D&D General Why defend railroading?

Been out of the thread for a bit, and you weren't talking to me. But my answer would be, "A haunted house is a location, and locations don't move, while creatures generally do." Now, I might invent a haunted house and place it in an area I expect the players to visit, but I see that as very different--that's populating a location before the players arrive. Yes, populating a location means me as DM inventing fiction to fill what was an empty unknown. I'm just dramatically less comfortable with "this location simply appears wherever the players decided to go" than I am with "once the players have decided to go somewhere, it needs to be a Something and not a Nothing." (Reading your most recent post, I'm deeply uncomfortable with palette shifting, but not with painting the background as I need it, so to speak.)
There is perhaps a grey area here. If the players expect that everything has already been mapped out by the GM, and that they are exploring a fully mapped out area, then the GM should not be moving things around.

But that's not always the case. In a procedurally generated hex crawl there may stocking of hexes on the fly. And in this case, I don't think random is fundamentally different to GM choice in terms of railroading (in other ways yes, but that's a different discussion). Now not every location will fit in every direction, but as you say that's the same for monsters, so I don't see a fundamental distinction here.

If the space in the game is more abstractly defined, then again, it may not be an issue. There may not even be a map to explore, or it may be so zoomed out as to leave the particulars of the local region undefined.
 

log in or register to remove this ad


Oh, and on the topic of not having locations be on random encounter tables or be otherwise mobile -- you're absolutely missing out on a great encounter tool. "Monster lair" is a great encounter idea. As is "fairy grove" or "thin place between planes." These don't need to be in fixed places, planned beforehand. They can be on encounter tables, or encounter lists, or just dropped in when it feels cool.
 

I'm not sure I follow this. Let's say we have path A and path B. The players learn that there is an encounter of a specific type on path A -- an ogre. They choose to go down path B explicitly to avoid this encounter. The GM determines (however) that they still have some encounter along path B. Okay, we have the setup. I see a few ways this can play out, and I'm not sure it's at all visible to the players:

1) The GM moves the ogre encounter, but reskins it as something non-ogre (I guess there's a useful set of stats)? The players do not encounter an ogre, but instead a stat-block like an ogre but with a few flavor changes and perhaps minor mechanical ones?

2) The GM rolls on their random encounter table, and gets an ogre. I'm not sure where this leaves us.

3) The GM has an entirely different encounter than an ogre or re-skinned ogre.

So, across 1, 2, and 3, the players still have an encounter no matter what. In 1) the GM "palette shifts" but the players may or may not be able to discern this, but they don't encounter an ogre, just a ogre-y stat block. In 2) there's nothing at all railroad-y or palette shifting or quantum ogre-y, but the players still encounter an ogre. In 3, the players don't encounter an ogre (yay!) but still have an encounter. This might looks the same as a successful palette shift to the players.

This appears that "palette shifting (1)" is somewhat overblown, in that you can have the same appearance without the attempt (3), and an even worse appearance without the attempt (2). I'm not sure this concept does a lot of work.
I think you're overthinking the trees and missing the forest here.

It's always a mistake to focus on the ogre rather then what it's supposed to represent.

Edit:
1) How much you can get away with reskinning depends on how much you reskin and what exactly it was the players wished to avoid.
2) If you roll a random encounter for an ogre, it should probably be at least connected to the previous ogre. Maybe it's the ogre's partner or another ogre that hates that ogre, or something. In any case the random encounter is within the agreed upon parameters of the game. It is not GM fiat.
3) If the players were specifically avoiding the ogre, then they can probably avoid the next encounter too. Then they can make a choice between the two or take a different direction entirely, or backtrack to where they started.
 
Last edited:

then my choice mattered because the direction I picked is the direction it was in. It’s the GM shifting things do it comes up no matter which direction I choose that it’s a problem

Even if that thing you decided about is two hexes out, and the haunted house is put in the intervening hex of whichever direction you go? Is it ok if it's avoidable once you find out it's there (E.g. it doesn't fill the hex).and becomes a permanent part of the map.
 

There is perhaps a grey area here. If the players expect that everything has already been mapped out by the GM, and that they are exploring a fully mapped out area, then the GM should not be moving things around.

But that's not always the case. In a procedurally generated hex crawl there may stocking of hexes on the fly. And in this case, I don't think random is fundamentally different to GM choice in terms of railroading (in other ways yes, but that's a different discussion). Now not every location will fit in every direction, but as you say that's the same for monsters, so I don't see a fundamental distinction here.

If the space in the game is more abstractly defined, then again, it may not be an issue. There may not even be a map to explore, or it may be so zoomed out as to leave the particulars of the local region undefined.
As I mentioned far upthread, some of this may lead to problems though, if you really are (in some sense) "making the world exist" only in the moment it's called for. A player felt there was no tension, no weight to choosing, if they were choosing Currently Faceless Void #1 vs Currently Faceless Void #2, when I would roll and whatever I would roll would end up being in whichever thing they chose.

More or less, he wanted to see a world that was randomly-generated, but randomly-pre-generated. One where choosing to go left didn't (OOC) manifest the Glassworks to the left, but rather, where the Glassworks already was to the left. Where choosing to go right might mean missing the Glassworks entirely, should the players fail to return to check the left direction too. I haven't done any "roll to explore the location" content since then, but if I do, I'll be factoring in this feedback. (Likely, I'd pre-generate as many rooms as I thought could be reasonably hit within one session, and only generate new rooms if required--a balance between addressing that perceived absence of tension and not mapping out a huge space that might go unobserved.)
 

Even if that thing you decided about is two hexes out, and the haunted house is put in the intervening hex of whichever direction you go? Is it ok if it's avoidable once you find out it's there (E.g. it doesn't fill the hex).and becomes a permanent part of the map.
I think one non railroad moment doesn’t cancel out the railroad moment. It would maybe make it less of a railroad that I can at least try to avoid it. But having it be me my path no matter what seems to me the text book definition if railroad
 

As I mentioned far upthread, some of this may lead to problems though, if you really are (in some sense) "making the world exist" only in the moment it's called for. A player felt there was no tension, no weight to choosing, if they were choosing Currently Faceless Void #1 vs Currently Faceless Void #2, when I would roll and whatever I would roll would end up being in whichever thing they chose.

More or less, he wanted to see a world that was randomly-generated, but randomly-pre-generated. One where choosing to go left didn't (OOC) manifest the Glassworks to the left, but rather, where the Glassworks already was to the left. Where choosing to go right might mean missing the Glassworks entirely, should the players fail to return to check the left direction too. I haven't done any "roll to explore the location" content since then, but if I do, I'll be factoring in this feedback. (Likely, I'd pre-generate as many rooms as I thought could be reasonably hit within one session, and only generate new rooms if required--a balance between addressing that perceived absence of tension and not mapping out a huge space that might go unobserved.)
Yeah some players are never going to enjoy exploration under these circumstances.

Although part of the art in doing this is working out as you go how it all interacts with every else that has already been established.
 

How do you know what the DM wants to happen? You mean, things like have interesting encounters? And you don't have anything marked on that hex - do you suspect DM shenanigans whenever you have a non-trivial encounter that wasn't foreshadowed before you chose a direction? (Volo travels guide subscription?)
because in this case the GM in question is telling us so. But like I said you don’t notice it in one instance. Over time though I find it often becomes clear what is going on
 

Oh, and on the topic of not having locations be on random encounter tables or be otherwise mobile -- you're absolutely missing out on a great encounter tool. "Monster lair" is a great encounter idea. As is "fairy grove" or "thin place between planes." These don't need to be in fixed places, planned beforehand. They can be on encounter tables, or encounter lists, or just dropped in when it feels cool.
I admit, this is diving right into the grey area, as Don Durito said, but I still see some major differences here. A fairy grove makes sense as a thing you could just sort of...randomly run into wherever. It fits as a random thing, in part because it's a very simple, straightforward location: a grove, where fairies live. Likewise for "monster lair" (very simply, "a monster sleeps here") or "planar thin place" (which could be almost anything).

A haunted house, by comparison, is a much more involved thing. I mean, purely from the brute facts, it had to be built by someone, or in some other way created (magic, illusion, whatever). And it's almost certainly going to be filled with a lot more stuff than "a fairy grove" or "a monster den" is going to be. The whole point of this haunted house example is that it was, in essence, a mini-adventure being forced into the players' path, no matter what they choose. Having a fairy grove, monster den, or other small surprise crop up along the journey...that's pretty well expected, if you're travelling through Dangerous Wilderness.

Again, there is no proper formula for avoiding railroading. But it seems pretty clear to me that a "haunted house" is in a different category from these small, simple, natural phenomena. It's much more complicated and implies far more stuff about the world around it than the fairy grove or the monster den does.

Even if that thing you decided about is two hexes out, and the haunted house is put in the intervening hex of whichever direction you go? Is it ok if it's avoidable once you find out it's there (E.g. it doesn't fill the hex).and becomes a permanent part of the map.
I don't run a hex crawl, so this is hard for me to comment on. I imagine it might depend in part on the group.

Perhaps I should ask a question: Is the house being put there purely IN ORDER to be in the party's way? Or is it being put there, alongside many other things, so that there is a "there" to go to? Because I'd definitely argue that intent and purpose matter a LOT for this situation. If the intent is "I'm trying to passive-aggressively make the party do this thing I made," then...it's soft railroading where the DM permits an out but is really really trying to make it happen. If the intent is, "Ooh, this could be fun, let's plunk it down over here and see what the players do," then it's not railroading, it's just painting the background and giving the players their own paints to add to it.

It's a lot like trying to deal with Mary Sue/Gary Stu as a writing problem. No matter what description you give based on the specific implementation, it is almost always possible for an author to still squeeze out something that gets a false negative. Because the problem isn't "OC inserted into an existing universe," nor "character with tons of special abilities," nor "character everyone likes," it's "character that causes the whole narrative to be about them," and that's an inherently abstract concept. Like the Mary Sue, railroading is a problem with a nigh-infinitude of specific implementations, because the problem doesn't exist at the implementation level.

I see a world of difference between "populating a location so that the party has interesting things to do or see" and "making sure that, whatever the players choose, my DM plan goes off without a hitch." The former is, as I said earlier, invention: you're inventing things to populate the space. The latter is illusionism: making the players believe they're calling shots that they just, flat, aren't. Every DM employs invention regularly. I don't believe any DM needs to employ illusionism--and that the risks and costs of doing so outweigh the benefits.
 

Remove ads

Top