D&D General The Great Railroad Thread

[Gimby]

Read the best of the Old School modules. The best of them do not even attempt subtle manipulations; they come up with a solid premise with logical hurdles and problems. Those are my preference. Let the players who spent resources shine. Let the final scene play out organically. If the villain has been given a well-developed personality and appropriate abilities it will probably be a dramatic ending no matter how it plays out. But deciding on a conclusion first and then trying to figure out how to get to that conclusion is not how I run games.

Could you give an example of one of these? I'd be very interested in a worked example of the kind of thing that handles this well

 

[Thomas Shey]

Well, I just don't think it is that clear cut. What is "willingly"? How the GM frames things has a huge impact on what actions can reasonably even be attempted. If the GM constantly just frames situations where there is only one direction they can plausibly go (assuming the players don't try to do "I attack the king" type crazy stuff, which they usually don't) then the GM doesn't ever need to "block" actions that would take the players off the tracks, as none can be plausibly be made in the first place. The players might indeed have vague feeling that their actions do not matter much, but there are no specific instances of GM blocking happening that they can point out.

Usually it because the players understand how things are going and don't need to be forced at all once they've signed on.

As I noted before when this came up, when I ran Scion 1e, in practice the PCs needed to stay in their lanes and the players were well aware up-front, because in their areas of specialty there was such a vast gap that if things thrown in the direction of one of them were attempted by another it'd either be impossible or a given that they'd succeed (usually the former). This was particularly pronounced in combat where opponents intended for the heaviest combatants would be overwhelming to the lesser ones and vice versa.

There was no need to apply any force because everyone understood the reality of the situation and was sensitive to not trying to bite off more than they could chew (or worse, direct things at other PCs that would exceed their capabilities).

 

[EzekielRaiden]

The existence of exceptions, outliers and many data points that are all over the place does not invalidate the existence of patterns in the data. Not to speak for DEFCON 1, but he doesn't have to be right all of the time about this to still identify it as a useful pattern to talk about.

Let me rephrase.

Do you really think that most situations in real life have a singular, obvious, correct solution?

Do you really think it is good storytelling to have characters that consistently see the singular, obvious, correct choice?

Because that excludes the possibility of temptation, for example. No character can ever be tempted, because the correct choice will always be obvious. Indeed, characters never been really need to think or plan or the like. There will always be one and only one correct thing to do. And that one and only one thing will be trivial to identify.

That sounds like an eminently boring story, where the only things that happen can be foreseen instantly as soon as you have a few sentences of exposition.

 

[Thomas Shey]

I don’t know if I fully agree but I like this perspective. There’s no railroading test unless the players do something not in the predicted campaign. Then it depends on the deemed unfairness of the dm handling IMO.

This ignores the case in a sandbox campaign where the players simply decide they're not going to engage with the problem at hand any further no matter what would do that and bail. In a linear adventure its sort of assumed they won't do that, that the players will attach motivations to their characters to stay within the adventure.

 

[EzekielRaiden]

I'm happy to adapt my statement away from "most obvious" answer to "most likely" answer. Because that has nothing to do with obviousness or what is most "correct". Rather... every situation is going to have one choice that gets selected more often than any other. The scenario gets played 100 times... one decision that any of the various tables can make will almost always win out and be chosen most often. Because again... players like playing smart.

I agree that they like playing smart. I disagree that this then means there will be one and only one dramatically more common response to any given situation. Some players are severely risk-averse (like my current ones) and will accept an inferior result if it protects them from danger. Others are risk-tolerant and will take whichever seems more likely to pay off. And some will be risk-seeking, eager to take risk even if it is unlikely to pay off just to see what happens.

We can play this out over a dozen metrics. Good vs Neutral vs Evil. Pragmatic vs idealistic. Practical vs theoretical. Intrigue vs violence. Diplomacy vs brute force. Etc., etc., etc. It is in these variations that we get the enormous variety of responses from one group to the next.

And thus even more open-ended scenarios with many different routes still have smarter and more logical decisions than others, and more often than not that one or two most logical choices will win out more often than not. And I maintain that most authors and DMs can still predict that fairly well and thus can plan for those choices beforehand. We know what the most logical or most likely choices would be, even without them being "obvious". So having those more likely choices written or prepped for beforehand-- a more "linear" path-- is not a waste of time. More often than not, that linear path is going to find its use.

I think this significantly overestimates the degree of singularness involved here.

 

[Thomas Shey]

I know, and I don't at all disagree. I'm just not sure how else you write that kind of thing to begin with.

You could set it up so, as the saying goes, "all roads lead to Rome" not in the sense where everything the PCs do will continue the plot as intended, but so that if they miss or destroy one node, there's other routes allowing them to reconnect.

Its just that its both a pain in the behind to constantly do, wastes a fair bit of effort potentially (as you're providing encounters and other material that no one group will ever use all of), and in some cases narrows the kind of adventures you can do more than you may want to.

I'm not really a fan of running adventure paths for exactly the reason that you note. But I'm also not clear on any other way to write them. My own preparation to run isn't something that I could clean up and make presentable as a salable product.

Well, I'm with Frogreaver; as long as I get into an AP knowing that's what it is, I don't mind following all the breadcrumbs. But I do think there's alternate methods, they're just rarely done for the reasons I mention above.

You could always do site-based adventuring, I suppose, like in the old days, but that market is well served by the modern OSR scene, and the reason that the mainstream trad scene isn't into that is because they were never really satisfied with that type of adventure in the first place. They want the game to feel more like a fantasy novel or movie and less like a fantasy-themed exploratory board game or wargame or whatever... but honestly, the trad playstyle has never cracked the code in terms of how you write adventures in a way that doesn't come across as a railroad. So GMs who are trad style but you don't like railroads have to pretty radically alter how they actually run adventures quite often, or at least be willing to in the event that, as you say, the PCs refuse to do what they're assumed to do for whatever reason.

Well, to some extent you can get around it with a low-prep approach if you assume the PCs are at least going to engage with the adventure in a consistent way and not just walk away from it the moment they don't like the cut of its jib. But not everyone is good at that approach, and not all of the problems I address earlier go away just because you do.

It's just the risk you always take, I think. Although in my experience, it's not really a very high risk. Most players intuitively understand the social contract of "this is the game I've prepared" so they tend to engage with it.

Yeah. The only exceptions you're going to get are people who really are expecting everything to be a freeform sandbox, or have a very self-centered view of play anyway.

 

[SkidAce]

This ignores the case in a sandbox campaign where the players simply decide they're not going to engage with the problem at hand any further no matter what would do that and bail. In a linear adventure its sort of assumed they won't do that, that the players will attach motivations to their characters to stay within the adventure.

I feel that if the "players will attach motivations" then we still have linearish adventure, but if the DM forces motivation or specific encounters/actions, then its a railroad.

 

[Crimson Longinus]

So you want your players to not play smart? To make bad decisions based on the information they have in hand? Or is it you just don't want there to be a throughline to your adventures and you want your players to just move randomly around (because to do otherwise is 'railroady')?

If that's how you prefer to play and your players prefer to play, that's cool. But hopefully that's just because you like that style and not because you have a need to 'not be railroady' first and foremost.

No, I want there to be complicated and nuanced situations that do not have one obvious correct solution.

 

[Thomas Shey]

I feel that if the "players will attach motivations" then we still have linearish adventure, but if the DM forces motivation or specific encounters/actions, then its a railroad.

I think it matters when you do that. Its not a railroad if, at the start of the campaign the GM says "You need to all design characters who want to do X when it comes up, because that's what the campaign is about." That's constraining the initial character-set, but isn't forcing the point in play.

(Of course you can have problems where the players' understanding of what that means and the GM's are different, but, "communication is hard, yo dog".)

 

[RealAlHazred]

Could you give an example of one of these? I'd be very interested in a worked example of the kind of thing that handles this well

Sure.

Let's check out Holy Mountain Shaker by Luka Rejec. It's statted for OSE, but is easily portable to just about any OSR game. Spoilers for that adventure.

Spoiler: Major spoilers ahead.

Some say there's a powerful creature, a God Fish, stirring in a dungeon under a nearby mountain. Others don't think it exists, and that the earth tremors are caused by something else -- a wizard did it, or those pesky whatevers. The townsfolk feel the earthquakes and many are concerned about the town being destroyed.

In the classic module style, they might have written that there are a few NPCs specified in the scenario, in various parts of the dungeon. You'd need to go to one, then the other, to build up a group of allies, and then maybe you'd successfully defeat God Fish in a set-piece battle with whatever reinforcements you were able to get.

But the scenario isn't written that way. All of the NPCs have their motivations, but they're characters, not cogs in an adventure-machine. There are factions in the town who want adventurers to go in, but for different reasons, at different times, for different amounts of compensation. There's a time limit on the adventure -- if the party doesn't resolve the problem by the time the countdown ends, the load-bearing God Fish departs and the dungeon collapses, causing a massive earthquake that will devastate the town.

There are good and bad things you can find all over the dungeon, but they aren't placed there to support a specific end-goal; instead they get there organically, as a result of the other creatures in the complex.