How do I resist the urge to railroad my players when I GM?

Stop preparing plots. Run a dangerous and interesting world instead.
Just come up with a fantasy world full of cool shit, and constantly come up with ideas for opportunities, threats, and calamities - near or far - that the PCs can learn about in one way or the other, and then use those whenever you can.
As your campaign develops, you can tie these to known actors/factions/nations and their established or secret motives, etc.

If you GM this way, stories will happen on their own - and they will belong not only to you but also to your players.

The moment you plan for any particular kind of outcome, you've implicitly decided to railroad them and take away part of their agency.
So stop preparing plots like you're writing a book or something. Run a world full of shit to do.

Make the party the crew of a train.

Planning things out and forcing things to happen just the way you planned it. Especially egregious if the players come up with stuff for their characters to do that's completely reasonable, but you start to do serious mental gymnastics to make even those things impossible via asspulls and pedantic "No-Man"isms.

Railroading is leaving no room for improvisation or player choices because you've already decided what happens next in terms of session-to-session storyline progression. Resorting to GM fiat, deus ex machina NPCs and other things in which the GM dictates everything that happens and players are ignored, bounced back into track or outright punished for not going the way the GM presented it. Or having a single solution to a problem that the players must find as if it were a videogame with states and variables rather than a collaborative narrative effort.

Contrary to popular belief, having a plot, main goal or questline isn't railroading. Many people don't like sandbox games, and a structured adventure tends to be the norm rather than the exception. However, there needs to be flexibility within this plot in order to take in account player actions and leave room for improvisation.

Seasoned GMs know of these differences and, unless the players WANT a fully sandbox game, know how to place premade elements and hide the tracks to leave room for player choice, reusing concepts and descriptions that don't necessarily need to be in one or another specific place, etc.

Having a main goal for which to work towards without dictating how and why is the key for a non-railroaded structured adventure. Build a possible destination and see if the players take an interest to it, build a world to work as the stage of the adventure, and the rest will start filling up through player actions and your own improvisation.

Hell, my main quest is directly related to a character's backstory, but he needs not be the centerpiece of the adventure because it relies on another NPC related to him that sparked the whole idea.

>tl;dr:

I appreciate this advice, because I'm fairly new to role-playing in general but I run Delta Green for our group and I usually run prewritten scenarios. I drop clues on places I'd like them to go or places they could go, but sometimes they'll surprise me with the order or what they want to do and I just try and roll with it. For example, currently doing Lover in the Ice.
>they're "supposed" to check on the status of a storage facility and the guy who was watching it
>have a deputy mention that he could use their help with a John Doe he found, at some point
>they decide to immediately check out the John Doe
>which ends up ultimately leading them to a place where two chicks, the dead guy's roommates, that had been infected by some monster and left two corpses that were "pregnant" with more monsters
And from there I was kind of playing stuff by ear. Everyone seems to enjoy themselves, but I do find that I occasionally drop too heavy hints.

Instead of strapping the players into a car going down the tracks, lay the tracks in the party's path for them to follow.
If you're really bad at improvising, have the baddy's goons chase them down the tracks.

So on a scale from 1 to 10 how railroady is forcing the players to kill a baron and the guards arresting them in a different country when the players had all managed to get race changed?
The dm for our last game did that

like a 12

Not everything has to be figured out. It's fine if the next interesting thing to do drops into their lap every now and then, as long as it's not literally everything, or you forget about using skills (or the equivalent) because you just give everything away like candy.
You're probably not doing that; I think you're probably just severely overthinking that particular aspect of the GMing job.

After all, most player characters have at least basic human senses, and sometimes shit is just visible or audible.
Not everything has to require Sherlock Holmes-ing to be engaging. In fact, there's too much of that then your game's pacing slows down a lot.

Railroading is about perception. Give them the illusion that they're choosing between three or four different things, which will, with minor adjustment, put them where you wanted them anyway.

The basic plot outline is the same, just the trappings are different.

It isn't like a video game where they can go back and replay it and realize that all the choices are essentially the same.

If your game is engaging and interesting then there's no reason to even attempt to change. You need to get better at delivering your game, then it won't matter if the players are on the rails they'll be too busy having fun to even notice.

Don't think of the solutions to any of the problems created in the world. If you have no idea how to fix the problem they will try something, then let the dice fall where they may.

Alright, the game kinda fell apart after that and I'm taking a crack at dming, what's a 7? I know the extremes because I've been through them but I don't know the middle ground too well

This. Planning ahead isn't bad- refusing to accept any changes to your plan is.

A 7 would be the GM banking on the idea of the baron being killed, and trying their best to manipulate the players in that direction, while also having no real ability or desire to deviate from their own expectations. It's okay to have an ideal outcome, but it becomes a problem when you can't adjust it.

Middle ground is mostly occupied by deciding to say "no" to a lot of "probably"s and "maybe"s.
Often with some sort of semi-acceptable (but not strictly necessary) justification.
Also, fudging dice rolls.
Also, refusing to improvise and just "blank page-ing" when they could probably have tried. (legitimately coming up blank is different though. That's an honest thing that can happen to any DM when the prep runs out or fatigue sets in.)

The main point being that the "no"s, fudging dice, and blanking tend to noticeably intensify when the DMs planned outcome is in jeopardy of not turning out how he wants.

- oh i almost forgot. Making select questgivers or "important" NPCs incredible, undeniable badasses, even when it doesn't feel necessary or reasonable.

Planning for one particular outcome is the path to the dark side, honestly. If you prep a "storyline", you've inadvertently created a huge sunk cost for yourself in that particular version of events.
I mean, sure, you can prep for a couple of the most obvious possible outcomes, but if you go "when the players defeat the evil wizard, then X happens, then Y happens, then they do Z", you're gonna have a real tough time allowing anything else to happen.
All that time and effort will have gone to waste, after all.

>Making select questgivers or "important" NPCs incredible, undeniable badasses, even when it doesn't feel necessary or reasonable.
That always seemed stupid to me. I'll usually throw in a friendly into DG games that is a point of contact and possibly extra help, but they're in the background and really only serve to allow someone to keep playing if their character dies(or messes with a puzzlebox and completely disappears from the known universe).

except maybe the players.

If you feel an urge to railroad your players, that's usually a sign that your adventure is designed wrong. Don't build them with an assumption that the players will go through any specific sequence of events: simply give them a location and a target that's hidden in the furthermost point of the location, and let them figure it out.

I've run two very successful games from 1st level to epic shenanigans, and in both of those i just plopped the PCs down into a world that was ending for some reason or another, and was like "nobody knows what the fuck to do about this, go nuts".

I think giving the PCs a really obvious overarching goal can be quite fine as long as you let them reach that goal however the fuck they want.

Go into a few sessions without any prep. Your players will be able to drive the game more, as it should be, and your role will return to neutral arbiter of the rules and gameworld rather than 'storyteller.'

I make it very clear before we start a campaign there IS a plot I have in mind and the game is not a sandbox to just fuck around in randomly. There isn't anything wrong with those kinds of games, in fact I enjoy playing in them sometimes. However if you don't want to play in a that kind of story driven game its no skin off my nose, we can play a sandbox another time.

And even then with my plot driven games, instead of railroading I think of my games as an eight lane highway. It goes to the same place as the train but there's room to maneuver. I don't meticulously plan every interaction and encounter to contribute specifically to my plot, rather I have several fairly general plot points to meander toward. I allow reasonable side tangents and the PCs doing unexpected things but use them to my advantage to present them with guidance along the plot.

Part of this is actually on the players. If your players are unimaginative murdohoboes who treat Veeky Forums like they're playing vidya, you're not going to get anything out of them unless you highlight interesting things with a big fuckoff "GO HERE" arrow.

With good players? Just make an interesting setting and situation and they'll do the rest.

PbtA has two pieces of advice:

>Play to see what happens
>Be the character's greatest fan

You don't play to have the players go through your plot - you should be as naive to the plot as the players, maybe only slightly more aware and that's it. When players decide to go to a certain plot point, follow them. Let players railroad you, not the other way around!