The Nature of Railroading

[zinycor]

But why? What is the ''look" of a level?

A bandit might be wearing say: fur boots, leather armor, a cloak with a hood and have a sword. And, really, that will describe a bandit of any level.

Because creating expectations is the job of the GM. Especially for combat encounters. Players should be able to tell if a combat is going to be hard or easy (unless you on that particular case are going for a surprise or deception), and prepare if they want.

[Thrudd]

Well, sure this is a big problem: when players see anything they simply don't like, they will cry railroad.

The thing is, it's not even close to a railroad...it's just the players using a Trigger Buzz Word. You know, it's a lot like that loser up at bat that swings and misses the ball...and then whines and cries that they ''were not ready" or ''the sun got in their eyes" and that the strike should not count.

I agree. That's not railroading, that's called "whiny players." That they use this buzzword doesn't make the claim accurate.

GM's have to prep scenarios, and there's an acceptable gradient of planning/linearity vs open world/sandbox. One of the earliest and most traditional ways of playing D&D is for the DM to tell the players that they will be exploring X dungeon today: ie "you find yourselves at the entrance to a cavern," and expect the players to go along and explore the dungeon. Because that's the game. Of course, this would have been expected by the players, they know that to be the format of the game.

So, as almost always, it comes down to setting expectations and communication. Simple as that. Instead of struggling and hoping to get players to go where the prepped material is, you can set up an episodic game, where non-adventuring time is minimized and participating in each session's adventure scenario is assumed with a DM-fiat pre-game set-up. Adventures can start in-media res if you want- a shipwreck on an island, waking in a prison, at the entrance to a lair with a dictated goal of finding am artifact, etc. If the players know this is how the game works and agree to play it, there should be no complaints of railroading (unless you railroad them during the adventure in some other way.)

If one wants a totally open-world game with player-directed quests and adventures, there is necessarily going to be quite a bit of improv on the part of the GM, in addition to quite a bit more prep. You should anticipate players going off in any/every direction and may need to invent things on the spot when they inevitably go somewhere you didn't prep for.

I think the best mix is a totally open environment of limited and predictable scale, with expectations for the game set in such a way that there won't be confusion about how and where the adventures will take place.
IE, You're all treasure hunters who explore dangerous ancient ruins hoping to get rich (with whatever personal motivations for doing that), with a base of operations at the border town of the kingdom. Travelers and explorers pass through town with rumors and news of places that are promising loot sites. Collect rumors and intel, pick a destination, stock up on supplies and begin your expedition. That's the game. I can expand the scope of the game any time I want to, but to start the players would know exactly what's up. If you don't want to be a treasure hunter, too bad. That's what the game is about.

[Xuc Xac]

Take for example two scenarios:

Scenario A:
The PCs witness a group of low level bandits raiding a village. The DM knows with 99% certainty that the players will attack the bandits, and barring horrendous luck or terrible decisions defeat them easily, kill or drive them off, and take their loot.

Scenario B:
The PCs witness a group of high level bandits raiding a village. The DM knows with 99% certainty that the players will attack the bandits, and barring incredible luck or brilliant decisions the players will likely be forced to retreat or beaten up and left for dead. The players are then almost certain to regroup, plot revenge, prepare a bit better, and then track the bandits down, kill them, and take their loot.

Neither one is railroading.

Scenario A would be railroading if the GM decided in advance that the PCs would attack and win in spite of any "horrendous luck or terrible decisions". If the players roll badly or choose to do something stupid, the GM will change the facts so they win anyway. "Oh, you only rolled a 1 for damage? *looks down at notes, sees Bandit Chief has 12 hp left, crosses out the 2* The bandit leader squeals like a pig and looks panicked. He was all show."

Scenario B would be railroading if the GM decides in advance that the PCs will attack, lose, get left to regroup, then attack again and win.
PCs see that the bandits are too much for them and try to sneak away? "The bandits see you and chase you. Roll for initiative!"
PCs get some lucky rolls that should turn the battle in their favor? Well, maybe instead of average hit points, the bandits should have max hp instead... "Wow, that was a good shot, but these guys seem tougher than average."
PCs get beaten and left for dead as planned but decide to cut their losses and run instead of regrouping and seeking revenge? "Uh, after you heal up and start heading down the road--away from the bandit hideout?--uh, the bandits appear on the road behind you! They came back to finish the job. You run? Well, they all have boots of jaunty skipping (with one charge left so you can't loot them after you win the battle) so they catch up to you. Roll for initiative!"

If you're hiking, a trail through the woods is not a railroad, because you can turn and walk into the forest. You don't have to follow the trail.

If you're driving a jeep through the desert, a road is not a railroad, because you can turn and drive cross country if you want. Staying on the road might be the best option, but you can still do something else.

If you are riding a train, the railroad is a railroad because you can't leave the tracks. That's why game scenarios with a predetermined outcome are called railroads: you can't choose to go another way.

[Quertus]

Because creating expectations is the job of the GM. Especially for combat encounters. Players should be able to tell if a combat is going to be hard or easy (unless you on that particular case are going for a surprise or deception), and prepare if they want.

If I grab a random Knight from history, and Teleport him and, say, Bruce Lee next to each other, will they know how dangerous each other is? Opinions vary. I've stood in the same room as numerous dangerous people - some of them, I could tell; others, I couldn't. Shrug.

The Simulationist says that everything should appear exactly as dangerous as it appears. But opinions vary on what that means.

Conventional wisdom says err on the side of giving the players too much information, giving it repeatedly in different ways, rule of 3, etc. But that requires a baseline of how much is "enough" to begin with.

The Narrative gamer cares about what makes the best story, but what is that? Is it the players always acting knowledgeably, or getting surprised?

The Gamist player would care about game play - but, again, public vs hidden information both have their place in games (Chess vs Stratigo, for instance).

The Rules Lawyer says that we should follow the rules - but what system we're playing will change whether the rules cover this. And the OP is making his own system, so who knows what the rules say.

The CaS player thinks both of these positions are idiotic, as of course the bandits have to be exactly dangerous enough to be a "sporting challenge"; otherwise, why would the GM waste their precious gaming time on them? If the GM didn't calibrate the encounter exactly right, to ensure the exact correct amount of resource expenditure, they'll tell their mom.

The CaW player knows that any one of those bandits could actually be an ancient dragon polymorphed into human form, or a demon host, or any number of other things, and wants to spend the next 5-6 sessions carefully researching the background of each bandit while observing their actions for any signs of unusual behavior before choosing whether or not to engage them. Oh, wait, the GM bothered to describe their clothes? Better make it 7-8 sessions, and have at least 3 emergency contingency plans readied, just in case.

In short, there's lots of ways to play the game, but, afaict, none of them mandate that the GM absolutely *must* telegraph the danger of a challenge. So, IMO, it feels wrong to call BadWrongFun on a GM for not doing so.

[Pippa the Pixie]

Because creating expectations is the job of the GM. Especially for combat encounters. Players should be able to tell if a combat is going to be hard or easy (unless you on that particular case are going for a surprise or deception), and prepare if they want.

I don't agree this is a job for the DM.

The players should really have no idea how hard or easy a combat could be. It's really the worst sort of metagaming.

You really can't see ''levels". What is diffrent between a low level person and a high level person? They will look the same, and their equipment would look the same.

Exactly what do you do to telegraph levels in your game to make things easy for the players?

[zinycor]

I don't agree this is a job for the DM.

The players should really have no idea how hard or easy a combat could be. It's really the worst sort of metagaming.

You really can't see ''levels". What is diffrent between a low level person and a high level person? They will look the same, and their equipment would look the same.

Exactly what do you do to telegraph levels in your game to make things easy for the players?

We agree to disagree then

[Pippa the Pixie]

We agree to disagree then

Sure.

I still wonder how you describe a human, for example, of different levels. What does your 10th level human bandit look like that makes them SO different then the 5th level bandit? How do you even describe say 1st level human bandits vs 5th level human bandits?

I can't see anyway yo do it myself unless your just going to metagame and tell the players.

[Xuc Xac]

Sure.

I still wonder how you describe a human, for example, of different levels. What does your 10th level human bandit look like that makes them SO different then the 5th level bandit? How do you even describe say 1st level human bandits vs 5th level human bandits?

I can't see anyway yo do it myself unless your just going to metagame and tell the players.

Look at the opera house battle in the movie "The Fifth Element". Corben Dallas is cool, calm, professional, and deadly accurate. Ruby Rhod is screaming, panicked, and flailing about. The Mangalore soldiers are obviously trained in the use of their weapons, but they're also obviously regular soldiers and not elite special forces.

Or go to a boxing gym and watch people in the ring. It should be obvious who is new to the ring and who is an experienced fighter.

[Resileaf]

Look at the opera house battle in the movie "The Fifth Element". Corben Dallas is cool, calm, professional, and deadly accurate. Ruby Rhod is screaming, panicked, and flailing about. The Mangalore soldiers are obviously trained in the use of their weapons, but they're also obviously regular soldiers and not elite special forces.

Or go to a boxing gym and watch people in the ring. It should be obvious who is new to the ring and who is an experienced fighter.

Yes, but how do you tell when they are not fighting at the time. When they're sitting around reading the newspaper?

[Pippa the Pixie]

Look at

Sure you can look at the vague context clues...and be wrong about half the time or more.

And, sure, if you watch a character fight in game....and cheat and metagame....then sure you can figure out the level, or CR or power of a character. If you want to play the game that way....

[Xuc Xac]

Yes, but how do you tell when they are not fighting at the time. When they're sitting around reading the newspaper?

That's irrelevant. The two situations presented were both about seeing bandits attacking a village and deciding to get involved.

Sure you can look at the vague context clues...and be wrong about half the time or more.

And, sure, if you watch a character fight in game....and cheat and metagame....then sure you can figure out the level, or CR or power of a character. If you want to play the game that way....

Just because you're wrong over half the time, that doesn't mean everyone is that bad.

And figuring out how well a character fights by watching them fight is not cheating or metagaming. Cheating would be sneaking a peek at the GM's notes to read the enemy's stats during a bathroom break. Metagaming would be doing things like "he hit my AC when he rolled a 12 so he must have at least a +5 to hit and the DM rolled an extra 2d6 when he flanked me so he's got some Rogue levels", but not "he hits me a lot more often than he misses and he does a lot of damage when he hits, so this guy is more dangerous than his buddies".

I don't like to play idiots. I like to play perceptive warriors. A gunslinger who stares an enemy in the eye during a poker game to gauge how steady his nerves are. A detective who watches the way someone moves and listens to their accent to recognize that they were trained by the British military. A wandering kung fu warrior who estimates another warrior's strength and skill by politely insisting that he take the last dumpling at dinner and engaging in a minor duel with chopsticks to put the dumpling in his bowl when he politely refuses and tries to turn it away.

[Resileaf]

That's irrelevant. The two situations presented were both about seeing bandits attacking a village and deciding to get involved.

Yeah, but you're not watching the bandits fighting, you're watching them cutting down helpless peasants.

[Quertus]

That's irrelevant. The two situations presented were both about seeing bandits attacking a village and deciding to get involved.

Yeah, but you're not watching the bandits fighting, you're watching them cutting down helpless peasants.

Oh. Reading comprehension is not my strong suit. Yes, if the bandits are fighting, and *all* the bandits are fighting, and the party watches long enough to see how well all of them fight, then, yes, I would agree that the party could accurately judge the combat capabilities of the bandits.

But, then, I've only ever seen one PC played that way. I was the player, and I spent my every action just making observations (while hiding) and asking the GM questions while the other PCs ("village heroes") fought the orc invaders. I never got involved in the opening fight (which lasted long enough for one hero to have fumbled his long sword, short sword, and 3 daggers. I think he was fighting with a rock & a pointy stick by the end).

Most PCs are made of more impulsive stuff.

[Xuc Xac]

Yeah, but you're not watching the bandits fighting, you're watching them cutting down helpless peasants.

Do the bandits hold their swords properly and cut with correct technique while covering each other and watching for any surprise counterattacks while working together as a team like a serious military unit?

Or are they a bunch of uncoordinated individuals who hold their pistols sideways and one-handed to look cool swords point down and swing wildly with a lot of unnecessary flourishes because they don't take the peasants seriously and aren't keeping their guard up in case a real threat appears?

Even if they are unopposed, you can tell how skilled they are. Even if they use the same gun to shoot the same target on a practice range, you should be able to easily tell a special forces soldier from a new recruit.

[Lorsa]

After having thought about it a little, I think I've reached a somewhat decent description of railroading in terms of [b]kyoryu[b]'s premise of basic play where the GM presents situations which the players then have to navigate somehow.

Under this model, railroading occurs when the GM presents a "fake" situation. That is, a situation which is only meant to lead to a different situation, "the real situation" if you wish. This sequence can last for any number of situations. For a purely railroaded adventure, maybe the real situation is only the boss fight. But rather than simply introducing that, the GM first drags the players through 30 or so fake situations with an already determined outcome.

Another way in which railroading occurs is when the GM forces the same situation to happen again, after the players solve it in an unforseen or undesired way. Like if they have planned the situation to be a hard struggle to travel past a mountain range and when it turns out the players simply fly over, somehow changes the situation, or presents a very similar one in order to force it on the players. Rather than simply coming up with an entirely new situation.

Which means that the best way to avoid railroading is to simply present situations with no plan for what the next situation will be, and without forcing the same situation on the players if they avoid it in an elegant way.

[King of Nowhere]

But why? What is the ''look" of a level?

A bandit might be wearing say: fur boots, leather armor, a cloak with a hood and have a sword. And, really, that will describe a bandit of any level.

Not really. At second/third level they will have some masterwork gear. At higher level they will have magic items. Unless you fluff that magic items are absolutely indistinguishable from mundane trinkets, the difference should be clear.
Or unless you have a 20th level bandit still working for copper.
Or unless this specific bandit has taken a vow of poverty. Maybe he just robs people to donate the money to the fund for victims of robbery :)

EDIT: another hint of someone's level are scars. Someone with a couple of nasty scars is likely a veteran of several battle.
On the other hand, someone without scars may be receiving high level magical healing regularly.
In my campaign world, if somebody has nice equipment but no scars, he is likely level 10+.

[Earthwalker]

Weather I think the original post is railroading or not seems unimportant. What is important is the player felt this and so working out why and what can be done should probably be the next step and hopefully a chance to learn. For GM and Player.

Why does he think it was railroading and also the question of, what should have been done different can reveal a lot I suspect?
Now the player seems to be of the standard stock that Talakeal gets, and I am sorry for your troubles even if your posts do keep me entertained reading them.

[Snipped for length]
Scenario A:
Normal Level Bandits
Scenario B:
Tough Bandits

Again why would the players think it was railroading is a good question to ask.
Oddly in this scenario it brings up a lot of questions for me on the nature of player agency and how players can find out information.

Also on what rules are in place for running away from a losing fight.

So can the players tell how tough the bandits are ? Is it simple ? Is finding out part of the game or are the players just told by the GM these are lvl 7 bandits and the players are lvl 3. (I need to say I have no problem with this, its beta gaming but the advantage of levels it’s a simple shorthand of how tough something is, so why not use it when trying to get information across of how tough something is)

If the players are unable to find out how tough the bandits are, can they make meaningful choices about what to do?

If they can’t make meaningful choices, then it impacts on player agency.

[Quertus]

Oddly in this scenario it brings up a lot of questions for me on the nature of player agency and how players can find out information.

Also on what rules are in place for running away from a losing fight.

So can the players tell how tough the bandits are ? Is it simple ? Is finding out part of the game or are the players just told by the GM these are lvl 7 bandits and the players are lvl 3. (I need to say I have no problem with this, its beta gaming but the advantage of levels it’s a simple shorthand of how tough something is, so why not use it when trying to get information across of how tough something is)

If the players are unable to find out how tough the bandits are, can they make meaningful choices about what to do?

If they can’t make meaningful choices, then it impacts on player agency.

Huh. This definition of "player agency" as the ability of the players to intentionally make informed, meaningful choices seems optimal. Defining "railroading" as the act of removing player agency seems optimal. Which would imply that your definition of railroading would be optimal.

After having thought about it a little, I think I've reached a somewhat decent description of railroading in terms of [b]kyoryu[b]'s premise of basic play where the GM presents situations which the players then have to navigate somehow.

Under this model, railroading occurs when the GM presents a "fake" situation. That is, a situation which is only meant to lead to a different situation, "the real situation" if you wish. This sequence can last for any number of situations. For a purely railroaded adventure, maybe the real situation is only the boss fight. But rather than simply introducing that, the GM first drags the players through 30 or so fake situations with an already determined outcome.

Another way in which railroading occurs is when the GM forces the same situation to happen again, after the players solve it in an unforseen or undesired way. Like if they have planned the situation to be a hard struggle to travel past a mountain range and when it turns out the players simply fly over, somehow changes the situation, or presents a very similar one in order to force it on the players. Rather than simply coming up with an entirely new situation.

Which means that the best way to avoid railroading is to simply present situations with no plan for what the next situation will be, and without forcing the same situation on the players if they avoid it in an elegant way.

I mean, it can be irritating, to keep fighting ever-stronger versions of bandits until the party *has* to use the McGuffin to defeat them, like they're "supposed" to.

But the funny thing is, the reason that the GM has to pull out more, stronger bandits is that they *aren't* railroading the individual encounter - they're giving the players full agency to deal with that encounter however they want, without changing game physics.

That they are then changing facts of how many bandits of what power level exist at what location is where the rails form.

[Lorsa]

I mean, it can be irritating, to keep fighting ever-stronger versions of bandits until the party *has* to use the McGuffin to defeat them, like they're "supposed" to.

But the funny thing is, the reason that the GM has to pull out more, stronger bandits is that they *aren't* railroading the individual encounter - they're giving the players full agency to deal with that encounter however they want, without changing game physics.

That they are then changing facts of how many bandits of what power level exist at what location is where the rails form.

I am not sure if you agree or disagree with me or not. We both conclude that there is railroading (it seems), and we both conclude that it will appear different in play compared with where a situation is really just "fake" in order to lead the players to the next one.

[zinycor]

Sure.

I still wonder how you describe a human, for example, of different levels. What does your 10th level human bandit look like that makes them SO different then the 5th level bandit? How do you even describe say 1st level human bandits vs 5th level human bandits?

I can't see anyway yo do it myself unless your just going to metagame and tell the players.

From the top of my head:

1- Reputation: A 10 level bandit would probably be well known.
2- Gear and equipment: There should be substantial difference in the equipment used by a level 1 bandit and a level 10 bandit.
3- Techniques: any martial character should be able to get an idea of how proficient someone is with a weapon, even if he is only fighting villagers. Hell, even if he is only doing cutting practice.

This is all not to say that the players would get know the exact level/CR of the opponents, but to get an idea of the encounter's difficulty. To get back at the example, in it the GM expects with 90% certainty that the players would be defeated, if so, the players at least the players should get a sense of it.

[King of Nowhere]

After having thought about it a little, I think I've reached a somewhat decent description of railroading in terms of [b]kyoryu[b]'s premise of basic play where the GM presents situations which the players then have to navigate somehow.

Under this model, railroading occurs when the GM presents a "fake" situation. That is, a situation which is only meant to lead to a different situation, "the real situation" if you wish. This sequence can last for any number of situations. For a purely railroaded adventure, maybe the real situation is only the boss fight. But rather than simply introducing that, the GM first drags the players through 30 or so fake situations with an already determined outcome.

Another way in which railroading occurs is when the GM forces the same situation to happen again, after the players solve it in an unforseen or undesired way. Like if they have planned the situation to be a hard struggle to travel past a mountain range and when it turns out the players simply fly over, somehow changes the situation, or presents a very similar one in order to force it on the players. Rather than simply coming up with an entirely new situation.

Which means that the best way to avoid railroading is to simply present situations with no plan for what the next situation will be, and without forcing the same situation on the players if they avoid it in an elegant way.

interesting concept. However, more restrictive than it should be.

take the "fight" example. You are ambushed by assassins. You defeat the enemies through a cunning strategy.
Now, assuming that you are still dealing with the same big bad, or that you are dealing with a new big bad but they made their research, it is completely reasonable that you'll face a new group of assassins. And equally reasonable that they will be prepared for whatever strategy you used last time.
When my players curbstomped an ambush by a larger number of mid level character with four (quickened) firestorms, the next group (that happened a few weeks late) had protection from fire, and none of the players complained; they were all "right, it makes sense that they would do that".

Take the "mountain" example. Ok, you can fly past mountains without effort. At this point, it becomes completey pointless to present you with more mountains. I either give them some feature than you cannot fly over, or I don't give you an obstacle in the first place. As DM, my job is to challenge the party, and that requires adapting to the party strategies.
Now, of course if the DM withdraws the prize for getting over the mountains the first time because he wanted them to pass them in another way, that's railroading. but simply presenting a new obstacle that is "like your old obstacle, but without that weakness that let you exploit it the other time" is perfectly legitimate, especially if you don't do it straight away. Sure, a good DM will try to vary the obstacles, but in along campaign reciclying a few old ones is acceptable. And not all DM have a huge creativity to always come up with something new.
Furthermore, having always something new riisks making the world less cohesive. You keep introducing new stuff instead of delving deeper into the stuff you have, which is what I'd rather do.

Finally, for the "it only has one possible outcome", sometimes it's unavoidable. And if there is enough freedom to create meaningful consequences along the way, it may not be bad.
In you given example, the "real" situation is only the boss fight. but then, this applies to pretty much any villain. You know that when your party picks an arch-enemy, it will lead eventually (almost always) to a boss fight. it is not railroading, because you still influence everything else. Did the villain destroy that little village, or did you save it? did you leave in charge the former henchman, who is evil but makes the train run on time, or do you give a chance to someone more honest but less competent? did you sway to the side of good that miniboss who had somewhat nobler motivations than his peers?
I'm thinking of the Mass Effect 3 videogame. You have lots of choice, and in the end, no matter how you get there, no matter what you did before, you get three basic choices, and that was it, and I read some criticism for it becauuse "the choices you made before were pointless". but they were not. Did you help the quarians, or the geth? did you manage to persuade them to stop their war? did you cure the genophage, leaving wrex in command of the krogan? or maybe you cured the genophage, but wrex died and left his dumb brother in charge? or maybe you refused to cure the genophage because you were afraid of what they'd do? how many of your friends did you manage to save? when you had to choose, which one you left to die?
those are all the choices that matter, and it means nothing that in the end they do not affet the three options you are given. It does affect that the new galaxy will still have both quarians and geth, which it wouldn't have if I didn't persuade them to make peace. And to do that I had to complete a long list of loialty subquests spanning three games, take the time to go talk with all my companions through three games, and pick the right dialogue options through all of them, and it was probably the single most rewarding thing I ever did in a videogame; especially after I did it and then looked at a guide and saw how actually difficult it was to get that outcome. And that didn't have any meaningful effect on how the "reaper" main plot was resolved, but it doesn't matter.

So I'd still go with "it is not railroading if the players can make meaningful decisions that affect the plot long term".

[Lorsa]

interesting concept. However, more restrictive than it should be.

take the "fight" example. You are ambushed by assassins. You defeat the enemies through a cunning strategy.
Now, assuming that you are still dealing with the same big bad, or that you are dealing with a new big bad but they made their research, it is completely reasonable that you'll face a new group of assassins. And equally reasonable that they will be prepared for whatever strategy you used last time.
When my players curbstomped an ambush by a larger number of mid level character with four (quickened) firestorms, the next group (that happened a few weeks late) had protection from fire, and none of the players complained; they were all "right, it makes sense that they would do that".

One thing I wonder is how the enemies learnt of the characters' tactics? Were there any survivors or witnesses who reported the incident? Just because a group uses a tactic, it doesn't mean word of that tactic gets spread around.

Anyway, it matters what you put into the phrase "situation". It does not necessarily have to be any individual encounter. For example, the situation could be "a big bad is trying to get you killed". This could involve continually sending assassins until the PCs are gone. This situation could be solved by either killing the big bad, disguising themselves so the assassins can't find them, setting up fake rumors of their whereabouts or paying off the assassins (or any other number of ways).

Sometimes the encounter is the situation, but sometimes it is not. What should always happen though (in a non-railroad game), is that the players can affect the next scene. They can control how and where they get attacked by assassins, based on where they go or what they do.

Take the "mountain" example. Ok, you can fly past mountains without effort. At this point, it becomes completey pointless to present you with more mountains. I either give them some feature than you cannot fly over, or I don't give you an obstacle in the first place. As DM, my job is to challenge the party, and that requires adapting to the party strategies.
Now, of course if the DM withdraws the prize for getting over the mountains the first time because he wanted them to pass them in another way, that's railroading. but simply presenting a new obstacle that is "like your old obstacle, but without that weakness that let you exploit it the other time" is perfectly legitimate, especially if you don't do it straight away. Sure, a good DM will try to vary the obstacles, but in along campaign reciclying a few old ones is acceptable. And not all DM have a huge creativity to always come up with something new.
Furthermore, having always something new riisks making the world less cohesive. You keep introducing new stuff instead of delving deeper into the stuff you have, which is what I'd rather do.

Exactly, the mountains should not be the obstacle if they are, in fact, not an obstacle. And if they DM had planned for "something awesome to happen in the mountains" which they simply fly over, then too bad, that thing won't happen. Figure out something else instead.

Finally, for the "it only has one possible outcome", sometimes it's unavoidable. And if there is enough freedom to create meaningful consequences along the way, it may not be bad.
In you given example, the "real" situation is only the boss fight. but then, this applies to pretty much any villain. You know that when your party picks an arch-enemy, it will lead eventually (almost always) to a boss fight. it is not railroading, because you still influence everything else. Did the villain destroy that little village, or did you save it? did you leave in charge the former henchman, who is evil but makes the train run on time, or do you give a chance to someone more honest but less competent? did you sway to the side of good that miniboss who had somewhat nobler motivations than his peers?
I'm thinking of the Mass Effect 3 videogame. You have lots of choice, and in the end, no matter how you get there, no matter what you did before, you get three basic choices, and that was it, and I read some criticism for it becauuse "the choices you made before were pointless". but they were not. Did you help the quarians, or the geth? did you manage to persuade them to stop their war? did you cure the genophage, leaving wrex in command of the krogan? or maybe you cured the genophage, but wrex died and left his dumb brother in charge? or maybe you refused to cure the genophage because you were afraid of what they'd do? how many of your friends did you manage to save? when you had to choose, which one you left to die?
those are all the choices that matter, and it means nothing that in the end they do not affet the three options you are given. It does affect that the new galaxy will still have both quarians and geth, which it wouldn't have if I didn't persuade them to make peace. And to do that I had to complete a long list of loialty subquests spanning three games, take the time to go talk with all my companions through three games, and pick the right dialogue options through all of them, and it was probably the single most rewarding thing I ever did in a videogame; especially after I did it and then looked at a guide and saw how actually difficult it was to get that outcome. And that didn't have any meaningful effect on how the "reaper" main plot was resolved, but it doesn't matter.

So I'd still go with "it is not railroading if the players can make meaningful decisions that affect the plot long term".

Generally speaking, most story-based computer games are railroads. Even (or especially) Mass Effect. While it does give you some choices, they really don't matter in the grand scheme of things for the gameplay itself. The sequence of events in the game is very well set, and it doesn't really adapt its scenes to what you want to do.

If you try to model your game to ME, it will inevitably become a railroad. Very few things actually has "only one possible outcome". If you think that way, it's mostly because you can't see all the other possibilities.

[Quertus]

I am not sure if you agree or disagree with me or not. We both conclude that there is railroading (it seems), and we both conclude that it will appear different in play compared with where a situation is really just "fake" in order to lead the players to the next one.

Appropriately oddly worded, because "or not". That is, while I may technically be agreeing in the general sense of it being railroading, what I am actually doing is neither agreeing nor disagreeing, simply exploring the interesting example of the nature of railroading.

[Lorsa]

Appropriately oddly worded, because "or not". That is, while I may technically be agreeing in the general sense of it being railroading, what I am actually doing is neither agreeing nor disagreeing, simply exploring the interesting example of the nature of railroading.

Indeed it was oddly worded, and accidentally so. But perhaps accurate in this case.

After having thought it about it some more, I've come to realize that railroading is best understood in examples. All the ways to try and define seems to fail. I think my idea using kyoryo's flowchart of preferred game style comes close, but like all definitions it breaks down at the boundaries and may require too much abstract thinking.

So, the best we can really do to capture the nature of railroading would be to construct a large data set of examples, hoping people will understand the concept through that.

[King of Nowhere]

One thing I wonder is how the enemies learnt of the characters' tactics? Were there any survivors or witnesses who reported the incident? Just because a group uses a tactic, it doesn't mean word of that tactic gets spread around.

of course

Generally speaking, most story-based computer games are railroads. Even (or especially) Mass Effect. While it does give you some choices, they really don't matter in the grand scheme of things for the gameplay itself. The sequence of events in the game is very well set, and it doesn't really adapt its scenes to what you want to do.

If you try to model your game to ME, it will inevitably become a railroad. Very few things actually has "only one possible outcome". If you think that way, it's mostly because you can't see all the other possibilities.

I'm not using mass effect as an example of not railroading, merely an example of how something can have deep and meaningful consequences and yet not affect some other parts of the plot.

and while it's true that nothing is completely unavoidable, some things are really hard to avoid. If you have a big bad, you may be able to persuade him to change plans, you may make some ally to go fight the big bad in your place, but actually we all know that 95% of the times it means that there will be a boss fight somewhere in the future.

[Pippa the Pixie]

From the top of my head:

1- Reputation: A 10 level bandit would probably be well known.
2- Gear and equipment: There should be substantial difference in the equipment used by a level 1 bandit and a level 10 bandit.
3- Techniques: any martial character should be able to get an idea of how proficient someone is with a weapon, even if he is only fighting villagers. Hell, even if he is only doing cutting practice.

Yup, as I though, you use homebrew rules to make metagaming mini games. If that works for you, then great.

I just wonder if such metagaming counts as railroading?

Like if the DM says ''the random grup of bandits you encounter are well known to you and have a reputation of seven, and they have the number seven on their hide armors and they hold their swords in the seven style...so you know the bandits are seventh level" is that taking away agency and railroading?

[awa]

Yup, as I though, you use homebrew rules to make metagaming mini games. If that works for you, then great.

I just wonder if such metagaming counts as railroading?

Like if the DM says ''the random grup of bandits you encounter are well known to you and have a reputation of seven, and they have the number seven on their hide armors and they hold their swords in the seven style...so you know the bandits are seventh level" is that taking away agency and railroading?

that's quite the straw man there

how do you picture a fight happening that its inconceivable to you that you could tell that a foe is wildly out of your league?

A level 7 character wearing non mw armor and hide at that come on seriously.

edit
magic weapons would cut through flesh and bone more easily

if their rogues they should be striking vitals with every blow

if fighters they should have multiple attacks
and how could providing additional info possible be railroading

[zinycor]

Yup, as I though, you use homebrew rules to make metagaming mini games. If that works for you, then great.

I just wonder if such metagaming counts as railroading?

Like if the DM says ''the random grup of bandits you encounter are well known to you and have a reputation of seven, and they have the number seven on their hide armors and they hold their swords in the seven style...so you know the bandits are seventh level" is that taking away agency and railroading?

What? what are you talking about? I... I don't think we are talking of the same game xD... What sort of game are you talking about with mini-games ... And what even is metagaming according to you? I am incredibly confused by your post.

Did I imply that I assign a numerical value to things like reputation? Is that common at your table?

[Koo Rehtorb]

People should probably stop responding to Darth Ultron's alt account...

[Florian]

@Quertus:

Ok, this is why I have a bit of a problem with your answers to this particular topic. You are advocating a style (SimWorld) that has been dead and irrelevant to RPG mainstream for decades, provided it actually was ever part of the hobby beyond the starting years (which I doubt). The creation, maintenance and exploration of a fictional world is not the main activity most people associate with TTRPGs as a hobby per se. It is a by-product of the actual activity, which is exploring content that is there to amuse you in some way. As in, when your main activity is, say, exploring a mega-dungeon (Rappan Athuk), you still will want to answer some questions (How is elven culture different from human culture? Where do we sell all the loot?), which is why you need a setting. When people talk about "Sandbox style", it comes down to the players being able to freely select what content to engage and how, but the sole reason d´etre for this content is there to be explored and interacted with. Trying to go for verisimilitude is not really a requirement and more a form of courtesy towards people needing it for their enjoyment. (You might remember one of our previous dialogues about stage and backdrop?)

The sole definition of the act of "Railroading" is still the GM overriding player agency and choices to force an predetermined outcome."Illusionism" is the act of hiding that you did it (see Quantum Ogre).

In a certain sense, strictly linear styles, like that used in Adventure Paths, are also not "Railroads", because they will mostly allow for exploration and options, but only in a very limited "game world", which is centered about exploration of this one story. That can change when the GMs acts like/maintains the illusion that there is any relevant outside world beyond that story (actually seen that quite a lot: GMs keen on playing a certain module or campaign, but hiding that from their players. Personal opinion: When you want to gm Expedition to Castle Ravenloft, say so.). Here, too, the act of "Railroading" only happens when the GM overrides player agency and choices.

@the magic topic that came up:

I´m quite open towards my players about that one. We use game rules for convenience sake and as a means of communication/settling issues at the table. This can often lead to a certain disconnect between a game setting and the rules system used to play in that setting, but this is something that the players have to solve by themselves. To use Pathfinder and Golarion as an example, I generally explain that this world is post-apo twice over, Dying Earth style, with "magic as science" being slammed in the first apo, the follow-up civilization scavenging much, but with incomplete knowledge and an esoteric spin, going down in the second apo, the actual civilization being stuck in the equivalent of medieval times and scavenging bits and pieces, what we call "spells", of that lost knowledge. This being like collecting print-outs of single Wikipedia pages.....