Here’s how the Tyranny of Dragons campaign begins:
For the past several days, you have been traveling a road that winds lazily across the rolling grasslands of the Greenfields. Sundown is approaching when you top a rise and see the town of Greenest just a few short miles away. But instead of the pleasant, welcoming town you expected, you see columns of black smoke rising from burning buildings, running figures that are little more than dots at this distance, and a dark, winged shape wheeling low over the keep that rises above the center of the town. Greenest is being attacked by a dragon!
Although the boxed text doesn’t mention this, the dragon is accompanied by an entire army. Greenest isn’t just being attacked by a dragon; it is under siege.
So… what happens next?
The adventure assumes that the 1st level PCs will head straight into Greenest.
Which… well… It doesn’t really make sense, does it?
It’s not just that it’s difficult for me to imagine any player responding to that scene by saying, “Guess, we head into town then…” It’s that, for whatever reason, the adventure just assumes that this won’t be a problem. It skips directly from “there’s an army and a dragon attacking that town!” to “okay, now that you’re in the town.”
This is bad design.
And the problem here is not just that Tyranny of Dragons is a railroad. I mean, yes, that’s a problem. I’ve told you that prepping plots is a bad idea. And I’ve given a detailed breakdown of why railroads are terrible.
But let’s lay all that aside.
Let’s talk a little bit about what makes a good railroad. Or, more accurately, what makes a railroad work.
THE TRICK OF THE RAILROAD
A good railroad, at a certain level, is like a good magic trick: The players won’t really believe that magic is real, but a good magic trick will let them suspend disbelief just long enough to be amazed.
Players in a railroaded scenario will almost certainly know that they’re being fed a predetermined sequence of events over which they have no meaningful control. (Players are not, by and large, stupid. Even if they’re polite enough not to tell you that they can see you behind the curtain.) But a good railroad will disguise this; it will allow them to suspend their disbelief and get lost in the wonder.
There are a lot of different tricks the railroaded scenario will use to do this: It will use illusionism to make meaningless choices look meaningful. It will give the PCs a lot of control over stuff that is meaningless, but which is nevertheless given the appearance of meaning.
(Illusionism is a technique in which the players only have the illusion of choice. Regardless of which choice they make, they outcome will be the same. A simple example has the PCs come to a fork in the road: If they go left, they end up at the vampire’s castle. If they go right? They end up the vampire’s castle. Because they were always going to go to the vampire’s castle.)
But the most important technique for the railroaded scenario is to frame the meaningful choices in such a way that the players legitimately WANT to make the predetermined choice.
The GM never forces a card on them. In the end, they do the magic trick to themselves.
When a railroaded scenario pulls this off, the suspension of disbelief is perfect: Players never feel as if they were forced to do something. They’re able to remain completely immersed in their characters, feeling as if the world is unfolding in direct response to their actions.
And this is why the opening scene of Tyranny of Dragons is hot garbage.
BREAKING THE ILLUSION
When you show a group of 1st level characters that a town is being simultaneously ravaged by a dragon and besieged by an army, the choice you are signaling is, “Don’t go there. Stay away.”
This is true not only logically in the context of real people looking at that situation. It’s also true in the specific context of the entire D&D rule set, which is built around the clear expectation that 1st level characters DO NOT FIGHT dragons nor armies.
Every rational signifier is saying, “Don’t go there.”
And this would actually be doubly true in a well-designed railroaded scenario, because the well-designed railroad would be using those signifiers – BIG DRAGON, YOU CAN’T SURVIVE THIS – to force the logical choice, not its opposite.
The only context in which running into that town makes sense is, in fact, the one where a DM says, “This is a railroad. Get on the fucking train.”
And even if the DM doesn’t have to literally say that, the implication is absolutely crystal clear to everyone sitting at the table.
The illusion has been broken. The suspension of disbelief has collapsed.
It’s the very first moment of the campaign and the book has already set you up to fail. They have dug a hole and now you’re going to have to dig yourself out of it.
Now, the reality of a railroaded scenario is it’s quite likely you will be forced into this situation at some point. The complex web of choices and relationships and emotions of your PCs will eventually force you to push them back onto the tracks. That’s just the nature of a railroaded scenario. (And it’s one of the many reasons why it’s not a good way to prep or run an RPG.)
But this is, in fact, the very first scene of the campaign. There is no baggage. No complex continuity that needs to be impossibly accounted for. Avoiding this moment would be trivial: Just put the PCs in the town before the dragon and the army show up.
A good railroad designer will recognize when a mandatory choice doesn’t make sense and they will restructure the scenario TO REMOVE THAT CHOICE.
That’s not always possible in the convoluted middle of a scenario after the PCs have begun strewing chaos. But in the very first chunk of boxed text?
That’s bad design.
Even if you’re OK designing a railroad.
Which you shouldn’t.
A major problem is that so many players have been conditioned by scenarios like this that when presented with this scene they think not like their characters, but like players of a game who need to do certain things to move the action forward. They think (and sometimes even say aloud at the table), “well, the DM would not have presented us with this city if we weren’t supposed to go there. Let’s head in.”
As a DM, I’m discouraged to hear players say things like this, and I try to just address it directly by saying something like “There is nothing you are supposed to do. You are free to make whatever choice you think your character would make. You don’t have to go to the city if you don’t want to.” But in my experience, some players are uncomfortable with this, at least at first, because they are so accustomed to being led through an adventure by following the DMs prompts.
I adore your posts. Thanks for another insightful addition!
As to the subject of the post itself, my problem comes with *why* the designers started it this way. It HAD to be a conscious decision, so my question is what forced them into that position? Was it a Wizards of the Coast content briefing? Was it the lack of Monster Manual or Dungeon Master’s Guide at the time of writing?
I’m honestly curious. You wouldn’t think to expect this from freelancers Wizards is trusting to design to hardback books.
The Curse of Strahd, which is otherwise a reasonably good sandbox campaign, has a terrible piece of railroading at the outset to force 1st level characters into the Death House, involving mists: “As more buildings disappear into the mists, the characters are left with little choice but to seek refuge in the house.” Why the house wasn’t just moved so it’s on the way to the village of Barovia, rather than in it, is beyond me.
Thanks for a well considered and well written post, enjoyed reading it!
I 100% agree with the philosophy you are espousing here. I try to run ‘sandbox on rails’ kind of games where the players have agency to chase a number of goals but events move the story forward (sometimes these events are out with their control sometimes they are) on a story line.
I’m not a huge fan of most of the WOTC adventure offerings and this series in particular I find really pulpy. However, this was the first session I played of D&D after a 15+ year hiatus and it was easily one of the best session I have ever had (maybe absence makes the heart grow very fond). It’s also worth mentioning that we had a cracking DM which always helps!
I think the opening is risky and it could elicit a number of responses from players. I will speak on how it worked for us: 1. Seeing the dragon in a 1st level adventure made us feel that anything is possible and the world is dangerous, it made us question the understood meta game of challenge ratings. 2. It reinforced to us that the party isn’t the centre of the that dangerous world, this media res kinda worked in that way. 3. It gave us a strong motivation to go and help in any way we could (we were a good aligned party with strong motivations to help the down trodden).
If this scene unfolded in the way you suggested it could also be amazing, imagine the attack unfolding while the party is asleep in the Inn! However the way it stood means we were choosing how and when to interact with the event. We initially scouted around the outskirts of town and only intervened when we go a better look of what was unfolding. We then had a thrilling street to street battle, followed by a rescue operation and finally a duel with the half dragon. It was brilliant.
Sometimes the alchemy works despite the formula 🙂
I once heard a piece of DMing advice in response to PCs not biting the hook. “OK, these adventurers walk away from the dragon savaging the town, and a new group of adventurers show up who will save the town.”
I now feel 100% justified in making those new characters Riggs from Lethal Weapon (the first one), because a literal death wish is the only thing that makes sense in that scenario.
Justin,
I agree that this is bad design, but there is a certain set of assumptions that occurs when running an adventure path. First, that the PCs are all on board with the premise, and second, that they’re going to participate “in good faith” in the adventure. Both of these are important, and I think the expectation should be set out in advance before the game begins.
(a) We’re playing an adventure module, this adventure module will have you doing X.
(b) Choices within the adventure module are constrained by the nature of the module.
(c) Players are expected to “play nice” with the module, within reason.
A module is a bit like a scenic tour. Ah, take a look at the State Capitol. Next, we’re going to visit the statue of Abraham Lincoln! Snap some pictures with him. Now let us board the bus to the Vietnam Veterans’ Memorial. Lunch is scheduled for 1:00 PM.
I am not defending the module and I agree with your general sentiment that non-railroaded adventures are the more rewarding play style, at least in my opinion. But I’d like to point out a few things.
I am not certain that a majority of players would chose to not enter the city in the scenario you described. First and especially at the beginning of an adventure most players are willing to follow along with what the GM presents. This is even very common. “You have all been hired to guard the caravan” or “You are on a ship on the way to XYZ” should be very common examples of this. “You are heading into Greenest” is just a tad more extreme. And even if this were presented in the most blatant form this could still be interesting for the players: “Greenest is under attack by dragons. You are heading into town.” – “Wait, really, why?” – “You tell me. Why would your character do that?” – “Um, okay, let’s see, maybe …”.
This brings me to another point. Even if the players are on a railroad and even this is painfully obvious, it could still be fun. Even if the plot follows a predetermined chain of events that doesn’t necessarily mean that there are no “meaningful” choices. There is usually still plenty of creative space. How does _your_ character handle this situation? How do _they_ react to the NPC? How do PCs interact with each other? And so forth.
Now could Tyranny of Dragons be improved? Hell yes. Is it usually better to present players and/or characters with meaningful choices that shape the nature of the unfolding plot? I think so. But I also don’t think that anything else is necessarily “bad” or “wrong” or un-fun. In fact I know that there are players who would feel uncomfortable with this amount of freedom and who would be more happy to just follow along and enjoy the ride. As always it is up a GM to know their players and be aware that there are different kinds of preferred play styles and react to their group’s tastes.
“But what you’re missing is that the players will hold their nose and go along with the railroad anyway! Because they know that they have to!”
That’s not something that I missed. That’s what the entire post is about.
Many years ago a certain GM invited 5-6 players he knew, but seldom got to play with, to a convention game of Shadowrun. The scenario started out in the middle of the action, a too-hard fight where everyone died–it took an hour or so and it was gruelling. At this point he sprang the big reveal–this was a virtual-reality test from your would-be employer, to see if you were ready for the real scenario.
Deathly silence around the table, and then someone (might have been me, I don’t remember anymore) said “Uh-uh. Nope.”
“Nope?”
“I don’t think we do this.”
Around the table, “No, you’re right, we don’t.”
At that point we took the (rather shaken) GM out to dinner and did some brainstorming. When we got back to the convention, our characters got drunk and decided to go watch the poor fools who had to do this run instead of us. And then the poor fools got geeked, and the hostage ran into our arms, and we had to do something; and the scenario more or less picked up from there.
Wouldn’t have done it for a GM I liked less, though.
I think that when you railroad, you are making a contract with the players. The players agree not to go off course if they can help it, and you agree to make it as easy for them to stay on course as possible. Shadowrun first edition modules rather routinely violated the GM’s half of that (though I think this was a homebrew, it was very much in the same style). For one thing, they published quite a few modules in a row where your employer betrayed you; it got harder and harder to find reasons to take the next job when the last five had ended in betrayal.
The scenario you describe would not work for me personally. Yeah, I can go in there, but the player/PC link is probably broken, and also I play really badly if I am forced to do something that should get my PC killed–it is very hard not to start “playing to lose” to demonstrate why my view (“this is suicide”) is more correct than the GM’s.
A cardinal sin of bad railroaders is to expect the PCs to both go ahead with evidently too-hard scenarios when that’s where the train is going, *and* to exercise due caution with regard to threats not on the railroad. Unless you can read the GM’s mind you will get it wrong sooner or later, and the recriminations tend to be ugly. “Why did you go up against that? It was certain death.” “So were the dragon and the army. We thought we were supposed to.”
One more example:
I had a friend who sandboxed. His logic was that the world was full of all sorts of different threat levels and the challenges put to players would be based on their choices and would NOT be a continually escalating threat stream metered to align with character levels and capabilities.
So, we snuck in (as 1st level characters) to some dungeons under the City of Badabaskar. It turned out we were tip-toeing between various aged red dragons. Um…. there was a REALLY long shot that the GM had some deus ex machina that would help us drive off or kill all these critters, but that was much less plausible than us being waaaay over our heads here.
The GM didn’t have any wake up, although there were tense rolls. We got through that section into more ‘normal’ areas we were more capable to handle.
The trick: If the players are going to meet something beyond them, they need several clues, some decision time, and they will usually figure out what is the least lethal path. Don’t just drop something way overpowering without any warning or yes, the party will be a TPK.
But players knowing that these sorts of ‘over their head’ experiences meant that we didn’t take for granted everyone we met would be someone we could have a good chance again and we had to realize we were very mortal and that we needed to choose our fights carefully, recon our enemies, and not be afraid to hide, bribe, persuade, escape, etc. to avoid dying. If we were stupid and woke up one of the dragons, we’d probably have been it’s mid-nap snack. But that would have been a stupid choice so we’d have deserved that.
If you sandbox, let your PCs know that the world is full of different levels of threat including ones you can’t win a fight with at the moment. Listening to the GM for clues about the nature of situations and foes is key, playing carefully and with an eye to avoid surprises, and being smart enough to eject when an overwhelming threat is on the horizon are good parts of sandboxing (and hence they are common in OSR games). Your PC is mortal and probably wants to pay attention to whom he picks fights with.
This article has a misleading title. This is how a railroad does not work.
I was hoping for an example of what you talk about, a railroad that actually succeeds.