Consequences, Railroading, and Schrodinger's Dungeon

[HidesHisEyes]

All fiction is built on "contrivances" and I would postulate that IRL the people who managed to change the world all had more than their share of "contrivances" occur to help them along.

In a linear game if you told me you wanted to kill the gods and reformat the world I would make sure there is a path to do that. In a sanbox, I shrug, tell you that on the off change the game is still happening in ten years and you haven't already been killed (or had someone else beat you to the punch) then maybe you can do that when you are level 40; but until then it just ain't going to happen without a "contrivance".

I broadly agree.

A piece of fiction in any medium is literally one big contrivance. In an RPG it means you make choices about how to present information, organise your prep, keep things interesting without SEEMING too contrived, and all sorts of other things. And players make choices about their characters and their actions.

Even if it’s a good metaphor for one’s GMing style, no one ACTUALLY creates a fully functional fictional world and watches events in it unfold as if it were real. That would be impossible. We make creative choices all the time, just like in any medium.

One approach to making these choices is just to plan a linear sequence of events and try to make sure they happen. That’s railroading. Having the players just happen to meet the king on the road right when the assassination attempt takes place is a specific choice. It might be too obviously contrived, but it’s not railroading unless you force it to turn out a certain way.

I see a lot of people talking about railroading as if it means “any choice or structuring of game material that isn’t ENTIRELY based on the internal logic of the game world”, or sometimes even “any structure whatsoever”. I think it’s completely off base.

[Vahnavoi]

Fair point, let's use "DnD" then, instead of "RPG". I think my original point stands, that if someone shows up to play a game of DnD (or a game of 99% of systems), they're going to accept when the dice determine the outcome of an event, especially events like random encounters, like the original example.

Oh how I wish that was the case. Next you'll tell le nobody ever ragequits Chess.

Like, consider: entire trends of game design have come and gone based on some player disliking what the dice or their game master decided. For a while in boardgame and wargame circles, the dominant opinion was that any use of dice, at all, is bad, because games should depend on skill, not random chance. For a while in the roleplaying game circles, the dominant opinion was that D&D is a bad game, in fact all games with a game master are bad, because decisions should be made by everyone at the table, all authority is illegitimate, so on and so forth.

Losing gracefully, accepting defeat, accepting some decisions are made by others and not you, these are personality-dependent and to a degree, skills. I'm not kidding here. Games are used as teaching tools in schools to literally teach these things to children. Every once in a while, somebody slips through, fails to learn the lesson or forgets it, just as surely as some people fail to learn or forget general solution to quadratic equations. It's not given a person sitting down to play has accepted or even understood what the game rules are telling them.

Ok, I think we've found the distinction, to a lot of people, the Ogre does exist, in a way that a future choice does not.

Said people are fooling themselves.

I hope we can agree that an action you've already done exists more than one you haven't.

As a matter of fact, we can't. You haven't even told me if you're a presentist or eternalist. Though I'll argue for both cases that there's no difference between an ogre that hasn't been placed yet and a chess move that hasn't been made yet.

Even if you have this strategy in your head, until you actually make the move, you could always change your mind, for whatever reason, and make a different move. The same is not true for past actions. So, from what I can tell, the difference in our stances here (I'm going to put words in your mouth, correct me if this isn't your real stance), is that most of us in this thread seem to think that the DM has done an action during his prep, while you seem to be of the opinion that he has not until it has shown up on screen. My view sees the fictional world as something with continuity beyond what's shown up "on screen" at the table, as opposed to only existing as far as has actually been played out. To me, the Ogre "exists" in a way before actually being described to the players, because the DM has it in his notes (or at least his head) that he exists.

You seem to be forgetting that the exact location of the ogre isn't determined until the moment the game master puts it in the game. The prepwork the GM has done is not the same action as actually placing the ogre. For comparison a Chess player could actually play and write down sequence of multiple Chess games before facing an opponent. Then, when their opponent makes a move they have predicted and put down in their notes, they can just follow the response put down in their notes. As far as rules of Chess are concerned, the move in the pregame is the same as the move in the actual game, but obviously the move in the actual game isn't done until it's done.

If I'm wrong, and your point is that the entire universe is perfectly deterministic, and so future actions are just as concrete as past ones, that's a reasonable-enough metaphysical position, but I don't think it's particularly relevant to a thread about planning out Dungeons and Dragons encounters.

Or, I could argue only the present exists, and past actions are just as non-existent as future actions: they only exist as extrapolations in someone's mind.

But the actual argument I'm making is that the "quantum ogre" is useless fiction, the ogre is never in superposition, and you can replace that fiction with various other fictions. The "travelling ogre" in my last post was one, but you could just as well go with older "The game master is God and everything in the game exists by God's will". That isn't exactly correct either, but it does describe some things of the situation better: the existence of the ogre does depend on the game master exercising their will, that is, positive mental effort to have it appear in the game, and if they don't will it, it doesn't.

[Morgaln]

I think you are assuming a binary and that is the blocker you are facing when wrapping your head around my post.

Let's say there is a long costal road between city A and city B. There is a short well paved and patrolled forest road between city A and city C. The PCs can make an informed meaningful choice about destination, and travel time. Can they make an informed meaningful choice about the dangers they might face?

In campaign X the dangers on road AB are the same as on road AC. There will be 1 ogre (named John Doe) fight between start and finish.

In campaign Y the dangers on road AB are different from on road AC. The costal road is not patrolled since attacks are less frequent. It being a costal road means there is an increased chance of amphibious or flying enemies. The forest road is patrolled since attacks are more frequent. It had a greater chance of wolves or bandits attacking. The PCs know the coastal road will have more time between attacks and the attacks might be more dangerous than the attacks on the forest road. <Split the timeline> In one timeline the PCs go down the forest path and encounter an ogre bandit with a few orc bandits that ambush the party a few times that day. In the other timeline the PCs go down the costal path and encounter a flight of harpies on the first afternoon and a kraken while they camp on the 3rd day.

Campaign Y has more player agency than campaign X and all I did was 1) remove the quantum ogre and 2) allow the PCs to make an informed choice about what replaced the quantum ogre.

The amount of agency is a spectrum. Using a quantum ogre instead of an informed meaningful choice is a way to decrease player agency with respect to the encounters on that trip.

In part due to the amount of player agency is a spectrum, player preference will vary. This is why a quantum ogre is not inherently a bad thing, but it can be if it conflicts with the playgroup preferences. The same is true for the inverse, high player agency is not inherently a bad thing, but it can be if it conflicts with the playgroup preferences.

You are comparing completely different things. One is a short travel with exactly one encounter. In the other one you added additional encounters and events instead of having just one. That is more than just removing the quantum ogre. Please reread what I wrote above. I specially noted that the roads would be different in some way, just that both would have an encounter with an ogre. That's explicitly different from your example x. I can easily take your example y and insert an ogre into each road, either in addition to or replacing the existing encounters. That doesn't take away any player agency.

Obviously example y will provide more player agency. But not because of the existence or absence of an ogre. But because example y contains actual information for the players to base their choice on beyond "will you go left or right?" Meaning the players are informed enough to make an informed choice. Which is actually something you earlier specifically denied is relevant to the question, whereas now you use it as the reason for player agency. So you are contradicting yourself

Here's the quote I'm referring to:

The point of a quantum ogre is "Here is something that the GM prevented PC choice from having impact". It is separate from the question of "Were the PCs informed enough to make an informed choice". So if you have PCs choosing between roads, and you add a guaranteed random encounter that exists but will be placed in front of the PCs regardless of their choice, then the PCs can't have agency over that random encounter and there is no in fiction reason for the lack of impact.

If the GMs notes say "Ogres frequent both roads," then that is a valid piece of world-building and the GM is entirely justified to have the players encounter an ogre regardless of which way they choose. The fact that he'll use the same stats and encounter details either way does not change that. If the players can get that information ("there are ogres on both roads"), it still doesn't remove their agency. It just means that the existence of ogres is not a criterium they can base their choice on.
If you want to drive from your house to the store and there are two roads with the exact same length you could take, that doesn't take your agency in choosing which way to go from you. It just means that "miles driven" is not a useful criterium, but that you should base your choice on different reasons, like the quality of the streets, the expected traffic, the number of traffic lights, anything like that.

[Tanarii]

p
Like, I define "Railroading" as "Denying the PC's any meaningful choices" because that's the definition I find most useful, but it's frustratingly vague and hard to apply.

Thats why I prefer the Alexandrian's "the GM negates a player’s choice in order to enforce a preconceived outcome". Because it requires a player choice first, then a DM action to negate it after, with a purpose in mind. Or to quote from the first paragraph of his article: "The players must try to get off the train and the GM has to lock the doors."

[kyoryu]

Thats why I prefer the Alexandrian's "the GM negates a player’s choice in order to enforce a preconceived outcome". Because it requires a player choice first, then a DM action to negate it after, with a purpose in mind. Or to quote from the first paragraph of his article: "The players must try to get off the train and the GM has to lock the doors."

Agreed.

A completely random dungeon isn't railroading. It may not offer much agency, but it's not railroading.

My totally informal heuristic is "does the GM know what's going to happen?" If so, you're probably railroading. If not, you're probably not. The further away that you know, the more likely it is that it's railroading. (Knowing what NPCs are going to do providing nothing interrupts that isn't really part of this).

I find the arguments about railroading interesting because they quickly get so extreme - like "if the PCs don't have complete knowledge about everything, it's railroading" or "if there's anything that would stop a PC's plan from working, that's railroading". That just doesn't make sense to me. I think they're too micro-focused on the actions, without looking at the context behind them. Sometimes I feel like it is (and I've seen this) a bad faith argument in the "broad->narrow" category - basically, "by this incredibly broad definition, almost every game has railroading, therefore railroading in the more narrow definition is okay".

[FilthyLucre]

This problem is pretty solvable, I feel like, if you design scenarios/situations rather than plot things out. Example:

Last session of my game pirates had kidnapped a merfolk and a clutch of eggs and the PCs decided to rescue them. I designed the small island stronghold that the pirates were stationed at, figured out how many pirates would be lurking about, where the eggs were, and where the merfolk was. I did not plan or even consider how they would actually rescue the merfolk. They ended up giving her a shrinking potion that I'd given them months ago and yeeted her out a window back into the sea.

[Easy e]

First of all, I am 100% sure I am over thinking it! However, it does lead to more interesting discussion of a topic that is usually just "Railroading is bad M'kay!"

The idea that railroading is actively shutting down a player AFTER they made a choice is closer to the mark, but also not entirely accurate as the GM has to shut down choices that do not make sense in the context of the game all the time. If player A said, "I build a tactical nuke, and then use it on the Dragon" the GM should negate that choice. I think we all agree that is not railroading.

If we nudge players along to various encounters. That is not railroading. They are choosing which encounters or locations to move towards.

If we build a dungeon lay-out and the players follow it. That is not railroading. They are choosing to move left or right.

However, if a player says they want to climb the wall of the tower as opposed to the front door; and the DM contrives a reason they can not climb the wall. That is railroading?

I find the arguments about railroading interesting because they quickly get so extreme - like "if the PCs don't have complete knowledge about everything, it's railroading" or "if there's anything that would stop a PC's plan from working, that's railroading". That just doesn't make sense to me. I think they're too micro-focused on the actions, without looking at the context behind them. Sometimes I feel like it is (and I've seen this) a bad faith argument in the "broad->narrow" category - basically, "by this incredibly broad definition, almost every game has railroading, therefore railroading in the more narrow definition is okay".

Yes, this gets to the heart of it. I think we can all agree that some GM agency to guide the session is OK, as long as it does not break the player's sense that they are making decisions and guiding the story themselves? Is that the difference between railroading bad. or "not railroading"?

[Wintermoot]

This problem is pretty solvable, I feel like, if you design scenarios/situations rather than plot things out. Example:

Last session of my game pirates had kidnapped a merfolk and a clutch of eggs and the PCs decided to rescue them. I designed the small island stronghold that the pirates were stationed at, figured out how many pirates would be lurking about, where the eggs were, and where the merfolk was. I did not plan or even consider how they would actually rescue the merfolk. They ended up giving her a shrinking potion that I'd given them months ago and yeeted her out a window back into the sea.

This may work for you, but it doesn't work for every GM. Because it can lead to a very static bland experience depending on how fast the GM thinks on their feet.

What I mean is this. If you haven't put any thought into how the scenario may change or grow depending on how the players respond, then if they come up with a solution that bypasses everything then it turns out to be a very boring game.

So you detailed the merfolk village. You may have put some additional loot in or areas of interest for them to find. You have have inserted a hook for a future session or story. But if they bypass all of that with the shrink potion and through the window, the entire session takes 30 minutes to do and ends up very non-fun for both players and GM.

Granted, this isn't ALWAYS true, but is SOMETIMES true. Some GMS run a better game by being over pre-prepared so they don't have to do so much thinking on their feet. Some Players enjoy a game where it evolves as they go to give a satisfying rich experience.

As always, your mileage may vary.

[Telok]

My totally informal heuristic is "does the GM know what's going to happen?" If so, you're probably railroading. If not, you're probably not. The further away that you know, the more likely it is that it's railroading. (Knowing what NPCs are going to do providing nothing interrupts that isn't really part of this.

When I DM the players are sometimes absolutely predictable, and sometimes insanely unpredictable. Usually due to how they have specific habits and reactions to certain situation or NPC personalities.

Due to the unpredictability and my own preferences I tend towards a bounded sandbox setting/scenario style of game. However, when I can predict some/most of what the players will do, I try to manipulate them into situations where they have to choose what & how much the thing they want is going to cost. Generally only statting things out about a session and a half ahead, although I often have semi-detailed setting outlines and a stock NPC backup cast.

I have been accused of railroading when I successfully predict their actions and they get faced with choices or consequences. Although I think its interesting that I get more accusations of railroading when they get consequences of their actions that arise from the insane random crap they sometimes pull. Maybe I make a weird mirror reversal of railroading somehow. They seem to feel that physics/NPCs reactions to their lol-random detours are railroading, while if I plot 2 paths with 4 encounters each that lead to the same end scene it isn't a railroad.

[kyoryu]

Yes, this gets to the heart of it. I think we can all agree that some GM agency to guide the session is OK, as long as it does not break the player's sense that they are making decisions and guiding the story themselves? Is that the difference between railroading bad. or "not railroading"?

Ultimately, I look at a (I consider it the primary and the first) fundamental interaction of RPGs to be:

GM: "This is the situation." (A)
Player: "This is what I do."
GM: "This is the new situation."

It's a useful tool, and contains more information than it looks like at first.

The GM has the absolute right to define the situation, within limits of the fictional world (and since the GM defines those, that's more about consistency than anything). This is their input into the situation. (A)

The player has the right to respond within the limits of the fictional world and what their character can do (B).

The GM has the obligation to respond to that (C). The resulting situation should be a combination of the initial situation, the player action, and any other new information the GM wants to introduce.

Where you get into railroading is where C is not dependent on the player action. That's the quantum ogre, right?

GM: "You're at a fork in the road, left or right?" (presumably more info is given, left brief for example purposes)
Player: "I go left"/"I go right"
GM: "You fight an ogre."

Note that it's possible for the result to be the same for multiple actions and it not be railroading. However, for that to be true, at least part of the result needs to be due to randomization (I attack with my dagger/I attack with my sword can both result in 1 damage) -or- there needs to be other potential player actions that would result in a different result ("I head back to town" in the case of the QO).

It's when the GM already knows what step C is, and will enforce it that you're really railroading.

I ran into a not-as-subtle-as-he-thought version of this with a friend. There was a local "primitive" type person guarding a tomb thing. We had teh option of attacking or not. This person was willing to let us go even though we had killed half her family so long as we agreed to not open the tomb thing.

Every question that we asked where the answer could have led to a reasonable answer of "don't open it" was met with "I don't understand you." Every question that was asked that could have led to "yes, you should totally open it" was understood. Even though it was clearly a Sealed Evil In A Can scenario, we were prevented from getting info that would have led to it.

We ended up opening it and releasing the Sealed Evil, of coruse. And the GM was all "you could have made any choice you wanted!" Maybe..... and maybe he believed he wasn't railroading, but the information was clearly controlled to get us to make the decision he wanted. It's still railroading, but just more subtle. That doesn't make it better.

[OldTrees1]

You are comparing completely different things. One is a short travel with exactly one encounter. In the other one you added additional encounters and events instead of having just one. That is more than just removing the quantum ogre. Please reread what I wrote above. I specially noted that the roads would be different in some way, just that both would have an encounter with an ogre. That's explicitly different from your example x. I can easily take your example y and insert an ogre into each road, either in addition to or replacing the existing encounters. That doesn't take away any player agency.

Campaign X has two roads that are different in some way, just that both would have an encounter with an ogre, and the ogre was the same quantum ogre named John Doe. This addresses the quantum ogre case but not the "copy paste there are now 2 ogres" case. Those cases are different with respect to verisimilitude, I don't think they are difference with respect to the potential to have agency over the encounter.

Double checking: When you say the roads are different in some way, I assumed they are different in some other meaningful informed impactful choice way. That is why I set up road AB and AC to be a meaningful informed impactful choice. Campaign X and campaign Y differ about whether the encounter (or encounters in campaign Y's case) are part of the meaningful informed impactful choice OR if there is a quantum ogre.

Obviously example y will provide more player agency. But not because of the existence or absence of an ogre. But because example y contains actual information for the players to base their choice on beyond "will you go left or right?" Meaning the players are informed enough to make an informed choice. Which is actually something you earlier specifically denied is relevant to the question, whereas now you use it as the reason for player agency. So you are contradicting yourself

Campaign Y provides more player agency because I removed the quantum ogre (counter action against a choice) and made its replacement a possible informed choice (the players have the potential to be informed enough to make an informed choice). Yes, as part of this replacement, I did have the new encounter(s) tied to information the PCs had about road length length and in fiction frequency of attacks.

counteraction negates agency
lacking information negates agency
A quantum ogre technically only affects the former but frequently the latter is assumed by the example. Campaign Y replaced the counter action choice (which was also a blind choice in campaign X) with an informed impactful choice.

I never denied information was a prerequisite for a meaningful informed impactful choice. I affirmed information was a requirement but counter action also negates agency and the quantum ogre mechanic specifically deals with the counter action.

Here's the quote I'm referring to:

If the GMs notes say "Ogres frequent both roads," then that is a valid piece of world-building and the GM is entirely justified to have the players encounter an ogre regardless of which way they choose. The fact that he'll use the same stats and encounter details either way does not change that. If the players can get that information ("there are ogres on both roads"), it still doesn't remove their agency. It just means that the existence of ogres is not a criterium they can base their choice on.
If you want to drive from your house to the store and there are two roads with the exact same length you could take, that doesn't take your agency in choosing which way to go from you. It just means that "miles driven" is not a useful criterium, but that you should base your choice on different reasons, like the quality of the streets, the expected traffic, the number of traffic lights, anything like that.

Yes, if the GM notes "Ogres frequent both roads" that would be a valid piece of world-building.

Yes, the GM is entirely justified (subject to information about playgroup's preferences) to have the platers encounter an ogre regardless of which way they choose. That is true regardless of the GM's notes. More/less agency is not inherently bad, check the playgroup's preferences.

Correct, having less agency does not mean you have no agency.

However having less agency (agency over destination & travel time vs agency over destination, travel time, & expected dangers) is having less agency. If the roads from house to store are the same with respect to a variable, then I do not have agency over that variable. I still have agency over the other variables that do differ between the roads provided I am aware of the difference. Agency is not a binary. Agency is a continuum.


Summary:
1) An informed meaningful impactful choice requires the choice be informed (lack of information, aka blind choices, prevents agency over that aspect) meaningful (to the players care about it) and impactful (does the GM negate the choice or leave it stand?).
2) Choice that includes more informed meaningful impactful choices has more agency than one that has fewer.
3) The right amount of agency is decided by the playgroup preferences. Something having more/less agency does not make it inherently better. Insert boilerplate about playgroup preferences.

Unless you think agency is a binary, I don't see where we disagree.

[BRC]

Thats why I prefer the Alexandrian's "the GM negates a player’s choice in order to enforce a preconceived outcome". Because it requires a player choice first, then a DM action to negate it after, with a purpose in mind. Or to quote from the first paragraph of his article: "The players must try to get off the train and the GM has to lock the doors."

The reason I dislike that is because you can get the same effect without intervening beyond establishing the scenario.

You can drop your players into a linear dungeon full of enemies who will attack on sight, with magically indestructible doors that won't open until you've defeated every enemy in the room, locking your players into a series of pre-established encounters, and never needing to lift a finger to keep them "On-Rails". In my mind, denying the players any chance to make decisions is on-par with negating them, even if it's a bit easier to hide in the background noise of scenario building.

From a perspective of "what advice to give to a GM", I try to stay away from anything that could be interpreted as "If you do enough prep work and cover all your bases, you won't need to Railroad and everything will be fine", or "If the story you're trying to tell is good enough, your players won't try to get off the rails", because I feel like those are solving the wrong part of the problem.

Every Railroader thinks they've done enough prep work/ thinks that their story is so good the players won't try to go off the rails. The Alexandrian's definition is what happens when they're Wrong.

[kyoryu]

You can drop your players into a linear dungeon full of enemies who will attack on sight, with magically indestructible doors that won't open until you've defeated every enemy in the room, locking your players into a series of pre-established encounters, and never needing to lift a finger to keep them "On-Rails". In my mind, denying the players any chance to make decisions is on-par with negating them, even if it's a bit easier to hide in the background noise of scenario building.

I agree with this entirely.

From a perspective of "what advice to give to a GM", I try to stay away from anything that could be interpreted as "If you do enough prep work and cover all your bases, you won't need to Railroad and everything will be fine", or "If the story you're trying to tell is good enough, your players won't try to get off the rails", because I feel like those are solving the wrong part of the problem.

Sure. I think the viable options are "don't make a game that's a linear path" and "make a game that's a linear path, and get everyone's buy-in".

Anything else is sketchy.

[HidesHisEyes]

Agreed.

A completely random dungeon isn't railroading. It may not offer much agency, but it's not railroading.

My totally informal heuristic is "does the GM know what's going to happen?" If so, you're probably railroading. If not, you're probably not. The further away that you know, the more likely it is that it's railroading. (Knowing what NPCs are going to do providing nothing interrupts that isn't really part of this).

I find the arguments about railroading interesting because they quickly get so extreme - like "if the PCs don't have complete knowledge about everything, it's railroading" or "if there's anything that would stop a PC's plan from working, that's railroading". That just doesn't make sense to me. I think they're too micro-focused on the actions, without looking at the context behind them. Sometimes I feel like it is (and I've seen this) a bad faith argument in the "broad->narrow" category - basically, "by this incredibly broad definition, almost every game has railroading, therefore railroading in the more narrow definition is okay".

Totally agree, except I’m actually comfortable treating your informal heuristic as pretty much a strong definition. I think railroading isn’t just about player agency and meaningful choices, it’s about the narrative sequence of events. If the GM has that sequence planned out and actively resists deviating from it, that’s a railroad.

I suppose with my definition there’s a VERY limited sense in which it’s true that all games have railroads, because the GM merely presenting something - anything - as part of the game world is technically a “pre-planned event”. But to conflate “at some point the PCs meet a sad goblin” with “the PCs meet a sad goblin and agree to help him find his lost necklace, and they find it and it’s a powerful artefact, and the goblin uses it to destroy the city, etc etc” is… well I don’t know if it’s bad faith but it’s obviously wrong.

[Wintermoot]

Ultimately, I look at a (I consider it the primary and the first) fundamental interaction of RPGs to be:

GM: "You're at a fork in the road, left or right?" (presumably more info is given, left brief for example purposes)
Player: "I go left"/"I go right"
GM: "You fight an ogre."

In my opinion, the failure here is asking the players to make a choice that has no purpose in the first place. Why are you asking them to choose a path when that choice makes no difference?

If the players have decided, we are going to go to the temple of aahz, and you know they are going to have three "random" encounters on the journey just get to it. Skip over the parts where the interaction isn't meaningful.

"Okay, you set out for the Temple of Aahz along the old Yorick highway. Your map shows to travel four days, then turn off the road at an old lumber station. It notes the station is often besot by Ogres. On the third day as you are travelling there is a thundering crack from the trees to the right of the road and several boulders come flinging out at you followed by the war whoops of a pack of Ogres. Roll Initiative"

to me: as a player: this is preferable to...

"Okay you decided to set out for the temple of Aahz. What are you going to do?"

"Uh okay, do we have a map"

"You do."

"How does the map say to get there?"

"It shows the best path to be to follow the old Yorick highway for four days, then turn off the road at an old lumber station. There is a mark for danger at the lumber station."

"What kind of danger?"

"According to the notes in the margin, ogres often infest the station and attack travelers."

"Oh well we don't want that. Are there any other paths."

"Uh, well, you can go through the wilderness but it will take a few extra weeks and you risk running into other monsters."

"Hmmm... well we are in no hurry, let's do that."

"Okay... so you start slogging your way through the woods. After a week of travel you come across signs of recent passage by large humanoids"

"uh oh. Well let's try to veer away from the trail they are taking."

"Okay. *rolls a different random encounter* That night as you are camping you hear rustling in the woods... ten gnolls start throwing javelins at you. roll init."


While the second looks like the players have more agency, so what. you are trading one encounter for another at that point and spending a lot of time and energy on boring stuff to get there.

There is a happy medium here, but it mostly comes from the players recognizing they are playing a game that the GM has to pre-prepare for and the GM to realize that players can get frustrated by being asked to make meaningless choices.

[kyoryu]

While the second looks like the players have more agency, so what. you are trading one encounter for another at that point and spending a lot of time and energy on boring stuff to get there.

There is a happy medium here, but it mostly comes from the players recognizing they are playing a game that the GM has to pre-prepare for and the GM to realize that players can get frustrated by being asked to make meaningless choices.

For some people that isn't the boring stuff.

And, really, that shouldn't take more than a few extra minutes.

Really, though, the QO is a minimal example of a scenario and is often taken too literally. Is "gnolls vs ogres" really that important? Not really. But there's a lot of daylight between "you will take out the Duke, here's the path you will go on, and any deviation from this path will not be tolerated and I will use every technique at my disposal to get you on track" and "yeah, there's an issue with the Duke, what do you want to do about it?" where the answer can be anything from "ignore it" to "convince him" to "get political power to oust him" to "raise an army against him" to "join him, Go Team Evil" to probably a dozen other things I haven't thought of.

[Wintermoot]

For some people that isn't the boring stuff.

And, really, that shouldn't take more than a few extra minutes.

Really, though, the QO is a minimal example of a scenario and is often taken too literally. Is "gnolls vs ogres" really that important? Not really. But there's a lot of daylight between "you will take out the Duke, here's the path you will go on, and any deviation from this path will not be tolerated and I will use every technique at my disposal to get you on track" and "yeah, there's an issue with the Duke, what do you want to do about it?" where the answer can be anything from "ignore it" to "convince him" to "get political power to oust him" to "raise an army against him" to "join him, Go Team Evil" to probably a dozen other things I haven't thought of.

Yes. there is a world of difference. And if that's what you thought I was advocating then you have badly misunderstood me.

[kyoryu]

Yes. there is a world of difference. And if that's what you thought I was advocating then you have badly misunderstood me.

I mean, I'd agree with not offering a false choice for sure.

I'm not sure that means that your example of choosing which path you wanted to take has no value. Especially from a player perspective, and especially if there's a cost to the extra time for taking the longer path.

[HidesHisEyes]

In my opinion, the failure here is asking the players to make a choice that has no purpose in the first place. Why are you asking them to choose a path when that choice makes no difference?

…

While the second looks like the players have more agency, so what. you are trading one encounter for another at that point and spending a lot of time and energy on boring stuff to get there.

There is a happy medium here, but it mostly comes from the players recognizing they are playing a game that the GM has to pre-prepare for and the GM to realize that players can get frustrated by being asked to make meaningless choices.

I agree with you but I do think you’re taking for granted which choices will be meaningful. As GMs we are constantly making choices about choices: which ones to skip over, dwell on, make explicit, make implicit, go meta and discuss with the players, and so on. This is just a fundamental part of RPGs, at least RPGs that have a GM. Really the best we can do is decide what we think will probably be a meaningful choice, present that choice and then respect the players’ response and allow it to shape the narrative.

[Stonehead]

I keep talking about spectrums and continuums. I keep talking about player preferences vary and are not all clumped at 100% agency. This response sounds like a non sequitur. I am not claiming there is a binary between 0% and 100%.



Thankfully you are replying to my post instead of that strawman, and my post did make sense to you. Please take 5 steps back from that binary strawman and address the continuum.


Most players don't like 0% nor 100% player agency. They don't want to listen to an audiobook nor do they want perfect knowledge with omnipotent characters. Find out what your play group wants and aim for that. However knowing players generally don't want the extremes doesn't stop you from understanding what can move your campaign in either direction. Find out your play group's preferences, get player buy in, and be aware of which GM choices impact how well the game matches the playgroup's preferences.



Now on the topic of exploration, I have found that I personally don't like quantum ogres even when I am making a blind choice. I prefer the benefits verisimilitude has on exploration of blind choices. This ties back to what I said about player preferences can be nuanced.

So personally, for my playgroup's preferences, I focus on a sandbox with no "superpositions" and allow the PC power and PC information gathered to let the players calibrate how much player agency they want that campaign. Depending on the group I also give more or less guidance on suggested optional group goals. I can make more obvious goal opportunities if the group wants less agency, or I can ask the PCs about their goals when the group wants more agency. However your playgroup might (probably) has different preferences. So the ideal playstyle for your group might (probably) differ.

Continuums are all well and good, but I'm not super interested in a semantic argument about terms. Like I said before, I'm much more interested in talking about when, and more importantly why it becomes a problem. I originally thought "Railroading" carried the connotation of being a problem that people dislike. If the quantum ogres are a problem inherently, it would have to be for a reason that isn't subverting player agency. Based on your point that no one likes 100% agency, there have to be unknowns, and if there are unknowns, there have to be some degree of blind choices because you aren't forced to go find that thing you don't know exists yet, and if there are acceptable blind choices, is there an issue with shuffling around the consequences of them behind the scenes?

I'm not telling you the answer to that question because I don't know it. We've established that there's going to be some amount of blind choices in an enjoyable game, inherently, because of the nature of free choices combined with unknowns. So putting a "Quantum Ogre" behind those necessary blind choices doesn't subvert player choice at all. I'm not necessarily saying that there's no problem with doing that, but if there is a problem, I want to know why, because to me, it really looks like the problem can't be subverting player choice.

Oh how I wish that was the case. Next you'll tell le nobody ever ragequits Chess.

Like, consider: entire trends of game design have come and gone based on some player disliking what the dice or their game master decided. For a while in boardgame and wargame circles, the dominant opinion was that any use of dice, at all, is bad, because games should depend on skill, not random chance. For a while in the roleplaying game circles, the dominant opinion was that D&D is a bad game, in fact all games with a game master are bad, because decisions should be made by everyone at the table, all authority is illegitimate, so on and so forth.

Losing gracefully, accepting defeat, accepting some decisions are made by others and not you, these are personality-dependent and to a degree, skills. I'm not kidding here. Games are used as teaching tools in schools to literally teach these things to children. Every once in a while, somebody slips through, fails to learn the lesson or forgets it, just as surely as some people fail to learn or forget general solution to quadratic equations. It's not given a person sitting down to play has accepted or even understood what the game rules are telling them.

Maybe I've just been super lucky to only ever play with cool people who I'm friends with, but I'm having a hard time relating to this at all. Like, we're so far off into theory-land that it has no relation to any game I've ever been a part of. Maybe my experiences aren't universal, but every single person I've ever sat down at a table with would be totally fine with the DM saying "You find some wolves in the forest because that's what I rolled on my encounter table".

It's crazy to me that "People playing Dnd are ok with rolling dice" is such a controversial statement.

As a matter of fact, we can't. You haven't even told me if you're a presentist or eternalist. Though I'll argue for both cases that there's no difference between an ogre that hasn't been placed yet and a chess move that hasn't been made yet.

Right, because I'm not super interested in a metaphysical discussion right now. The time and place for that is in a philosophy 1001 class. I'm sure we both have self-consistent views about the nature of future objects, but their only relation to this thread is about where a pretend Ogre is standing.

You seem to be forgetting that the exact location of the ogre isn't determined until the moment the game master puts it in the game. The prepwork the GM has done is not the same action as actually placing the ogre. For comparison a Chess player could actually play and write down sequence of multiple Chess games before facing an opponent. Then, when their opponent makes a move they have predicted and put down in their notes, they can just follow the response put down in their notes. As far as rules of Chess are concerned, the move in the pregame is the same as the move in the actual game, but obviously the move in the actual game isn't done until it's done.

I didn't forget, that's kind of the whole point. The location of this Ogre hasn't been determined, but it's size, and it's hit points, and the silly voice the DM uses when it taunts the party have been determined. So some part of the ogre exists, at least more so than some other thing he has to improvise. And the dragon hoarding gold in his cave has all these things _and_ his position predetermined. The reason "superposition" is being used here, is because this one object seems to exist and have a location, while this other object seems to exist, but it's location hasn't been determined.

It really does just come down to whether you're viewing DnD as a game, or as a fictional world. A chess move you haven't made yet doesn't seem to exist in the way one you already have made does. But in a movie, a character doesn't "cease to exist" when they step off screen. In the canon of the story, characters "exist" before they appear on screen.

If you don't think Luke Skywalker "existed" in the Star Wars universe before the droids left in the escape pod, that's a reasonable, self-consistent stance to take, but so is the stance that he did exist, in the fiction of the story. I'm not telling you to look at DnD the way I look at a movie, but you can't make anyone else look at DnD the way you look at chess. And even if we were all "fooling ourselves", this isn't the kind of view that gets changed based on forum posts on the internet.

[kyoryu]

The point is that "Do we fight the wizard" is not the most interesting question here. It's not that "The PC's fight the wizard" is evil railroading, and "The PC's don't fight the wizard" is Player Agency.


The story doesn't end when the PC's show up at the top of that tower to fight the wizard. The PC's choices shaped how that fight goes.

In the "Railroad" metaphor, the PC's are not passengers, they are The Train (I guess they could be passengers stuck on the train?), constrained by the tracks, following a pre-determined path to their destination.

Yes, a Wizard Fight will always happen, but, depending on the PC's choices, It won't be the same wizard fight.

The PC's who sneak up the side of the tower are not the same PC's who fought their way through the interior. The Wizard who is at the cusp of completing his dark ritual is not the same wizard who lost his staff when the PC's convinced his apprentice to betray him.

The burning Tower is not the same Tower as one that isn't on fire.

Also, if the players agreed to play a game about hunting down the Wizard, then giving that to them doesn't impact agency at all. If the game is pitched as "be people in fantasy land" and the GM decides ".... and you will hunt the wizard, regardless of your desire to do so" then there's a railroading claim.

Imagine you are at a restaurant, you sit down, you are handed the menu.

Railroading is when the menu just says "Spaghetti", only Spaghetti. Your only choices are to leave, or eat the Spaghetti.

If the Menu says more than Spaghetti, that's not a railroad (Or, less railroady). Yes, you'll have a meal, but the nature of that meal will change.

Even worse is if you order lasagna, but get spaghetti. Or you go to the restaurant next door for sushi, but still get spaghetti.

[OldTrees1]

Continuums are all well and good, but I'm not super interested in a semantic argument about terms. Like I said before, I'm much more interested in talking about when, and more importantly why it becomes a problem. I originally thought "Railroading" carried the connotation of being a problem that people dislike. If the quantum ogres are a problem inherently, it would have to be for a reason that isn't subverting player agency. Based on your point that no one likes 100% agency, there have to be unknowns, and if there are unknowns, there have to be some degree of blind choices because you aren't forced to go find that thing you don't know exists yet, and if there are acceptable blind choices, is there an issue with shuffling around the consequences of them behind the scenes?

I'm not telling you the answer to that question because I don't know it. We've established that there's going to be some amount of blind choices in an enjoyable game, inherently, because of the nature of free choices combined with unknowns. So putting a "Quantum Ogre" behind those necessary blind choices doesn't subvert player choice at all. I'm not necessarily saying that there's no problem with doing that, but if there is a problem, I want to know why, because to me, it really looks like the problem can't be subverting player choice.

Clarification noted. Here is a clarification on my end too.
Railroading sometimes does and sometimes does not carry the connotation of necessarily being bad. This depends on if they are describing what is mechanically happening or if they are labeling what they dislike and then trying to explain when the events cross that threshold. I am use Railroading in the neutral tone because I find it easier to focus on the mechanics rather than redefine the word each time a player has a slightly different preference.

I do not think quantum ogres are inherently a problem. I do inherently dislike them and I can/will elaborate why, but there are players that appreciate quantum ogres. There are even players that appreciate quantum ogres for basically the direct inverse reason of why I dislike them.



The most common objection players raise (based on listening to different players on this forum) is comparing a quantum ogre to an informed impactful choice even though there are two changes between those cases (information & impact). If you have a blind impactful choice you could convert it into an informed impactful choice (granting player agency), leave it as a blind impactful choice (no agency over that aspect), or double down and make it a blind non choice. So often players object to quantum ogres because you could have offered an informed impactful choice instead. This objection is more common / louder if the choice was about something the players cared more about. A quantum ogre with nothing special is going to be criticized less than a quantum ogre carrying the next part of the plot device.

Also most players appear to be okay with some shuffled blind choices but not others. I think this scales with how much the player cares about the consequence being shuffled. The more the player cares about the consequence, the more they tend to prefer a blind impactful choice over a blind quantum non choice. Of course the more the player cares about the consequence, the more they prefer the choice be informed instead of blind.


Personally I have an additional (non agency) criticism of quantum ogres. I like verisimilitude and quantum ogres (assuming no in fiction explanation) feel like a metagame glitch imposed on the world rather than a natural consequence of the actions of various characters existing in that world. This kind of verisimilitude critique of quantum ogres is more common from players that have preferences towards the sandbox end of the continuum because the more reliable the verisimilitude of the campaign world, the easier it is to have informed choices. So I don't want the blind choices to be shuffled and thus turned into blind non choices. I prefer they remain blind choices (with the balance of informed vs blind choices depending on my playgroup's preferences).



Finally there is a 3rd consideration. The dishonest GM and historic associations. The phrase "Quantum Ogre", despite Vahnavoi's critique of its scientific accuracy, was coined in response to a GM that wanted to railroad their players more than the players claimed they wanted to be railroaded. So they bragged about how they could use perfect illusions to trick the players into playing the game that had less agency than the game players wanted to play. They did this as part of their thesis that player preferences don't matter because the GM understands what players enjoy more than the players do. So like "Railroad" there are potentially negative connotations attached to the term. This might make a player dislike a quantum ogre more than disliking separate ogres on each road.


Those are the 3 main critiques. The first is about a lost potential for an informed meaningful impactful choice. The second is about the decreased verisimilitude and its impacts on the potential for other informed choices. The third is about the historic associations with illusionism to abuse player trust. Which means quantum ogres are perfectly fine as long as the playgroup is okay with them.

[kyoryu]

Which means quantum ogres are perfectly fine as long as the playgroup is okay with them.

As with all things.

As I've said many times, having a linear game is great - just tell the players that's what they're getting. Once everyone is on that page, do what you need to do (within whatever limits people agreed to)

[Talakeal]

I use the same logic to explain quantum ogres and other "contrivances" as I do with HP and rerolls; we don't tell the stories about people who led ordinary lives; therefore PCs are always getting into unlikely but interesting situations as a sort of narrative survivor bias.

[Quertus]

All fiction is built on "contrivances" and I would postulate that IRL the people who managed to change the world all had more than their share of "contrivances" occur to help them along.

In a linear game if you told me you wanted to kill the gods and reformat the world I would make sure there is a path to do that. In a sanbox, I shrug, tell you that on the off change the game is still happening in ten years and you haven't already been killed (or had someone else beat you to the punch) then maybe you can do that when you are level 40; but until then it just ain't going to happen without a "contrivance".

And this is why you start the party at level 50, so that they have the agency to pursue their goals during the campaign. (Color blue to taste)

More realistically, if there's a "McGuffin of god slaying" that you would have worked into the hands of the linear party, nothing prevents the sandbox party from getting it… either/too (I'm pretty sure one of those words can finish that sentence). More generally, whatever method the linear party could use, could be used by the sandbox party. Now, despite my "creativity", I might never come up with and successfully implement a valid "god-slaying" strategy, especially under a GM whom I lack sufficient ranks in Knowledge: GM, or who lacks creativity themselves, and can only think in terms of "the only way to…".

So… if your world has the McGuffin of god slaying, that you would have made sure fell into the laps of the low-level linear party, then the sandbox party could investigate rumors, peer through space, time, and narrative causality, or otherwise track down said relic, and use it to slay gods themselves. A little later than the linear party, perhaps, but a well-deserved victory, free of contrivance of the McGuffin just so happening to fall into their hands.

That said… I'm perfectly happy with "T=0" contrivances, like, "grandpa *made* the god-slaying McGuffin, and bequeathed it to me in his will". I'm choosing to play *that* particular farm boy / 7th son of a noble / whatever generic background turned adventurer rather than *some other* random character turned adventurer.

That said, regarding "people who managed to change the world all had more than their share of "contrivances" occur to help them along"? I'm told (I hope I've got this right - any Playgrounder know what I'm talking about, and care to be awesome and provide a reference?) that… after the Holocaust, it was found that one town did an amazing job evacuating people. So researchers went to investigate, to find out *why* this town did so well, what their secret ingredient was. And what they found was… nothing. That, as far as they could tell, anyone could have done the same thing, they just… didn't. That the existence of this town wasn't a road map for success so much as it was condemnation to everyone who had failed.

So I don't think everything happens through contrivance. For example, what contrivance (beyond "being smart") allowed my dad and myself to be among the first in the world to report solving our respective mental challenges?

Huh. Apparently, I am a figurative "7th son of a 7th son", as I believe we were each the 7th person to report our respective findings. Other than "rules lawyer cred", what abilities in what worlds might that grant me?

Thats why I prefer the Alexandrian's "the GM negates a player’s choice in order to enforce a preconceived outcome". Because it requires a player choice first, then a DM action to negate it after, with a purpose in mind. Or to quote from the first paragraph of his article: "The players must try to get off the train and the GM has to lock the doors."

I like definition for its simplicity, but mine is both a little more broad and a little more narrow: "('Railroading' is when)… the GM changes established facts or negates game physics… to force or prevent a particular outcome.".

On that note,

The reason I dislike that is…

Every Railroader thinks they've done enough prep work/ thinks that their story is so good the players won't try to go off the rails. The Alexandrian's definition is what happens when they're Wrong.

How do y'all feel about my definition? (@BRC, I'm hoping that your answer will let me understand your post / what you disliked about Alexandrian's definition. Reading comprehension continues to be… not my strong suit.)

This problem is pretty solvable, I feel like, if you design scenarios/situations rather than plot things out. Example:

Last session of my game pirates had kidnapped a merfolk and a clutch of eggs and the PCs decided to rescue them. I designed the small island stronghold that the pirates were stationed at, figured out how many pirates would be lurking about, where the eggs were, and where the merfolk was. I did not plan or even consider how they would actually rescue the merfolk. They ended up giving her a shrinking potion that I'd given them months ago and yeeted her out a window back into the sea.

Awesome! That's what I like to see. Moments like that are what I GM for.

This may work for you, but it doesn't work for every GM. Because it can lead to a very static bland experience depending on how fast the GM thinks on their feet.

What I mean is this. If you haven't put any thought into how the scenario may change or grow depending on how the players respond, then if they come up with a solution that bypasses everything then it turns out to be a very boring game.

So you detailed the merfolk village. You may have put some additional loot in or areas of interest for them to find. You have have inserted a hook for a future session or story. But if they bypass all of that with the shrink potion and through the window, the entire session takes 30 minutes to do and ends up very non-fun for both players and GM.

Granted, this isn't ALWAYS true, but is SOMETIMES true. Some GMS run a better game by being over pre-prepared so they don't have to do so much thinking on their feet. Some Players enjoy a game where it evolves as they go to give a satisfying rich experience.

As always, your mileage may vary.

Uh… that *scene* might only last 30 minutes (and possibly be the best 30 minutes of the campaign), but the *session* can continue beyond that scene, and, coming off the high that scene, the session is IME *more* likely to be fun for everyone.

[BRC]

I like definition for its simplicity, but mine is both a little more broad and a little more narrow: "('Railroading' is when)… the GM changes established facts or negates game physics… to force or prevent a particular outcome.".

On that note,

How do y'all feel about my definition? (@BRC, I'm hoping that your answer will let me understand your post / what you disliked about Alexandrian's definition. Reading comprehension continues to be… not my strong suit.)

Your Definition is pretty close to the Alexandrian's definition, so I dislike it for the same reason, it puts "Railroading" (Generally considered to be a bad thing) at the moment the GM has to change things in-session.

If A new GM comes up to you and says "How do I be a good GM and not railroad" and you say "Never change established facts or negate game physics to force or prevent a particular outcome", that leaves them room to take away the bad lesson that how one is a "Good" GM is to exclusively create watertight scenarios, where their desired outcome happens without them needing to change any established facts or negate any game physics. That's where you get the PC's being babysat by superpowered NPC's who exist to win all the fights that need winning and clap the PC's on the ear if they ever deviate from the intended path.

We go on and on about railroading being bad because giving players agency is good. If your definition of railroading focuses on the moment you force the players back onto the tracks, that doesn't actually contain the lesson of "Give players agency".

Your definition is a good one, and could be considered part of a wider piece of advice, E.g:

1: Give your players the ability to make meaningful choices.
2: Don't Railroad (change established facts ect ect).

But the distinction between "The GM intervened to force an outcome" and "The GM created a scenario with only one possible outcome" is pretty academic in my mind? If we're going to be defining "Railroading" as the thing to not do, I'd like to include both.


Edit: Note my definition is "Give the Players Agency" not neccessarily "Give the PC's agency". If the Players buy into the scenario in which their characters are forced to do something, that's fine.

[Easy e]

I keep coming around to this simple question:

What is a meaningful choice?

I have a definition I have shared in this thread. What is everyone else's definition?

[JNAProductions]

I keep coming around to this simple question:

What is a meaningful choice?

I have a definition I have shared in this thread. What is everyone else's definition?

Mind sharing it again? I missed it, or have forgotten it.

[Vahnavoi]

A meaningful choice is one that changes the game state and cause the game tree to begin, diverge, converge or end.

A meaningless choice is one that doesn't influence anything outside itself.

In a complex game with incomplete rules, distinquishing between meaningful and meaningless choices can be non-trivial. However, a good rule of thumb is that in a human run game, in order for a choice to be meaningful, people have to pay attention, because their minds are the medium for the game.

For games where a game master has final say on the state of the game, the fiction of "Game master is God, everything exist by God's will" can provide one useful heuristic. A choice is meaningful when God permits it and allows it to change the world.

[jayem]

Some 'real life' situations are naturally railroady, some are naturally Schrodingery, some are naturally open and similar. And it obviously makes sense to try and play to their strengths.
It's probably better to say they are a combination of layers.

E.G my Grandad's school day was proscribed (limiting)
Well he could bunk off (freedom)
But the teachers would try to catch him (limiting)
but he could plan and avoid that (freedom)
And whatever he did, that school was getting bombed on X/X/1941 (limiting).

And if it were a game, the players would pretty much pick up if that was the game being the game, or railroads/etc...

__

On the dungeon, this would also apply.

Yes there are totally some aspects that are just naturally begging for Schrodinger to help to lighten the burden, and it would be totally wasteful to duplicate things. (The original Gollum would be an ideal thing to put in whatever tunnel you run down, rather than going "in this tunnel is Gollum, and in this tunnel is Phlegm and he's survived using his heavy scale armour, and in this tunnel are some vines that...")
Yes there are some aspects where it would totally over-rule player choice and it would be damaging.
In between you have some that are hard to call, and it depends on what the game is. (Having the One Ring just happen to be in that tunnel, for instance)

Similarly with the other types of rails, there are some areas that they fit nicely (Caradhras/Moria for instance) and some where they really don't.

__
Regarding Meaningful choices
I think the key thing is that it is worth the players time to make the choice (and the GM's time for the player to).

1) The players must have at least 2 options (a choice)
2) The outcome of the picking different options should be different (it's not meaningful if you have Hobson's choice)
2b) .. and sufficiently valuable and long term to be of interest
3) The outcomes can not be ranked by some single 'best' metric (it's not meaningful if the choice is cake or death)
4) The player should have some idea what the expected outcomes are (it's not meaningful if the player may as well roll a die)

I would say that not all choices have to be fully meaningful, it's a starting point.