Railroading vs Sandboxing

Time to talk about real role playing games. You know, pen and paper stuff. It should be a nice break from the ongoing discussion of Oblivion and things related to it. How do you feel about railroading? It is an interesting topic because there is really no clear answer. If taken to the extreme levels it can be a game breaker. On the other hand, it is a valuable tool in GM’s toolkit that allows him to tell more compelling and tightly plotted stories.

In case you were to lazy to click on that link, let me explain. What is railroading? Imagine that the GM has prepared a very linear campaign which depends on a sequence of events unfolding in the exact timed sequence. The tightly scripted plot demands that the players are in the right place at the right time giving them very little wiggle room. If they would fail to appear at some crucial place, or talk to a crucial NPC the whole plot would fall apart. In order to keep things going, the GM essentially prods and guides players, trying to keep them on this narrow path. He essentially puts them on rails. They can only move forward in GM approved direction, and any attempts to diverge from it will be made impossible by a sudden avalanche, an emergency or a via NPC imploring the characters to stay the course in a very direct manner.

As you can imagine, if taken to the extreme, railroading may turn into a game of “guess what GM wants us to do today”. Players may begin to feel that they have no control over the unfolding events, and become apathetic about making any kind of decisions. Eventually they will cease all attempts to actively participate in the action, and just wait for the next big plot twist to be revealed to them.

A good GM try to mask his attempts at railroading while giving the players an illusion of control or modify his campaign on the fly to adapt to what the players are doing. Or both. As you can imagine, these things are difficult to do while running a game with a strong narrative arc. Sometimes you have to railroad to tell a good story.

Recently I stumbled onto an interesting idea that proposes to adopt a very different style of play. What if your campaign was completely non-linear, and open ended? What if there was no over-arching plot, and no direct goal the players would need to work toward? People love games such as Morrowind and Oblivion, so why not take a similar concept and bring it back to your gaming table?

This stuff works in single player video games, but would it really translate well when adopted at your gaming table? How would it work?

The same as usual – the players arrive to a location, they are broke, hungry and looking for work. They start asking around in the local tavern and they get few tips on how a group of adventurers could make money in the area. But instead of shoving a Very Important Mission™ down their throats why not make finding work a core element of the game play. Have them talk and develop relationships with local NPC’s and let them know that they are free to take on any assignments they want.

If they want to go clear out goblins from an old, abandoned property outside the city walls, and help the owner move there then roll with that. If they want to investigate the ruins to the south of town and see why there are strange lights at night appearing there, figure out a good story and encounter to go with it. Make small quests, and exploration rather than story to be the focus.

If you want a real life example of such a campaign, go check out the notes for West Marches Campaign. It’s a set of articles written about a sadbox game that was actually implemented and played through. It is a fascinating read, chock full of ideas and good suggestions. West Marches is set on a new frontier – a newly founded town at the edge of unexplored wilderness. PC’s are the only people crazy enough to venture into the uncharted west and discover ancient ruins, dungeons and hidden treasures. It shows that the sandbox model is workable and can be implemented successfully.

Then again, perhaps West Marches worked because of the unusual logistics behind it. The game was played in ad-hoc groups that shared maps, resources and information. The sharing and collaborative exploration aspect of the game seemed to be a big part of it’s allure. Perhaps it would have flopped in a more traditional setting with a stable group and regular meetings. Perhaps it’s open-ended style of play would be too unfocused, and directionless. Sometimes a little bit of railroading comes a long way. Take what you will from West Marches but I think that if you are planning to run a sandbox game it is best to aim for a hybrid. Combine the sandbox model with a more traditional, linear “quests” offering strong narrative and even inter-connected story arcs – just try not to shove them onto players. Don’t snap them onto rails – make them pick the threads of the story and follow them onn their own time. If they loose a thread, don’t worry about it. Make a note of it, and try to introduce it at a later time when it makes sense story wise.

Dungeon Mastering site has some tips on how to get a non-linear game going. Gnome Stew has even more including suggestion on including pre-written modules within your sandbox.

The census there seems to be that a GM should not go on a world building binge. The temptation to flesh out the setting, draw detailed maps and populate towns with well rounded NPC’s may sometimes backfire. But that is true for most campaigns. Anyone who has ever sat in the GM chair and did any kind of prep work for it is probably painfully aware of the world building pitfalls. It is a very alluring time sink that may either pay off big, or turn out to be a complete and utter waste of time. Guess which one happens more often?

Still, having some rough sketches and ideas and a rudimentary map is usually a good idea. If I was running a sandbox game I would probably make a heavy use of the stuff like Oracles.

What are Oracles? They are random generators that produce sort of seed concepts or plot hooks you can use to construct interesting adventures. It’s a little hard to explain, so just go look at the page and read couple of them.

They were originally created for the indy game In a Wicked Age. Before the session the GM would draw 4 cards from a deck to get his Oracle. This would yield a set of 4 vaguely described ideas, events or conflicts that he would then flesh out into the unfolding story. The idea took off, and people create Oracles for all sorts of other systems and settings. You should be able (or even to make) one for the system you are playing.

You could easily pre-generate one or two oracles for each notable town in your sandbox. You don’t even need to treat these things as fully fledged adventure seeds – unless the players decide to investigate them further. Treat them as generic current events that are going on in the background. If each town has some intriguing gossip it will appear as if you did much more prep work than just pressing refresh few times on a website that randomizes plot hooks.

Seriously, check these things out. The original oracles are a fucking goldmine of ideas and can usually be easily tweaked to fit into any fantasy setting.

What do you think? Have you ever played a sandbox game like that? Would you play one? Do you think it would work? Or would it quickly lose focus, and direction and become boring and disjointed? Let me know!

Update 01/28/2009 10:50:46 AM
Out of curiosity – how many people here play or played pen and paper RPG games? I have this suspicion that like 90% of people who read this blog regularly don’t, and couldn’t care less about this post.
This entry was posted in rpg and tabletop. Bookmark the permalink.



2 Responses to Railroading vs Sandboxing

  1. Jake UNITED STATES Google Chrome Mac OS says:

    I actually made something like Oracles once. It was pretty much exactly the same. Seeing Oracle is weird, it’s like what I would have done if I had pushed the idea further.
    Thanks for this post, I clicked on EVERY LINK. It was awesome.

    Reply  |  Quote
  2. Luke Maciak UNITED STATES Mozilla Firefox Windows Terminalist says:

    @ Jake:

    Thanks! It’s nice to know someone appreciates this particular post. :)

    Reply  |  Quote

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>