A big problem with many role playing games is that the player characters can make their own decisions and then take actions based upon those decisions. Another problem with RPGs is that the game master may want to control the scene and direct the game session towards a predetermined plot point.
If you think I am a complete moron after reading those two sentences I don’t blame you. It hurt my brain to write them.
Yet it is not that hard to find a game where the GM is dictating the scene so that the PCs are forced to take a certain action (commonly referred to as railroading), and it is just as easy to find a game where the PCs can take any actions regardless of how they hinder the game as a whole (let us refer to this as off-roading so that I can stick with my analogy).
The funny thing is that the GM being able to control where the game is going and the PCs being able to to follow their interests are both good things to have in your game. And despite appearing to be in conflict with each other, when your group focuses on using PC actions and GM storytelling as complimentary tools you usually end up with one very kick ass game.
You need to address both the GM’s and the players’ needs in these situations. Between the railroading and the off-roading approaches lies the multi-lane highway. The multi-lane highway approach is when the GM has a destination for the game, but the players get to choose what lane takes them there.
Let’s take a common scene from many fantasy games where the PCs are assigned a quest by the NPC king. We’ll see how the scene may play out with the railroading approach, the off-roading approach, and the multi-lane highway approach.
Here is the basic scene.
The Scene
GM: Your characters are brought before the king. From his throne he says “There are orc raiders pillaging the villages in the Southern parts of my kingdom. Will you brave souls go and vanquish them from our lands?”
Player: Man, I was hoping that we would go check out that port on the map instead. “Your majesty, my comrades and I must refuse. Please accept our apologies as we must attend to business at the port.
And here are some possible scenarios using each approach.
Railroading
GM: The king leans forward and in a very serious tone says “I’m afraid that the port has been closed for repairs. There is another port to the South, and since the only road that takes you there passes by the villages where the orc raids are occurring, you should be able to take this quest.”
Player: I guess we are going South then…
Here the GM has made it clear that the players don’t have a choice. What is of interest to the players is blatantly disregarded so that the GM’s plans are stuck with.
Off-Roading
GM: The king grimaces, but then speaks. “Fair heroes, by all means attend to your business at port first. When you have completed your tasks return to my court, and we may negotiate payment for your services for taking up this quest.”
Player: “Your majesty is most kind.” We head to port and pay a charter to get out of this kingdom. I want to see what else there is in this world.
GM: I planned a scenario based upon the party fighting the orcs in the South though.
Player: Dude! Let me play my character, okay?
Here the players are obviously uninterested in the GM’s scenario. While this is a valid problem itself, the approach of just ignoring the intended plot in game is not a very effective way to deal with the problem. The GM is a player too, and forcing the GM to scrap a prepped scenario is the same as the GM ignoring the player’s character sheet.
Multi-lane Highway
GM: The king seems displeased but then speaks. “Fair heroes, from port one of my finest ships sets sail to bring supplies to our forces in the South. Attend to your business in the port, and once your tasks are completed seek out my ship. I will pay you handsomely for your services.”
Player: I don’t know. I really don’t feel like fighting orcs again. I was hoping for something with pirates and sea monsters. More nautical in theme.
GM: I see, but I didn’t prep anything like that. I can throw in some encounters while you are out at sea on the ship though. Also, write down some notes of what you want to see in the game and I’ll try to work them into future sessions.
Player: Sounds good. Let’s go check out the port then, and when we’re done we’ll go to the king’s ship.
In this scenario the GM doesn’t just disregard the player’s interest in the port, but instead the GM makes it a possible path for the plot to continue moving forward with. The player in turn explains why the plot isn’t appealing and offers an alternative that is appealing. The GM plans to adjust the session with those suggestions, but doesn’t have to scrap the material that has been prepped for this particular session.
This is how you build a multi-lane highway, by having the GM take player input and then applying that input to the game sessions. It also helps to let the players know that if they work with you as a GM that you will shape the game according to their suggestions. The GM from the example might now decide to change a non-critical detail in order to make the current scenario more appealing to the players (a prepped orc raid may now become a raid by pirates against the king’s ship). The plot of the game can still be moved forward with some slight changes to incorporate the players’ interests.
That is my opinion on the matter, so what is yours? Leave your comments for others to read and share your own experiences with me and other members of the GnomeStew community. And no matter what happens, don’t forget that the GM is a player too! Have fun with it!
Personally, I don’t make railroads or multi-lane highways, because as a GM I don’t see my job as coming up with a “story” to “tell” to my players; I see “story” in RPGs as being very much something that happens as a by-product of the interactions between the players and the GM, and if the players don’t want to follow a particular route I won’t force them.
Here’s the ways I managed to run a game in this manner successfully in my recent REIGN campaign:
– Think in scenarios, not scenes. Rather than thinking “I want the players to confront the Dark Lord in the shadow of his tower – now, how do I get them there?”, I think “OK, one of the elements in the campaign world is going to be a Dark Lord with an expansionist agenda”. The players will probably get involved one way or another – precisely how they choose to get involved is, of course, down to them. The robustness of the scenario-based approach is that while a scene-based approach falls over if the players never get to the really cool scene you’ve built the adventure around – or if you find yourself forced to railroad or cajole the players into getting there – whereas the scenario-based approach throws up dozens of opportunities to quickly improvise scenes as and when the opportunity arises.
– Make sure the campaign world is well-populated with potential scenarios. The players don’t really have a choice if their options are a) fight the Dark Lord and b) sit around on their asses.
– Don’t over-prep. If you’ve minutely prepared an adventure where the PCs go fight a bunch of orcs, you’re obviously going to be reluctant to scrap it. I actually found it easiest if the players decided what their next course of action was going to be at the end of each session, as opposed to at the beginning, so I could then prepare accordingly.
– Spend more prep time on the activities of NPCs and NPC factions as opposed to coming up with pre-scripted events. REIGN is particularly good for that – I could have nations battling it out in the background, generating all kinds of scenario opportunities in the process, just as easily as I could track the interactions between two noble families in a single city.
That is a valid approach that works well with some systems, but in my opinion may not work well with all systems. I personally prefer to improvise whole sessions and your approach works very well in that context. Yet having a predetermined plot isn’t a bad way to GM with some games.
It always comes down to what is it that your group wants out of the game. For some groups a dungeon crawl is a great way to spend a Saturday evening, and in that case having a story just accentuates what it is that the group is striving for.
I’d also say that your approach is a multi-lane highway approach. The prep work that you are doing is just adding many more lanes to choose from. Yet perhaps the key difference is that you don’t have a final destination in mind. This is a good way to run games with strong groups, but I wouldn’t use it with a newly formed group. I just find it works better to give new groups a clear agenda.
I’ll write more about it later, but I’ve never seen a true ‘sandbox’ game really work. The successful ones I hear about still have some kind of agreed-upon plot or theme.
Now if somebody were to come up with “Seinfeld, the RPG”…
I run a fairly sandboxed game, and my key is to asking the players ahead of time what they want. I usually seed them with 5 or 6 choices over e-mail and ask them to decide (OOC) what they want to do. Based on what they decide, I make plans for the next session. Then, when we get together for the game, we run through the IC conversations and play the game based upon the plans. Presto! I’m prepared for exactly what happens! (Well, not exactly, since players always do nutzy stuff, but close enough…)
During the session I throw out hooks and subplots, but none are very fleshed out. Then afterwards, I ask again what they want to follow up on, and make plans accordingly. Rinse and repeat.
The downside is that a lot of game decisions are agreed upon beforehand OOC, but it saves me prep time and makes sure that the players get to pursue the goals they want.
As some have suggested, player input before hand and agreeing upon a theme is essential. Below are the steps I take to avoid railroading:
1 – Use an in depth campaign setting, one that defines most aspects of the world.
2 – Know that setting well.
3 – Ask your players what they want before the session, and build a common starting point.
4 – From one session to another, the players let the GM know where they want to go, and the group builds a dynamic campaign in that fashion.
I GM a group of 6 players once a month and I guess I “semi-railroad” the session in that there is always an Objective or Goal at the outset. This keeps the players and game moving in the right (read: required) direction. The players are free to devise their own way to get from Start to Finish, but their actions are very very rarely unexpected/out of the ordinary. So I guess they’re either well trained or very accommodating. Either way, they seem to enjoy the simple structure of my GM style = Introduction – Body – Conclusion.
I’d also say that your approach is a multi-lane highway approach. The prep work that you are doing is just adding many more lanes to choose from. Yet perhaps the key difference is that you don’t have a final destination in mind. This is a good way to run games with strong groups, but I wouldn’t use it with a newly formed group. I just find it works better to give new groups a clear agenda.
Ah, but the group in the campaign I mentioned did have a clear agenda – they were the new rulers of a city-state, having swept to power in a popular revolution, and had the task of guiding it through choppy diplomatic waters to its rightful place on the world stage. The key thing was that I hadn’t decided in advance precisely where that place would be – would they be Swiss-style neutrals, using a combination of diplomacy and military strength to convince everyone to leave them alone? Would they be a force for good in the world, slapping down the evil empires? Would they become an evil empire themselves? Would they come up with a fourth option that surprised me? Any of these things were possible, and I was entirely happy for any of them to come about.
I’ll be interested to see what Kurt writes about this, but I strongly disagree with the idea that a commonly-agreed theme for a game makes it not a true “sandbox”. Of course you come up with some rationale for why the party are hanging out together and what they are trying to achieve – I find nothing more boring than a game where a bunch of strangers hang out waiting for fate (in the guise of the GM) to tell them what to do (a railroaded campaign) – or worse, stumble around aimlessly until they eventually find a plot strand that they’re interested in (a directionless sandbox). I would make a distinction between “directionless sandboxes” and “directed sandboxes” as opposed to “true sandboxes” and “false sandboxes”; in the directed sandbox, you still have all the player freedom that you do in the directionless one, you just have a particular agenda that the PCs want to advance. And crucially, as the GM you don’t have a stake in how that agenda pans out.
(In addition, I don’t like the approach apparently taken by many sandbox GMs that the players should always actively seek out the action. It’s my sandbox too, and if as the GM I decide that a particular NPC faction is going to meddle in the PCs’ affairs, well, they’re going to, not because I have a personal stake in seeing that faction screwing the PCs over, or because it’s important for some pre-arranged arc, but simply because it’s a logical consequence of the PCs’ actions, and I want to avoid the impression that the PCs are kicking about in a vacuum).
I like the post, good advice for the play style it’s designed to apply to.
I also agree with the “scenario” over “scene” technique.
Regarding sandbox style games, Ben Robbins has run a very successful campaign called “West Marches”. You can read more about it here: http://arsludi.lamemage.com/index.php/78/grand-experiments-west-marches/
It certainly seems like a lot of work, but it shows it can be done!
“The GM is a player too, and forcing the GM to scrap a prepped scenario is the same as the GM ignoring the player’s character sheet.”
i like that
This is a really good gamemaster tips article. I like the folksy tone and the examples really open up the issue. Railroading it one of the most common gamemastering ‘mistakes’ on the books and his treatment is a fair assessment and gives great advice for beginner GMs. I have a somewhat different take on the issue though.
Contrary to what a few respondents said, I don’t think the multiple-lane highway actually requires any more prep. I believe it is a powerful example of better improvisational techniques for maintaining game ‘flow’. It’s very workable and very well thought out. I like the way the author expresses his technique.
However, I don’t see any noticeable difference between multi-lane roads and railroading except for the amount of finesse. (Which is a necessary art; it’s good to hear sound advice on that.)
The essence of what we are really discussing is the value of the players’ input. Or more correctly, how much the players believe they ‘make a difference’. The ultimate complaint about railroading is that, in the end, the players don’t feel like they’ve really accomplished anything, being only spectators of play. In many occasions, advice on gamemastering works out to various methods of ‘don’t get caught’. Which is nothing new, but still vitally important for that technique. (There is nothing inherently wrong with railroading, except the rare case where it breaks down. It is, by far, the most popular gamemastering style.)
You see, I’m the guy who first explored the ‘No Myth’ gamemastering style (This link is an excellent summation of what I have written). For those without time to read the other article, ‘No Myth’ means that, of everything done outside of play or that has not been said in-play yet, not one bit of it exists. From there, instead of railroading, you can move the planned conflicts ‘in front of’ the players. There is no ‘southern kingdom’ (nor even a port city) unless someone has mentioned it.
So, in the example given, with ‘No Myth’, you have the orc attack in mind, but does it have to be ‘in the south’? or ‘of the kingdom’? or even are they ‘orcs’? I’d say that you hit the brick wall of railroading the minute you say ‘your characters are…’ or perhaps ‘…brought….’ If the players are familiar with being called upon by the king, these statements don’t really cause problems, but as the example plays out, it reads a little heavy-handed. (In other words, don’t tell the players that the world happens to their characters, let them make the decisions.)
Alternatively, as the gamemaster, you could start with ‘okay, when we left off, you were at the capitol. What’s next?’ According to the example, they would say they ‘go check out the port’. So? Why drag the king into it? Just have the orcs are menacing the port. Is that too off-beat for how the group expect the milieu to be? Okay, there are pirates attacking the port who happen to be orcs…or are pirates that happen to have the same stats as those orcs (if it’s the orc design-work you fear wasting) or maybe they are half-orcs.
The real question should be ‘what do I accomplish with the orc attack?’
From there, you can change any detail of the attack to suit. Put the scene of the attack in the place the players chose to go. Adjust a few details to better suit the expectations of how the game world works (such as ‘orcs wouldn’t be at the port’). And even better, shortcut to the situation, exactly where the players want to go, where the action is already happening. Then let them sort it out. Since dice aren’t involved in ‘No Myth’ gamemastering, there is no ‘fudging’. In the end, you run the awesome battle content you created, make the players feel like they ‘just walked into it’. And there is no railroad.
On the railroad, the next action would also be scripted, but instead, the players can choose to go or do whatever they like. They feel like they are in control because they are. You feel like you are the storyteller because you do. The conflict here is false; never existed, never could’ve. A story is a structure which creates additional relevance to a series of events. By structuring the details the players generate, you remain the storyteller, but not ‘the writer’ (since you don’t create those details).
The responses to this article all suggest really great points, I’m glad Gnome Stew has such insightful posters. I’ll definitely reflect on these when I next set up a game. Like I said before, though, they are great addendums to the premise of ‘how to better run the railroad’.
No matter how clear an agenda is, it can still rob the players of sense that they are a part of the game. They may feel like mere dice rolling spectators then. And it is very true that many gaming groups are so well acquainted with the railroad, that they even patiently wait for the next train.
The alternative given by the respondents is the sandbox game, but it’s a false alternative. When all is said and done a sandbox is simply something which takes a lot of work to fill and all you come away with is sand in your pants; it’s still just a (limiting) box.
While front-loading choices may save prep-time, and appear to involve the players, it’s still railroading (only with the vague effects from player input). The players may still feel that they could’ve just phoned (or emailed) their play in. Even switching to a more looser agreed-upon theme or plot can suffer from being as effective as having a committee make decisions while still feeling the same amount front-loaded. Scenario-based gaming is only slightly different, using a black hole approach rather than a railroad, but neither allows ‘escape’.
Going all-out and preparing an in-depth campaign requires a huge investment of time, primarily on material that won’t ever be used (any other way means you’ve simply built many more railroads to reach every station on the line not to be wasted). Even getting together with the players ahead of time does not really co-create the game; see ‘clear agenda’ problem above. Not to mention that I’ve always felt that, if the unit of player input is the character sheet, the gamemaster unit of input isn’t the prepped scenario, it’s the campaign’s ‘elevator pitch’ (or a milieu sheet like I’ve been trying to streamline for my game, Scattershot).
I’m most impressed by the idea of following the ‘introduction – body – conclusion’ script; this is great, exactly what I’d suggest. However, choosing the goals for this ahead of time still makes player-input of marginal impact. I prefer to think in terms of rising tension, climax and resolution. The ‘No Myth’ approach then takes me to adding tension or importance to the result of what the players add and what details they create. (Decry the anti-climax!)
Everything in ‘No Myth’ comes from 1) what the players have done before, 2) the expectations of how the game world is (or the genre) and 3) what the players choose. I might create the idea of a hidden fortress, but I have no problem turning into a floating castle if 1) that doesn’t conflict with anything the players already know and 2) it moves the ‘bad guy’ to a place where the players are going already.
In the end, I’m left wondering, was the hyperbole really necessary to start off this article? The tips given aren’t so much a solution as better ways to maintain the status quo. I hope this writer keeps going at it! For a beginner, he’s got a great style, very communicative, and I want to see more articles from him.
Fang Langford
p.s. And for people who have problems with improvisation (or are afraid to ask), try my article on setting up scenes.
These comments are all great, and I think that they are more beneficial to the Gnome Stew community than the actual post is. Thank you all for expanding upon the material and for sharing your views and approaches. That is one reason why I enjoy writing for Gnome Stew, because the knowledge that the readers share with me helps me to improve my own GMing skills with.
Great comments, folks. Just like to add listen to the players. If they grow attached to minor NPC, then make that NPC more important. If they get focused on chasing down the a minor villain then make him a Big Bad. When faced with decisions like Ocs to the North vs Port to the South, then I ask myself, which will make a better story that the players will enjoy and go from there.
In developing my current campaign world I’ve drawn upon some of the “whole” aspects that I liked about Forgotten Realms when I was running it under 2nd ed. – there’s a lot going on in the setting both openly and behind the scenes which can be utilized in a wide variety of ways. Players can look further into whatever they wish, but realize that what they may find could be very different from what they initially expect – or could be exactly what they expect. This influenced how I used to run Shadowrun which in turn influenced my approach to the background dynamics and information of setting. Players know up front that they can certainly influence events, sometimes radically (and at times unwittingly), but can’t influence everything – some news they will make, the rest they will hear. (There is no guarantee that they will always be able to accurately tell which is which tho’.) e.g. – there’s a war brewing in the region where the campaign begins. Starting characters have no way to prevent this – too much is already too well in motion outside of their sphere of awareness. Certain pieces of information soon come into their possession tho’ and what they do with it/who they tell can potentially change the course/timing of events significantly. In the absence of the junior Intel Officer I had in mind when I was first prepping some of this, they will probably have no clue as to the significance of what they stumbled upon. That turned a more carefully constructed initial story arc into a complete crap shoot hinging upon the big mouth of an over-trusting and incredibly oblivious bard – less story depth for the PCs but possibly much more random fun as the backdrop shifted accordingly. The group’s later decision to travel elsewhere (most coincidentally were followers of the god of wanderers,) meant leaving chaos in their wake – they would only find out about the war later, after it began, making their choice in hindsight seem very fortuitous in both act and timing. . .
Over time “railroading” becomes pointless anyway, as Fang points out, the effects and consequences of their actions can ultimately result in characters essentially railroading themselves – characters who choose to swing first and leave negotiation to survivors can hardly say they were railroaded into combat. Likewise when the legitimate owner of (fill in the blank) comes looking for it, the rest of the Hatfields want to have a few pointed words with the sumbitch as knocked up their sister, or the law tries to execute a warrant charging the bard and “known associates” with “incitement to riot resulting in the destruction of numerous valuable properties both public and private. . .”
There’s rarely any real need to scrap a carefully constructed scenario, again as Fang pointed out, the adventure itself can be set pretty much anywhere. If players are tired of fighting orcs in their underground lairs, the material you spent so much time and effort on can, with only a little work to adjust it, become a carefully constructed running battle with the thieves guild through the sewers of the city located in the opposite direction from the orc stronghold that the players *want* to embroil themselves in.
I like to dig through old, used adventures (preferably out of another system to prevent possible recognition) in game store bins, as well as old map books (60 keyless generic town maps, buildings, etc.) which can usually be had for a buck or two. Customizing is quick and I then have a library of potential adventures to throw out on the fly. I’ve found a wealth of material in the past which I could tie into the setting in great detail, easily fleshing out entire campaign story arcs from a bunch of completely unrelated sources. I generally make some preliminary “assignments” of material to various plot hooks and their developments, setting up the groundwork of a ready story path to follow for any choice the players make. (Except of course for the one they will *actually choose,* cuz St. Murphy the Optimist loves DMs and wants to be sure we know that he still cares no matter how many blessing he bestows upon computer programmers and politicians.) Once it’s certain that a particular story path will not be needed I can simply summarize the resulting news as the campaign develops(having already determined what evens will be occurring in the world where no variables are introduced as their patterns unfold,) and move the reserved material back into the pool, so that orc lairs can reincarnate as thief controlled sewers.
Ultimately, the campaign begins with players being certain that they *will* have significant effects upon the game world. From this follows both the knowledge that I will “let them play their characters” and the clear responsibility to consider what their characters will do as carefully as their characters would “in reality.” (A lot of my comments in the RolePlaying Intensive thread have direct bearing on all of this.) Players then are participating in the creation of the setting and the bigger story while happily expecting both to surprise them when they see the results of the collaboration.
In accordance with Prophecy!
Have Fun, Play Well,
Amergin O’Kai
One thing to bear in mind is the railroad scenario is in some ways necessary for some adventures, primarily one-shots. Also certain groups of players who are, shall we say, uninspired to be proactive in the campaign may need that kind of special prodding. In those cases, the multi-lane approach is superior to a direct-line adventure. I agree most whole-heartedly.
Outside of one-shots and assuming a group of motivated and creative players, however, I prefer the stick-and-carrot method of party management. The carrot is the approach of giving both the players and their characters some buy-in to the game and a reward for certain actions. For instance, in my current Viking campaign, I laid out several adventure scenarios in the order I wanted them done in. With the understanding that my players enjoy turning my game on its ear, I looked at each player’s personal interests and the motivations they gave to their character and wove them into each scenario in ascending order. For instance, several of the players wanted their characters married, the unofficial leader in the group wanted to work on his crafting skill and learn new skills, and the titular leader wanted to reinforce the town they came from. I wanted them to raid the southern lands as youths, deal with the counter-raid by the Romans and then meet the Dark Dwarfs of the east. Therefore, the players learned that the best way to win riches to build up the town and attract good mates was for them to become successful raiders. The players threw their characters to the task willfully and with great enthusiasm. After the raid was ended for the most part successfully (their fellow raiders in their longship pretty much died, while they escaped with much loot), they then returned to begin preparation for the town’s defense, several PCs met good women (daughters and kin of fellow raiders, mainly) and got married, but the one character still wanted to go to the dwarfs to learn how to craft superior armor.
That’s when we come to the stick. For the world to make sense, there would have to be some retribution for the raid they staged, and it would only be fair after they had caught the Romans with their pants down several time in the raid! The recalcitrant player was leveraging the other players to get their characters to leave to find the dwarfs who could teach him how to craft plated armor. When the group learned, however, that their town was going to be assaulted by a vengeful legion of Imperials, their minds were suddenly brought to bear on the consequences of their leaving. Each character had family in that starting town, their prestige was at stake, and now several had wives or investments to guard! They then threw themselves into the business at hand without me having to beat on anyone with the plot hammer. Even the one with his own agenda put it aside for a moment to concentrate on repelling the Imperials at his doorstep. Once they had turned them away, his appetite had been whetted successfully, and the next phase of my story falls into place because it’s now what the characters and the players want as well.
Other good carrots to have on hand are: the search for training beyond one’s normal advancement (3rd Edition had those wonderful prestige classes for that, and anyone who just handed a PC a prestige class missed out on some opportunities to have fun with the characters ;P), political sway, prestige and glory (sometimes this is more effective than dangling a new magical weapon or other treasure before the party), the favor of a god/king/nation, revenge (and this one is why it is soooo essential to have players in a long-standing campaign make a back-story for their characters), or perhaps the opportunity to interact with a famous NPC (like getting to meet with a member of the Circle of Eight in Greyhawk as a treat). Sticks that I have found effective are: protecting one’s past, be it family, holdings, or even sometimes something as intangible as the sanctity of a place of reverence; personal honor; hoarded wealth and its protection; the displeasure of one’s superiors in faith or title (though this sounds heavy-handed, it’s certainly no worse than holding a paladin to his alignment, and it makes sense in a real world); creating opportunities to play to a party or character’s weaknesses (like an orphanage in trouble for a soft-hearted party or character); the preservation of the status quo (this obviously only works on some characters or players). Note that these “sticks”, if overused, become the plot hammer that a railroading DM might use, but if the players even know of their existence, you can steer things subtly enough that they still feel as though they have a choice freely. Rather than saying “No”, offering risks and rewards can make certain decisions easier, and more in tune to the flow of the game you have designed. If done well, the players accept such things as the way the world works. But be prepared – many players may still flout these decisions and take the road less traveled. This is fine, as you are not, in fact, clamping down utterly on their characters’ lives, but they must be willing to accept the consequences.
I agree with Fang in that this is just a railroad with few options; the fundamental problems (and strengths) remain the same.
My personal method is not exactly a sandbox, but neither is there much preparation of any kind. The first step is to construct a starting situation: Say, a village threatened by a powerful giant. The second step is to see the characters; create them for this situation, generate randomly and tweak, or use existing characters. The third step is to change the emphasis of the situation to be on the characters. The fourth step is to simply play; with well-built situation and characters the play produces more content by itself and there is no need to prepare an adventure: It will manifest itself. When it does not, the situation has been adequately played through.
Works just fine with one-shots, which need no railroads for this very reason. They are probaly the easiest to run without any planning.