointed by those that I have played in.

Story hooks may be in a bad sandbox game, but I have found those story hooks to lead into simple and short adventures. A good sandbox game would do what you have described, and would include an overall sense that the PCs are making some form of impact upon the gameworld as a whole. Some sandbox GMs just don’t pull that off though, and they need to understand why a sandbox game can still fail.

#17 Comment By whateley23 On August 25, 2009 @ 11:00 pm

[5] – “A good sandbox game would… include an overall sense that the PCs are making some form of impact upon the gameworld as a whole.”

good point. this is something so central to the idea of sandbox play as i understand it that i hadn’t even considered that it might not be so.

“[S]andbox GMs… need to understand why a sandbox game can still fail.”

i suppose that’s so. it is certainly worthwhile to examine what can make a sandbox succeed or fail (plot-driven games can look to the extensive body of literary and film criticism for ideas on this issue). but, i don’t agree with the premise that “sandbox games are better” is a “myth”. given two games of otherwise equal quality, i would prefer the sandbox (and would even if the quality were somewhat less, but that is due to my personal preferences and philosophies) over the plot-driven one. i prefer rpgs to reading novels or watching movies (though i love novels and movies, don’t get me wrong), and i think that many of the strengths of the rpg lie in its sandbox potentials.

#18 Comment By blalien On August 26, 2009 @ 1:27 am

whateley23, I have never personally seen a campaign that goes, “Okay, you’ve all rolled up your characters? Go nuts!” What I have seen, however, is, “Sir Doomsalot is causing havoc all over the land. Go nuts!” And then we spend the next six sessions arguing over how we’re going to defeat Sir Doomsalot, or why we actually want to do so. And any plan that the DM didn’t account for beforehand will arbitrarily not work.

#19 Comment By whateley23 On August 26, 2009 @ 2:39 am

[6] – that’s what i’m saying (except for that last – if the gm didn’t account for it, then she’d better be thinking on her feet as to what the bigbad is going to do as a countermeasure). i’d also add that the players might decide that Sir Doomsalot has some good ideas, and hitch their wagon to his star.

what i’d argue against is (to exaggerate for effect) the plot-driven game where scenes are plotted out by the gm in advance, and the characters are going to participate in those scenes regardless of the players’ desires. in other words, traditional storytelling doesn’t work in rpgs as well as the new storytelling allowed by the rpg medium.

a great game, in this sense (and to pick one which i’ve grown to really like), might be Dogs in the Vineyard, which is really about how the players and their characters approach a sandbox situation, attempt to resolve its inherent conflicts, and maybe leave the gameworld a little better. or else learn something interesting about motivations and such.

although, to pull back a bit, if you’re having fun, then it’s good. if you’re not, then it isn’t. and those are the only rules that matter, in the end.

#20 Comment By kenmarable On August 26, 2009 @ 7:01 am

So maybe the proper metaphor isn’t just “Here’s a box of sand, have fun!”, but “Here’s a box of sand, some shovels and buckets, now build a castle!”

The toys provided and the overall purpose are as important as having a big box of sand. As long as the GM doesn’t dictate where the castle will be built or what it would look like or prevent players from bringing their own toys, then it’s all good.

Plus, it seems for many people the best games tend to be in middle of the sandbox-railroad spectrum. The biggest difference is just which side of that middle they skew. Whether it is 55%/45% or 70/30, it is much more likely to be a more enjoyable game than something that is 90/10 in either sandbox or railroad direction. But of course, epically awesome GM and players can make ANY game fun – no matter the system or bring pure railroad or sandbox. Most of us are just in somewhat awesome groups, however. 🙂

#21 Comment By Kurt “Telas” Schneider On August 26, 2009 @ 7:46 am

Like most gaming buzzwords, “sandbox” has a different definition for every gamer. But the pure form of the sandbox game is often held up like some bright, shiny grail that is somehow “more pure” than all other approaches. (Cue angelic chorus) “This is a sandbox game, and I am a better GM for running it.”

The reality is that except for some statistical outliers and the occasional VLARP, the true sandbox game simply does not work. It either turns into a pissing contest between alpha-players or a “I dunno, what do you want to do next?” snooze-fest.

Sure, there are sandbox-ish techniques that work, such as taking your cues from the PCs, or reacting flexibly and intelligently to an unexpected PC action, but this does not make a sandbox any more than having some kind of plot makes a railroad.

#22 Comment By LordVreeg On August 26, 2009 @ 10:03 am

Wow. Take a few days of reading, and see what you miss?

Not only a worthwile OP, but great responses. But a lot of questions have been raised and I feel like I’ll be retilling some soil…
Ah, well, no harm in trying.

I run a 25 year old Sandbox game. I did not call it a Sandbox game in ’84 when I transitioned from the campaign before it, I was just ready to graduate on a bigger, Divset world (Diverse setting, as opposed to a Conset, or single/main concept setting). Currenlty playing 2 Live groups and one online group.

A few opinions and Observations, based on this perspective.

1) A sandbox Setting is a setting in motion. Things are happening with or without the players. And the players will react to this. A sandbox game bogs if the PCs ever feel like it only moves based on themselves. One of the hardest things a GM has to do is have large and small storylines and events that lend versimilitude to a Sandbox game if it is to ever work. I see a lot of conversation here about them being boring or about the need for plot, but little about the underlying idea of a Sandbox game, which is, “The world is all here and moving, what interests you? What part of the world do you want to effect.”
So the GM must copnstantly give the impression of a vibrant, moving world around the players. However, to ‘reward and motivate’ the PCs, as was mentioned, they HAVE TO have an effect on this constant weaving of events. You have a river of plotlines and events, and the players have to be able to se where they are damming and diverting the flow.
It is also the way to surprise them. It is not a snooze fest if the players go to an Island and start a war with the local thieve’s Guild in the capital, only to find out the reason it was easier than they thought is that the Master thief had been away taking part in a coup and that the guy who ran the guild they were trashing now was the power behind the throne in the whole island.

2) Divset vs Conset
It is vital, in any Sandbox world, to have some overarching plotlines and themes zooming overhead. It has been said before, but it is worth repeating. This helps that feeling of the world not being in place for the players to romp in, that it would be there with or without the players.
However, Conset worlds make poorer Sandboxes. A game with a single overarching plot/reason for existence (such as LotR) is less of a sandbox, as unless the players latch onto plots that affect the ‘big overarching’, their actions will have less meaning (see motivation in point #1).
A divset world, with many major plotlines and themes, by contrast, allows the players that important luxury of choice that makes a Sandbox game what it is.
It is not direct, pure railroading to have a pretty normal world with (To use Patrick’s example) “the Dark Overlord’s army is taking over the continent”, but if that is the only overarching, big-picture plotline, then the Sandbox game is less of a Sandbox and has a lower chance of success becasue you have still set up a situation where the players still have to respond to that one plotline or be somewhat inconsequential.
I see this mistake a lot (and see it mentioned in many responses here), and while it is fine and maybe even preferred in most types of games, it is a mistake in a long-term sandbox.

3) Another hard one that will cause a lot of controversy, but Sandbox games with a high amount of social interplay have a much better longevity than ones without. I know this sort of flies in the face of the ‘West Marches’ stuff, but often those campaigns are not long term.

4) I could go on and on about this (and I might, later, but I’m at work), but I still really call my setting a 90% sandbox. This is for my own motivational purposes, but I agree that having the ‘players write the adventure’ is the path to fail. The GM has to be excited about the game as well, and so when the players totally avoid all the cool stuff we create, we are not motivated as well. In a good Sandbox, however, the GM has aot have a LOT of adventures made up ahead of time.

#23 Comment By Patrick Benson On August 26, 2009 @ 11:13 am

Good points all, but I want to make something clear about big overarching plots in response to [7] comment. LotR really doesn’t have a big overarching plot. At the core it is just a hobbit taking a ring to a volcano. That’s it. Once ring is in volcano the story is done.

Your sandbox game should not have that kind of a finite plot. It is not open ended, and it is incredibly inflexible.

WWII is a great real life event that you can create an overarching plot from. The Axis is against the Allies. Many different forms of antagonists and protagonists exist across several types of settings. The war could have ended in several different ways. The end of the war only marked the beginning of a new stage in history. No one event started or ended it.

That is what an overarching plot really is, and it is IMO what LordVreeg is describing. There are plenty of possible adventures in such a setting. The PCs can be part of something large (WWII) through smaller events (The Manhattan Project, or sinking the Bismark).

So when you have the “Dark Overlord’s army taking over the continent” as a major plot in your sandbox is it LotR or WWII? Good sandbox games have a GM who understand the difference, and does not rely on the idea that the freedom in a sandbox game is what provides the fun. A good GM gives the players that freedom in order to interact with the fun parts of the gameworld.

#24 Comment By ben robbins On August 26, 2009 @ 12:33 pm

@ LordVreeg

“3) Another hard one that will cause a lot of controversy, but Sandbox games with a high amount of social interplay have a much better longevity than ones without. I know this sort of flies in the face of the ‘West Marches’ stuff, but often those campaigns are not long term.”

[2] was extremely social, probably more so than the next five games put together, because all the players were cross-comparing their adventures, jostling to be in different groups, etc. Tons of roleplaying.

As far as using plots in sandboxes: [8]

#25 Comment By LordVreeg On August 26, 2009 @ 1:05 pm

I read (and commented, I think) on the West Marches set of articles on ‘Ars’.
Differet type of social interplay, my friend. And I will take the blame for being unclear. And I am not justvsaying that, I think I was unclear (posing at work again….).
I was speaking of ‘within game’ social interplay. And why I was willing to say it is controversial. Game and setting design wise, I think combat-heavy games, where every session involves multiple combats and the system used has most of the rules based on combat, become a little onerous in terms of Sandbox challenges. How the players interact with and affect the governments, politicians, artistic, and other populations in a game have a lot to do with the long-term immersion level of the game.

I did NOT mean to say the players, GM, or others were less social or that the roleplaying was any worse in a West Marches game.

#26 Comment By ben robbins On August 26, 2009 @ 1:25 pm

@ LordVreeg

Oh that’s definitely true: the West Marches charter was exploration and danger, not societal impact.

As a counter data point, I ran another “player initiated action” game before West Marches called Promised Land that was entirely national/cultural level — a societal impact sandbox basically. It was much less successful, but there were a lot of confounding reasons so a direct comparison isn’t so easy. Personally I think the societal impact sandbox sounds like lots of fun.

Another big point that I don’t think’s been addressed in this discussion is that a single group sandbox is an entirely different beast than a multiple group sandbox. It’s a critical distinction.

#27 Comment By Starvosk On August 26, 2009 @ 1:26 pm

Problem is that sandboxes try to tap into player motivations. Problem is players tend to generate characters that are realistic.

Most people aren’t very motivated to do anything. PCs are an extension of the players, and quite frankly, your players are happy to sit around eating sandwiches pretending. More than likely, their PC’s motivations will reflect this.

A leader is necessary to motivate the PCs into doing something, and if there isn’t one, good luck. Your sandbox game will wash away.

But yeah, more often than not sandbox is a synonym for lazy or indecisive.

A living breathing world should eat the players alive if they decide to do nothing. Popular archetypes with no motivations and nothing to lose (like the lone ranger or the street bum) will bomb a game faster than you can say splat.

#28 Comment By Patrick Benson On August 26, 2009 @ 2:42 pm

[10] – “Another big point that I don’t think’s been addressed in this discussion is that a single group sandbox is an entirely different beast than a multiple group sandbox. It’s a critical distinction.”

I think that is another article all by itself. You can have multiple groups in a non-sandbox game as well, and the problems with multiple group games go well beyond whether or not the game is a sandbox game IMO.

[11] – “But yeah, more often than not sandbox is a synonym for lazy or indecisive.”

Is it really? I think that it is a style of game with a lot of merit, but like any buzzword people adopt the label for their game without putting any thought behind it. Like people who call themselves environmentalists because they bought a Prius yet they own jet skis and a huge truck to tow them around with. Sandbox games do not equal bad games, or lazy GMs, they just do not ensure that the game is going to be fun.

#29 Comment By whateley23 On August 26, 2009 @ 2:58 pm

[11] – “Most people aren’t very motivated to do anything. PCs are an extension of the players, and quite frankly, your players are happy to sit around eating sandwiches pretending. More than likely, their PC’s motivations will reflect this.”

that’s not an issue with my gaming group. while some of the players tend toward this mode, the rest do not, and the active ones tend to inspire the passive ones. perhaps i’m lucky with the group of people i call friends, but i think that it just takes some discussion of what the players hope to get from the game and some pointed questions from the gm about character motivations to get the ball rolling. once it starts, it tends to keep going.

plus, anyone who makes mundane motivations for adventuring characters is doing it wrong. so your character wants to farm his land and have kids? then why is he sitting in this tavern with a bunch of other disreputable types and listening to the old wizard with the map to the dungeon? oh, he needs money so that he can buy a farm and get married… motivation solved. now the player and his character have reasons to act on the world (and maybe those motivations will change over time – once they get a taste of city life, you can’t keep ’em down on the farm, as they say).

#30 Comment By whateley23 On August 26, 2009 @ 3:02 pm

[12] – “Sandbox games do not equal bad games, or lazy GMs, they just do not ensure that the game is going to be fun.”

well, that’s true. i still say that, other things being equal, a sandbox game is better than a plot-driven one. after all, plot-driven doesn’t equal a bad game, either, but it also doesn’t ensure that a game will be fun.

#31 Comment By Patrick Benson On August 26, 2009 @ 3:35 pm

[13] – “i still say that, other things being equal, a sandbox game is better than a plot-driven one. after all, plot-driven doesn’t equal a bad game, either, but it also doesn’t ensure that a game will be fun.”

The problem there is that “other things being equal” and even how to measure which game is “better” are purely subjective measurements. The same GM and group that has a great time running “Tomb of Horrors” might miss the cheese of that old module when running a sandbox game.

So forget “X is better” entirely. GMing styles are like works of art. Despite advancements in techniques and technology the Mona Lisa is still a masterpiece, and the new world of digital multimedia art is just as intriguing. You can’t really compare the two at all, because it all comes down to the individual looking at the options. It is a purely subjective comparison.

Never think that adopting a particular style will make your game fun. Work on your GMing skills and know why certain styles work better for you to begin with. Don’t assume that one style is superior over another, but rather know which style is better for you.

#32 Comment By whateley23 On August 26, 2009 @ 4:55 pm

[14] – or, as i said above, “if you’re having fun, then it’s good. if you’re not, then it isn’t. and those are the only rules that matter, in the end”. i’m saying that the premise that “sandbox games are better” is a “nice myth” which has a converse “ugly truth” is just as mistaken as “all sandbox games are awesome and everyone should always play that way, regardless of the gm’s and players’ strengths and weaknesses”.

for me, sandbox is a style that i actively pursue because of my philosophy of gaming and life. some people prefer to have their characters pushed around by the whims of the world/the gm/fate, while i prefer to let the characters emerge from their stochastic environment (which may result in them being pushed around, whether that’s by the dice, the environment, or the gm characters, but equally allows the characters the opportunity to be actors on that environment – it’s a matter of the strength of character presented).

#33 Comment By Robert On August 26, 2009 @ 8:13 pm

Running a sandbox entails creating a world. Creating a world includes potential antagonists. A sandbox without (potential) antagonists is at best a boring world and a worst incomplete. To try to define sandbox as a game without antagonists strikes me as setting up a strawman.

Now, whether those antagonists were pre-designed or improvised—possibly inspired by some random rolls—doesn’t really matter.

#34 Comment By Patrick Benson On August 26, 2009 @ 9:34 pm

[15] – Offering antagonists is one thing. Every bad sandbox game that I played in had antagonists. But note that in the article that I said that you need to provide opposition.

In a Superman comic a mugger can be an antagonist. For a brief moment Superman must deal with the mugger, and then it is done with. But when the mugger is given atomic armor by Lex Luthor who is also funding several other wannabe super villains to keep Superman occupied while Lex works towards his ultimate goal we have opposition.

The difference is that an antagonist who cannot hope to beat Superman does not really offer an opposing force, but against Lex Luthor it is possible for Superman to lose. They are in a strange way equals despite being opposites (hence opposing forces).

Like I said in the article, if a sandbox game is defined by the amount of freedom given to the PCs to explore the game world then I run a sandbox game myself. I don’t fool myself into thinking that the sandbox quality is what make the game fun by itself. You still need an opposing force to the PCs in that gameworld in one form or many.

#35 Comment By whateley23 On August 26, 2009 @ 9:50 pm

i just realized one other thing that was bothering me about this blog entry. why are you implying that a dungeon crawl is not a sandbox? look at two of the prototypical dungeon crawls, In Search of the Unknown and Keep on the Borderlands. both of those are clearly sandbox environments, designed around locations instead of gm-centered plots. the players are given a general problem and allowed to solve it (or ignore it!) in any way they desire.

what definition of “sandbox play” are people using which would exclude such classic dungeon crawls?

#36 Comment By whateley23 On August 26, 2009 @ 10:01 pm

[16] – “he difference is that an antagonist who cannot hope to beat Superman does not really offer an opposing force, but against Lex Luthor it is possible for Superman to lose. They are in a strange way equals despite being opposites (hence opposing forces).”

that’s an issue of balance, not of sandbox vs. plot. whatever his exact characteristics, Lex Luthor is a story hook. so is the mugger. the only difference is that one is appropriate to Superman and the other isn’t. in a sandbox world, both the mugger and Lex exist, but Supes is going to spend his real efforts countering Lex because nobody else can do it.

similarly, in a fantasy game, the Keystone Bandits might challenge a beginning group, but as the characters grow in power the players are going to want to deal with Duke Evald the Malicious instead of spending their time on an issue that could be handled by local beginning adventurer talent (even if those adventurers are hypothetical – meaning that the gm should generally leave such “low-level” encounters aside as already dealt with by random npc adventurers). conversely, at the start of their careers, they aren’t going to be a credible threat to the Duke.

#37 Comment By Kurt “Telas” Schneider On August 26, 2009 @ 10:26 pm

[17] – Y’all are hilarious. Two examples of non-linear dungeons mean that ‘dungeon crawl’ can’t mean ‘linear adventure’?

Define your terms, please. Sandbox games mean that the players can do what they want without impetus from the GM. In other words, Superman ignores Lex Luthor so he can put the moves on Lana Lang because the player is more interested in the social aspect of the game, or maybe he’s just a perv. But if Lex clobbers him, how is that any different from the railroad of “you’re supposed to beat up Lex and his atomic armored muggers”?

#38 Comment By whateley23 On August 26, 2009 @ 11:55 pm

[18] – first, i’ll leave aside the unnecessary comment with which you lead off, except to ask why you think that’s a necessary way to conduct this discussion.

“Two examples of non-linear dungeons mean that ‘dungeon crawl’ can’t mean ‘linear adventure’?”

no, two classic examples of sandbox dungeon crawls are an indication that the opposition of “dungeon crawl” and “sandbox” is an oxymoron.

“But if Lex clobbers him, how is that any different from the railroad of “you’re supposed to beat up Lex and his atomic armored muggers”?”

in a sandbox world, Lex’s actions in relation to Superman are seen as consequences of Superman’s actions. since Supes is pursuing Lana, and that (for instance) interferes with Lex’s plans for her, then something happens. if Superman wanted to deal with no consequences, then he wouldn’t do anything, but that isn’t very heroic or interesting, is it, and why is Superman’s player playing at all? and, anyway, there are consequences to doing nothing, as well. alternately, Lex has no interest in Lana and pursues his own goals off-camera, as it were, without interference from Superman, becoming President and bungling the handling of Hurricane Kallisto in Mississippi or whatever background events the gm has in mind. meanwhile, all of the people playing learn something interesting about the person playing Superman and his vision for that character.

#39 Comment By Kurt “Telas” Schneider On August 27, 2009 @ 7:22 am

[19] – I’m sorry I offended you. I did not intend that first comment to be aimed specifically at you. You have my apologies.

I wrote the entire comment, put the cursor after the “Y’all are hilarious.” line and clicked the “Reply” link on your post, which pops your name at the top of my comment, and not where the cursor was. I then hit “Submit” without proofing it. I should know how this comment thingy works, but I apparently confused it with another one.

My intention was to say that the entire discussion is funny because it is way off track from the original post, and that it’s very typical of gamers in general to get into heated discussions without actually defining their terms. Almost every time I’ve seen a heated debate like this, it could have been avoided had the participants clearly defined their terms.

For instance, despite 38 comments and the original article, “Sandbox” still does not have a clear definition.

#40 Comment By Robert On August 27, 2009 @ 7:52 am


OK. Where I wrote “antagonist”, read “opposition”.

And Luthor goes after Superman either because of Superman’s rep or because Superman has gotten in his way before. If Superman never fought crime and just woos Lois, he’s not even going to be on Luthor’s radar. Although, that analogy seems really out-of-place since a Superman game isn’t really going to be a sandbox.

Even if you’re randomly rolling all encounters, the PCs are going to eventually run up against a potential Luthor. If the ref never follows through on such potential or prevents them from even showing up, that doesn’t make it a sandbox. That makes it a poor sandbox. ^_^

I don’t know that clear definitions are important, but I am beginning to think the discussion has been kind of pointless. I’m finding that the games I run fall on a continuum between sandbox and rails. There’s fun to be had at every point along the spectrum, and there is FAIL to be had at every point too. I’m not interested in clearly defining sandbox. I’m interested in figuring out what changes in approach should be considered as we move from one area on that continuum to another.

#41 Comment By Patrick Benson On August 27, 2009 @ 8:32 am

I have enjoyed reading everyone’s comments, but my I still hold to the point of the article:

“Just because you run a sandbox game does not automatically mean that it will be fun.”

Now for some of you that might be a “Well, duh!” statement. That’s fine. For some people it isn’t, and I try to write articles that address beginning GMs as well as those who use more advanced techniques.

Saying a sandbox game is better is like saying vanilla ice cream is the best flavor. You can’t prove it (and yes the same is true for saying any type of game is better). It all comes down to how people feel about the game during and after playing it.

The point of this article is that if you are a GM running a sandbox game are you offering opposition to the PCs in the gameworld? If not, chances are your sandbox game is less fun than a linear plotted dungeon crawl (and yes dungeon crawls may be sandboxes, and I never said that they couldn’t be nor implied otherwise).

A sandbox game, like any other style of game, is not a silver bullet to having fun. It is a technique, and you need to develop that technique. The point of the article was not to fool yourself into thinking that just giving the players the freedom to go anywhere in the gameworld and interact with anything in the gameworld equals fun. Can you improvise a good encounter? Does your RPG system make on the fly combat easy? Do you know how to introduce a memorable NPC, or can you introduce a simple plot that can work itself out after the PCs abandon it?

But I’ve said all that I want to say on this matter. I hope that the article and all of its comments provoke some interesting ideas amongst our readers, and I’m happy to see that it got so many responses.

#42 Comment By ben robbins On August 27, 2009 @ 11:26 am

The derogatory opening tone of the original post (“nice myth, ugly truth”, “I hear people brag about their sandbox games”, “boring as plain oatmeal”) made it a lot less likely that productive discussion would result. It’s a good way to get a lot of comments (and start an argument on the internet) but not shed light on a topic. You’ve buried your point in the noise.

#43 Comment By whateley23 On August 27, 2009 @ 2:33 pm

[20] – “despite 38 comments and the original article, “Sandbox” still does not have a clear definition.”

i thought that i was pretty clear in several comments above, but i’ll try another explanation. a “sandbox” game is one in which gm plots are secondary to player/character actions, and in which the characters are the primary focus. this is in opposition to a “story-driven” or “plot-driven” game, in which the gm plots take precedence (and in an extreme case is nothing more than the domain of a frustrated novelist). obviously, this is a continuum, but generally any game in which the plots are secondary to character desires (and development, etc) is on the sandbox end.

on another topic you raise, if you think that this is anything close to “heated”, you haven’t been on the internet very long.

[21] – as i’ve noted, we agree on that. we don’t agree on the premise with which you titled this piece, nor on several points in it.

certainly, like all things, it’s a matter of taste. however, it is possible to discuss taste. we are allowed to say why we like one thing over another. we just can’t argue when someone rejects those reasons for themselves. for instance, i could (and did) say that i prefer sandboxes for various philosophical reasons regarding the human condition, and that, to me, they are therefore better than an equivalent plot-driven game. i also noted that, for this reason, i tend to seek out sandbox games over plot-driven ones. you, on the other hand, have indicated that you prefer plot-driven games because the sandbox games in which you’ve played have been “boring as plain oatmeal”. fair enough, though i think that’s likely a gm/player group problem rather than a problem with sandbox games.

for myself, if what i wanted was plots, then i’d read a book or watch a movie. for the social interaction, i’d organize movie- or sports-watching parties or play some kind of sport (i do all of those, actually, but that’s beside the point). roleplaying has its own strengths and weaknesses, and shouldn’t be bound arbitrarily by the limitations of other media, in my opinion. other people want a more close translation from book or screen to game table, and that’s their right. give me the games that couldn’t exist in a story collection.

[22] – i don’t know about that. this discussion seems to me to have been fairly productive. among other things, the comments have caused Patrick Benson to clarify his message and back away from those original derogatory statements.

#44 Comment By Patrick Benson On August 27, 2009 @ 2:37 pm

[23] – “among other things, the comments have caused Patrick Benson to clarify his message and back away from those original derogatory statements.”

I wasn’t going to comment anymore, but this I want to clarify: I stand by the original statements. I do not feel that they were derogatory. If they offended people that is unfortunate, but I still stand by the article as written.

#45 Comment By whateley23 On August 27, 2009 @ 2:56 pm

[24] – that’s interesting. the tone i get from you in the comments is different, less confrontational, than the tone i get from the article. i interpreted that as backing off of some of the more contentious (would you prefer that to “derogatory”? it’s an overstatement to say that offense was taken, just as it would be an overstatement to imply that the article was offensive) statements.

but now things have gone “meta”, which can only mean that this conversation has little left to sustain it.

#46 Comment By Patrick Benson On August 27, 2009 @ 3:46 pm

[25] – I stand by the article. Sandbox games can be great. I said so in the original article, and even said that my game is can be called a sandbox game. Sandbox is a technique and not a guarantee for fun. I’ve said that from the beginning.

I have been in some very crappy sandbox games. The statement “Sandbox games are better.” just is not true. I never in the article blamed the technique for the games being crappy, I said that GMs must not expect the sandbox technique to make the game fun.

If my tone is derogatory to others I did not intend it to be, but I cannot control how my tone is perceived either. I am not going to say that I was wrong if I don’t honestly believe that to be the case.

I did not want others to think that I had changed my mind, because I haven’t. That is why I made the comment that I did in response to yours. I still believe in what I wrote. I wanted that to be perfectly clear.

I also will not say that others should not be offended by what I wrote. That is always a personal matter.

There is a saying in Russia a friend from Moscow once shared with me that translates to “If you want to be the good guy, be prepared to be the bad guy.” When I share my opinion I know that it will cause mixed reactions, and that is the consequence of sharing them. Better to deal with that consequence then not to be heard at all.

I think that the comments that you and others have made, whether for or against the article, all have merit. We’ve given others lot to consider. They will judge it and reach their own conclusions. That is all that I want any of my articles to accomplish. No one has to agree with me, some comments may even change my mind, but if I still believe in what I wrote my job is to defend it as best I can.

I just don’t see what else can be said on this topic.

#47 Comment By LordVreeg On August 27, 2009 @ 3:55 pm

Well, if we classify ‘productive’ in terms of high-level conversation, this thread has been good. Some very experienced GMs have found it worth posting about.

I also think Kurt’s Mini definition fits pretty well. I may feel a need to come up with lots of ideas what makes such a game more likely to succeed or not, but his quick definition is good.

@Patrick, you do make a lot of statements in the OP that are are at least challenging. You mention you do not label the guy running the dungeon module as an inferior GM just because he is NOT running a Sandbox game. A GM running a module should not be judged less than another GM running the Sandbox game.

However, Can both guys run a ‘canned’ adventure? Sure. Can both guys create and run the Sandbox? Maybe not. You can’t label one as better watching one game, but is there more to being a GM in creating a full setting and running it well VS running someone else’s module? I believe so.

This actually does tie in to a point you made about the automatic assumption about a Sandbox being more fun. It’s actually the opposite.

A real Sandbox game is harder to run than something more linear. Harder. Harder to keep track of, harder to create, harder to keep moving.
So all things being equal, by my reckoning, since it is harder to run a good Sandbox game, they are more of a challenge to run, and there are probably MORE bad sandbox games out there than other varieties.

#48 Comment By whateley23 On August 27, 2009 @ 4:22 pm

[26] – “We’ve given others lot to consider. They will judge it and reach their own conclusions. That is all that I want any of my articles to accomplish. No one has to agree with me, some comments may even change my mind, but if I still believe in what I wrote my job is to defend it as best I can.”

i think that this paragraph should be the goal of every internet discussion. so, i’d say that we were certainly successful, and that this has been a very productive discussion. thank you for the very civilized debate.

#49 Pingback By Open-Ended Campaigns: Sandbox Trouble | Moebius Adventures On September 14, 2009 @ 10:02 pm

[…] For another take on sandbox games, check out Gnome Stew here. […]

#50 Pingback By Is Too Much Player Agency a Bad Thing? | On September 23, 2014 @ 8:41 pm

[…] of articles out there warning of the pitfalls of poorly done sandboxing. Gnome Stew points out that without a strong opposition, sandbox campaigns can feel boring. They even have a post “In Defense of Railroading.” Reddit and StackExchange each have […]