Jun 4, 2009

Sandboxes are for kids and cat-turds. Railroads are for hobos and railcars.

Oh yes, he has been to your "sandbox" it was a good place for poops.

The sandbox. That iconic ideal we all strive for. The idea that the party should make their own decisions, and interact within an organic medium that reacts with their every action and choice. Let me tell you folks, it is over-rated as hell. Let me explain why.

First off, what is the antithesis of the sandbox? That's right, the railroad. That, to the uninitiated, is the literal "railroading" of the party to move along a pre-set course that the DM has charted out. This is prevalent in published modules and adventures that are usually light on any content not directly relating to the current adventure.

I am not aware of any other methods, if any of you would like to comment and point them out, I will edit or post a new article with them included.

First off, to quote a fellow blogger, "there is no spoon." We like to try and categorize ourselves like there are prizes awarded for most quirky, most old school, most new school, whatever. This is silly, and unfortunately very typical human behavior. GNS theory is a good example of this as well.

Truth is, no matter how much we try and pidgeonhole ourselves, we are so alike that any seppuku would serve only to take us all down. Back to the original topic!

The Simulationist Sandbox

Sandbox gaming is as far on the simulationist extreme as is possible to easily go. You are a member of a group of heroes(?) exploring an unknown area, interacting with it's inhabitants, and doing whatever in the hell they wish to do.

Where this really chaps my ass, is when the map or grid or whatever is being populated, the DM has to assume that a party that survives will "level up" and gain in power. Therefore, the encounters have to be truly random, spread out all over the place at all levels of play. This is effectively a double blind as the DM has no idea where the party will go, and the party has no idea what the DM has thrown down.

This fails in a couple of ways. The most obvious is the party encountering something that could de-flesh them with a mere thought. Does the party run away? - The dragon lazily flies over and nukes you from orbit. The giants are faster than you, etc. So the random direction you picked just killed you. How fun.

To those who would say, "but Donny, the dragon is full from eating a merchant caravan, and is willing to parley." I call Railroading! You have just metagamed yourself out of the sandbox. Unless that encounter was specifically designed around a dragon lounging on a bed of loot and skeletons - who just happend to be in a converational mood - you have failed. You had to change the parameters of the encounter to suit the story, commonly known as railroading.

The alternative to this is even worse. What do you get when you cross a "typical" 7th level party with a pack of kobolds? An extremely lackluster encounter. The kind of encounter where the highlight is the one rat that took TWO swings to drop. Yay, how awesome.

Randomness is just that, random. It is nobody's friend. It is nobody's mistress. It is a voluntary inclusion of chaos into a game defined by it's rules. It works, but too much randomness creates an incoherent game. For example, the party (3rd level we'll assume) moves north from the village.

Hex #c5 says, "A band of 2-12 orcs is terrorizing a farmhouse, it's residents still inside. The farmer has a magic sword, but doesn't know how to use it." The party has a tough time, but prevails.

Moving east, to hex #D5 says, "A pack of d4+1 dire weasels are eating berries off of a bush - the bush has 2D10 berries that will cure d4 HP each. The party manages to survive here as well.

Continuing on, they travel north again. Hex #D6 says, "Three hill giants are eating a woodsman they caught. They would love more meat." Guess what, the party dies. Wow, that was sure fun. Time for new characters.

Am I missing something? Are there actually people out there that LIKE this kind of game? Where you never know what is going on until you are suddenly dead? Here we had a total lose scenario - The giants are not only massively more powerful than the party, but also are faster than the party. What about an ochre jelly? It's mindless, it'll just keep following. A vampire? Another toasted party. How is this fun? Seriously, somebody let me know - I am at a loss.

It's not even about dying. THAT happens in D&D with clockwork regularity, it's a part of the game. However, a well designed encounter will ALWAYS have a way for the players to win. Before you start foaming at the mouth about the word win, I ask you who plays to lose? Any commenter who want to admit that they have secretly always played just to brag they've died or lost in every conceivable way will be trapped, tagged, and released with pity.

In my humble opinion. The sandbox fails because the very randomness it relys on to exist, is antithetical to an enjoyable game experience. This is not to say that randomness has no place by any means. Just that using randomness as a fudamental building block for a campaign is like playing russian roulette with your loved ones for giggles. "Gee dad, I thought this game was fun?" Thank you random troll pack encounter.

The Narrative Railroad

Railroad style gaming is Narrativism taken to the same ludicrous extreme. In this style, the DM reserves the right to do whatever it takes to "guide" you along a predetermined path to a specific event. Most game modules are good examples of this. They contain only information pertinent to the chain of events and encounters that lead from point A to point Z.

This is the method that currently gets all of the badmouthing. It's gonna get some more today. Railroading your party is a good way to have your party do stupid, silly, and otherwise idiotic behaviors - i.e. "Let's burn down the tavern!" or "I pick everyones pockets." They do this as a test of their boundaries. They know that the cage has bars, but cannot often see them. As such, they play with resignation.

This is why it is a BAD idea to allow a character in the party that is essential to the endgame of the campaign. Without them, the campaign collapses - and they will broom-handle you relentlessly once they figure that out!

As if that wasn't bad enough, when the party asks the DM where they should go - the DM is not doing their job. Either A. The information was not passed on properly. Or B. They honestly have no idea where your next set-piece event takes place, and don't want to piss you off.

The Gamist compromise

This is the game most of us actually play. Where the two circles come together into some common ground. This is where we can have a dragon encounter at level 5, and use it to further a story without the party feeling like well armed care bears. This is the game with an established plot line, random encounters, an A a Z and all points in between determined as needed. We get it ALL here. This is the game where your character is something other than a snack to a badly(?) placed encounter, or a wooden puppet marching to the last page of the module.

This is the game that gives you the tools to WIN! To accomplish the goals said character has set. This is the game where everyone is having fun. This is where the hex simply says, "Party level +2 encounter. Captive nymph in bad mood. May strip out of spite. No treasure." Is this old school? Is this new school? Does it matter? It's actually both. A winnable encounter, with a basic framework that (to me anyway) provides everything I need to have some good roleplaying as well! Is the nymph part of the encounter? Is it a trap? Does she kill the party out of annoyance? Who has her? All of these can be pencilled in on the fly, within the framework of an EL+2 encounter!

In my humble opinion, a lot of the "controversy" surrounding this leg of GNS theory has been the whole "win" thing. In fact, 4E was recipient to a lot of identical criticisms - lots of noise about every character winning everything or somesuch. I am not here to tell you how to game. Maybe your players are mongoloids that need to be herded. Maybe your group has a long standing bet relating to the total number of character deaths. I say that if you aren't in it to "win", what ARE you in it for? beer and pretzels? Funny looking dice? All the hot chicks? To die in an epically thematic way at the feet of the villain your party has spend the last 2 years of real world time trying to get to? To LOSE?

In my opinion, the biggest flaw in GNS is that the three exist in opposition to each other. Going in one direction distances you from the others. In my mind, they are all mishmashed together in a pile of nearly useless letters - until you form a group. At that point, you need to fish out the bits that will fit the group as a whole. A game "tailored" to 3 power gamers, one method actor, and two folks allergic to dice will look differently than, well...any other group on the planet - but I digress.

In summation, there is no one true way. The artificial divide between the "schools" of gaming thought is just that - artificial. We choose to build these walls around our narrow interpretations of the most subjective game on the planet. We choose to dig the moat, and how deep. We even choose when to fill it with water - and when to fill it with flaming trolls.

As to game design itself. Sometimes when you codify every minute detail into easily digested, modular little bites, you lose a little bit of perspective. D&D is not chess. D&D is not snakes and ladders. We get so pre-occupied with using formulae and rules that we forget there is likely not a single game ever published that is played RAW.

We all have different ideas as to how is "should" be. One person's Xanadu is another's steaming pile of poop. This summarizes the edition wars, Gaming theory and design, ans well as the silliness of the old school / new school movements. This is also the reason we have such a vibrant gaming community. Do we really need to build walls to feel better about our personal choices?

Long live gaming! And all of it's silly players.


Jonathan said...

Oh fer fucks sake... this is by far one of the best posts I've seen in a VERY VERY VERY long time.

Thank you for posting only when it's good. thank for being so kick ass. Solid - to the point ... dead on; RIGHT. I agree 1000%. Ok, maybe my two posts with posters poked too much fun at the whole stupid ass old school vs. new school torrent. But.. this post. BRAVO man... may this be in V2 of OGT.

Jonathan said...

I'm linking to this



can't wait for weekends.

Donny_the_DM said...

Thank you Jonathon :)

That means a lot coming from you. I hope there WILL be an OGT 2

I just wanted to say that I think everyone is wrong. But not for the reasons they think. While I may not necessarily be right, at least I will try to see the light :)

It's all intertwined, and the devil is always in the details.

Thanks :)

Tom said...


kaeosdad... said...

Hm, I didn't know the sandbox was all about simulation... I figured that even in a sandbox there could be possible adventure paths, but the point was to make a consistent open world for the players to roam in. I guess I misunderstood it, but it's still my first year of DMing. I figured why not use both?

Using a video game analogy I always thought that a well designed sandbox would play like an mmo without a budget and the digital limitations.

Paul Kasper said...

I think one of the best examples of the best of both worlds is the Savage Worlds Plot Point Campaigns. Basically, you have a set of main plot points that lead to a specific ending (railroading), but there a number of Savage Tales (mini adventures) that can be played at any time and at the whim of the players.

50 Fathoms and Necessary Evil are perfect examples of this approach. It's been said that these are a bitch to create, but they are a brilliant compromise.

Graham said...

@kaeosdad -

A well-designed "pure" sandbox should indeed resemble an MMO in certain ways, but not completely. For instance, there would be no distinct separation between the level 10 area and the level 20 area, and if it makes sense for a level 70 monster to be roaming about the level 10 area, it is.

But bear in mind that if you stumble into the wrong area in World of Warcraft and find the level 50 monsters at level 10, you die and respawn. In D&D, there is no respawn point, and you only survive if enough party members survive to resurrect you.

So if in a sandbox game, the players stumble into the grumpy, antisocial, murderous level 20 dragon's lair, you have two options:

1) everyone dies, or

2) alter the parameters of the encounter.

So strangely, the MMO, despite its limitations, ends up being more of a sandbox in many cases.

Brilliant post, Donny. Great stuff.

Anonymous said...

I play the sandbox style games you decry, and the "Gamist" approach would be boring and unsuitable for me as it still follows the 'Boss GM' and 'employee players', I don't like that. I find it spoils my spare time and them. Everyone at the table gets paid to be in employee v boss relationships, why would we sit around doing it for fun?

When there is sandbox, yes, you will die eventually (stupidly) if you just pick a direction and start walking.

This is akin to dropping down in a random town in the world with and wandering in any direction without getting any directions, asking locals, scouting ahead, spying from treetops or anything else to know where you are going before you get there.

That is know as leaping before you look.

If you run into a dragon..how did he spot you? Were you just tramping across an open field? Did you not pay attention to the smoke from burning villages?

If there are giants roaming around killing everyone...how did you get near them? Did you not notice the lack of other human habitation? If there are other human settlements are they not somehow fighting or avoiding the giants? did you not ask?

Ya, in a sandbox game sometimes people die. Sometimes you run out of luck and the giants find you. I've been in that situation, 4 of 6 party members died, 4 giants so we split in 6 directions. If we'd run into 6 giants we'd be dead. We took the risk cutting through giant territory and it failed. Now if the giant was just magically teleported in, then you've got a stupid GM.

Likewise if the giants are just ALWAYS in the hex never eating, never roaming away, never doing anything but waiting to be stumbled upon...

You have a railroad game, not a sandbox.

Sandbox games are not described in your article. You've described another "gamist' hybrid that merely combined the worst aspects of the Sandbox and Railroad, instead of the best aspects.

kaeosdad... said...

@graham: that dragon example is what I figure separates a bad sandbox setting from a good one. I've never been a fan of the "oh shit I picked the wrong door" type of deaths with no warning. Rolling up a lvl 20 dragon on a random encounter table when the lvl 3 party is adventuring through the plains is just bad design with no thought beforehand.

Rumors, information gathering and such should always be an option to give the players a good idea of what type of dangers they may face (in other words whats on the random charts for the area).

Just like a well designed encounter should always give an opportunity for the player characters to win a well designed sandbox should give the players fair opportunities to learn about their surroundings and enough clues to allow them to deduce what what they can and cannot handle.

Anonymous said...

1. GNS theory does not say what you think it says. Use GDS, where the D stands for dramatism. You are already using it, expect with one wrong word (narrativism).

2. There is a third way besides sandboxes and railroading. There GM creates a bunch on unpredictable and interesting situations where he does not know what the players will do and throws them at the characters at suitable moments, also improvising them on the fly. Difference from sandbox: It is story-oriented and the world is not quite as fixed. Difference from railroading: There are no rails; the players have genuine choice in what they do in the situations and each one takes to game to some direction; GM does not even try guessing where it will move.

3. In functional sandbox play it is possible to gather information ahead of time. Blindly stomping around means that either GM is doing it wrong or the players are failing.

Anonymous said...

Best thing I've read all year. Well done.

Helmsman said...

I always like your writing, you have a way of asking very obvious questions in a way that is enlightened at the same time. However, in this I can't say I'm in agreement with you.

Your assessment of sandbox seems to be from the perspective of someone who really hasn't ever played that style of game before. I run sandbox all the time and I can tell you that it's not even a little like how you describe.

Maybe I could see a sandobox game working how you're describing if a GM simply scattered encounters of a completely random sort all over a big map with no rhyme or reason, and then place the PC's with no background into this bizzaro world, then the PC's decided to close their eyes and spin in circles and then wander in the direction they were facing when they stopped... in that circumstance I see sandbox working as you describe, and yes that would be retarded.

But that is not how I roll. In my games things make sense, and there is a story, the PC's don't have to walk along a set of scripted checkpoints to make the story work. The story happens and they move through it and react to the things occuring around them, because mine are interactive worlds, not ones that simply sit and wait for the PC's to trigger events. And massively powerful things like Dragons should not be things that a group "happens upon" like tripping over a log or something. Dragons should be the center of 10 square miles of blasted scorched earth where no one goes because THERE'S A DRAGON THERE! When dragons go on the move kingdoms shake and kings start spending their fortunes to keep the dragon the hell away. Only the most blind stupid PC's would ignore the extremely obvious signs that this is not a place to go unless you really honestly have to, it would be obvious because the world worked in a way somewhat resembling reality where food comes from farms, kings rule because they have armies, and civilization exists where the monsters aren't because the two are incompatable with each other.

To me sandbox is about a world where the PC's can do what they wish, but it is a dynamic world where things happen with or without the PC's involvement, and the PC's should be able to effect and alter this world with their actions. The setting in-turn interacts back with the players, offering jobs, scenarios, assistance, admiration, lovers, allies and enemies as the game goes on.

This might be what you consider the middle ground, but to me this has always been sandbox. Not old school sandbox, not new school either to me this is the only proper way to run a game and when I'm a player I always assume that the game is this way.

Anonymous said...

One more thing.

4. Donny, you are assuming that there is lots of fighting. Unless the player characters are a bunch of bloodthirsty mercenaries, your objections with sandbox play, and the strengths of planning ahead, pretty much disappear.

Sham aka Dave said...

First of all, interesting read. You certainly have an effective writing style. Unfortunately I think the gist of the post ends up being that you can run a sand-box game poorly, and you can run a railroad game poorly. Sand-box does not equate to "pick a door and see what happens" unless the DM decides it does. Not running an encounter exactly how it is written does not equate to railroading, that's DM fiat. Honestly I don't think you've described anything remotely close to a game I've played in before, and I've played in more poorly run games than I'd care to remember.

Keosdad: Sand-box most definitely does not mean that there are no adventure paths. It means that the players are not forced to stay on a particular adventure path. Railroading means no matter what the players do, the campaign does not progress unless the players follow a predetermined adventure path. Sand-box is not pure simulation. In its highest form it is a setting that becomes defined by the actions of the players. It can begin with minimal description, and between sessions the DM can flesh out the areas that need it based upon what paths, angles or goals the players are following. It is a story not yet written. In a sand-box the players are the authors, and the DM is the editor. In a railroad game the DM is author and the players are characters in his story.

Donny, as I mentioned you have an effective writing style. Even though I disagree on your use of the terms and on the examples given, I did read the entire post and I do think there's a theme worth dissecting. Regrettably the theme is blurred by these incongruities.

Sham aka Dave said...

Oh, and Helmsman: Great friggin' comment. You said it better than I could have (and did).

Rob said...

Seconding the kudos to Helmsman. Well said, indeed.

A good sandbox has lots of ways for players to find out about most things long before they meet them. There may be a deadly deadly ghoul pack on the random encounter table in the Old Cemetery, but Old Sal the herbalist was there when it carried off Farmer Johann's son, and the town guards have posted flyers declaring the place off-limits. That's why old sandbox modules like Keep on the Borderlands all have those big rumor tables, and listings of what nearby secrets monsters do and don't know about.

Andy said...

I agree with you on most points, but I think you are pigeon holing the sandbox campaign a little bit. I've played in great sandbox games where there was a real chance of going to the wrong hex and being turned into adventuring party soup. The key for it to be a successful campaign is information. A stop at the nearby town and a round of ale for the townsfolk should tell you what major places to avoid and what kinds of wandering threats there are in an area. In a realistic game world, if there is a settlement nearby, its probably relatively safe to travel road. As long as the party is smart, and sticks to civilized lands while they are young upstarts they should be able to stick it out until they get to those higher levels. You might say that this kind of predictability is a form of railroading, but I'd say that because it is a realistic simulation of a world, you would be wrong.

Donny_the_DM said...

Holy crap! I go home for the day and the whole internet shows up to comment :)

Instead of individually responding here, I plan on posting a follow up article with responses this afternoon.

It would be now, but I've been recruited to do a menial task that very well may end up with me holding a bloody letter opener.

Thank you for the comments, even those that disagree - especially those that disagree :)

Anonymous said...

Great post - not sure I agree 100% but I enjoyed reading it. Have linked to it on my LJ.

kingworks said...

Two recent experiences I had:

A) DM'ing our last session: The PC's roll Nature/Dungeoneering and determine down path 1 lies Kruthiks (and probably no treasure); path 2 is mysterious and probably more plot-centric so they choose the later. I was prepared for the Kruthik encounter. I had to hastily re-draw the map and get set up for an encounter I didn't think we'd get to. Then they tried to cast two rituals after the encounter started . . . sigh. Maybe I should have let them run away back when the BBEG wanted to invade the town, instead of guilting them into being, y'know, heroes.

B) At the LGS, I ask when MM2 is being released. Some random guy comes up and says "4e sucks! They totally ruined it after 3.5e - that's the only thing we play." His group being comprised solely of classic RPG'er stereotypes and mine having two beautiful women as well as a professor and a lawyer in it, I decided that discretion was the better part of valor. Still irked me, though

Josh Roby said...

Congratulations! You have identified your preferred creative agenda. Now armed with the knowledge of what you enjoy, go forth and enjoy your gaming experiences.

...meanwhile, those of us who enjoy the "unplayable" other stuff will continue playing and enjoying the stuff you don't like. ;)

Xander Abbott said...

Hey Donny - That's some pretty great stuff you got there. But you did miss something...

As Gamers we do build walls, especially D&D Geeks. We divide ourselves into ways of thought and styles of play and build characters that reflect that. Division of thought is who we are, truly. The game remains the most imaginative game because of that division, and the guys who rage against it. I play Leader types, skilled with swords and shields and group tactics. I am my group's leader, I train in scale mail with my sword and shield alongside my group as a way to keep fit. And that is my point in a nutshell. If I had to warrant a guess I would say that you play instigators, Jonathon plays rogues and Direking plays leaders or healers. My friend who wanted to become a nurse never played a single game without being the healer. My artist buddy played really heavy Special FX combat guys. The division is for us a good thing. Where we live in the Fantasy Worlds we game in come to reflect in who we are. The point isn't to break down our divisions, but to accept them like the D&D heroes accept the different classes. We embrace our differences to build a good party, why not a good gamer community?
Xan Thunder Wolf