I get a lot of reactions to “Don’t Prep Plots” that basically go like this: “Oh, yeah. I don’t want to railroad my players, but…”
The justification or rationalization which follows the “but” varies. There’s a consistent recognition that roleplaying games aren’t really designed for linear, predetermined plots, but since the vast majority of the media we consume is purely linear (even most of the “interactive” stuff) those creative instincts get buried pretty deep in us.
One of the most common versions of the “but” I hear is, “But I don’t want them to kill my bad guy!” The justifications for this vary from a strictly predetermined “finale” that’s being ruined to the more seductive version of convincing yourself that your players won’t be “satisfied” if the bad guy is “prematurely” knocked off.
These instincts aren’t necessarily wrong: Pulp fiction is filled with scenes where the heroes impotently watch the bad guys escape, building the sense of rivalry between them and baiting our appetite for the finale of the story. It’s an effective trope.
But I don’t think railroading is the only way to achieve that trope at the gaming table. Nor do I think it’s the most effective: When you push your thumb onto the scales of fate in order to predetermine the outcome of your game, you deflate the value of that outcome. If you do it poorly (or simply do it often enough), the anger and frustration of the players will stop being focused on the NPC villain and start focusing on you. The value of the trope becomes depreciated when it’s achieved through artificial ends.
CREATING MEMORABLE VILLAINS
What I recommend instead is a three-pronged approach:
First, build tension between the PCs and the villain without using direct confrontations between them. Give the bad guy minions. Have the bad guy do horrible things to people, places, and organizations that the PCs care about off-screen. Social interactions in situations where the PCs won’t be able to simply shoot them in the head without serious consequences also work well to build a personal relationship. (As do taunting communiques and phone calls.)
Second, when you’re prepping your scenarios include lots of bad guys. You’re probably doing this any way, so the real key here is to simply refrain from pre-investing one of these guys as the “big villain”. Basically, don’t get attached to any of your antagonists: Assume that the first time they’re in a position where the PCs might kill them that the PCs will definitely kill them. (This attitude will help to break any railroading habits you may still be secretly harboring.)
Third, remember that people in the real world usually don’t fight to the death. Have your bad guys run away. And not just your “big villain” (since you won’t have one of those any way): Unless their back is truly to the wall, most of the people your PCs fight should try to escape once a fight turns against them. (If you’re finding it hard to break the “fight to your last hit point” habit, try experimenting with some morale rules.) Most of them will probably still end up with a bullet in the back of their heads, but some of them will manage to escape.
The ones that escape? Those are your memorable villains. Those are your major antagonists.
This is the crucial inversion: Instead of figuring out who your major bad guy is and then predetermining that they will escape to wreak their vengeance, what’s happening here is that the guy who escapes to wreak their vengeance becomes the major bad guy.
MEDIA FAUX-EXAMPLES
Consider Die Hard for a moment. As written, this film is a great example of our first principle: The antagonism between John McClane and Hans Gruber is established almost entirely without any direct interaction between the two of them: Gruber takes McClane’s wife hostage. They talk to each other through telecommunication devices. Gruber sends his thugs to fight McClane elsewhere in the building.
The exception to this is the scene where Gruber pretends to be one of the hostages. This is actually a really clever device that heightens the conflict between McClane and Gruber by allowing them to directly interact with each other. But if this was a game table, what would happen if the PCs saw through Gruber’s bluff and put a bullet through his forehead right then and there?
It doesn’t matter.
Remember our second principle? Lots of bad guys. So now Die Hard becomes the story of the hot-headed Karl Vreski taking control of Gruber’s delicate operation and blowing it up in a mad pursuit for vengeance. Maybe he starts killing hostages and becomes the most memorable villain of the campaign when he throws McClane’s wife off the top of Nakatomi Plaza.
Okay, so cycling through the org-chart of Villains, Inc. works when you’re facing a team of bad guys. But what if the PCs really are just facing off against a single nemesis?
First off, remember that not every challenge needs to be of epic proportions: Sometimes you run into some goblins in the woods and you kill them and you move on. You don’t need every goblin to murder the priestess’ cousin or become the sworn blood-enemy of the paladin.
Second, even the most memorable villains from fiction were often part of Villains, Inc. even when that isn’t immediately obvious. For example, consider Dracula: Wouldn’t it be really unsatisfying if Jonathan Harker sneaked into Dracula’s tomb at the beginning of the book and staked him through the heart before he ever went to England? I mean… this is the Dracula, right?
Remember, though, that Dracula is only the Dracula because that didn’t happen at the hypothetical gaming table. We didn’t know that he would become obsessed with Harker’s wife and kill Mina’s best friend in pursuit of her. We discovered that during play. So let’s pretend that play had gone a different way: Harker stakes Dracula and heads back to England, satisfied that he’s destroyed an ancient evil. It’s a beautiful, happy ending…
… until the Brides of Dracula pursue him to England seeking bloody vengeance.
As a final example, remember that you need to embrace the whole package: You have to allow your bad guys to die indiscriminately and you need to include lots of bad guys in your scenario. If you only prep the “big villains” and then allow them to die indiscriminately, what you end up with are the Star Wars prequels: Darth Maul is replaced by Count Dooku is replaced by General Grievous… and none of them ever achieve enough narrative weight to make you really care whether they live or die.
EXAMPLES FROM THE TABLE
As I’ve mentioned in the past, faux-examples from other forms of media can be useful due to the common understanding of the source material, but can be somewhat misleading because the official version of events from the original media lends a patina of canonicity that shouldn’t be true of actual tabletop scenarios. So let me also take this opportunity to offer a handful of examples from my Ptolus: In the Shadow of the Spire campaign.
SILION: Silion was a cult leader. Using our first principle, I built her up in a variety of ways: Her name was referenced in early foreshadowing. The PCs tangled with her thugs and were targeted for retaliation by her organization. She was also incorporated into the background of a new PC joining the campaign, becoming responsible for murdering the PCs’ family and destroying their village.
Eventually, the PCs managed to track down her lair. They snuck in, found her digging through a box of archaeological artifacts, rolled a critical hit, and put an arrow through the back of her skull. She literally never even got a chance to look them in the face.
My players gleefully tell this story at almost every opportunity. They love it. It’s one of their favorite moments from the entire campaign.
Why did it work? Because when you heavily invest a villain through foreshadowing, the payoff of defeating them is massively satisfying. It can be argued that this sort of thing might not work as well in other media (although consider that Luke’s actual confrontation with the Emperor in Return of the Jedi, after building up to it over the course of three films, lasts almost no time at all), but in a roleplaying game the audience is synonymous with the protagonist. Your players don’t want to be handed their quarry on a plate, but a quick kill shot isn’t a gimme: It’s a reward for all the work that got them to the point where they could take the shot.
ARVETH: Arveth was a mook. She was captured by the PCs, questioned by Tithenmamiwen, and then cut loose. When Elestra tried to sneak back and slit Arveth’s throat to stop her from warning the other cultists, Tithenmamiwen stopped her.
But then the cultists caught up with Arveth: Believing that she had betrayed them to the PCs, they tortured her and even cut out her eye. Eventually concluding that Arveth was still loyal to their cause, the cult gave her a team of assassins and sent her to kill Tithenmamiwen. This was our second principle: Use lots of bad guys and develop the ones who survive. (In some other campaign, Arveth could have easily been cut down randomly during combat and completely forgotten by the next session.)
Arveth nearly succeeded in her assassination attempt before the rest of the party showed up. While the rest of her team held the party at bay, Arveth managed to escape (barely evading Tithenmamiwen’s angry pursuit). This was our third principle: When they’re losing a fight, have your bad guys run away.
At this point, things transitioned to the first principle: Arveth used a magical artifact to send horrible nightmares to Tithenmamiwen (often featuring Arveth cutting out Tithenmamiwen’s eye). She issued threats to Tithenmamiwen’s friends. She placed a bounty on her head.
The PCs would fight her again. This time Arveth was teamed up with a medusa who turned two of the party members to stone. Arveth carved an eye out of each of the statues before making her escape once again.
By this point, of course, the PCs were absolutely furious. Tithenmamiwen, in particular, had a rage which burned so white hot that her alignment shifted: She had shown this bitch mercy and she was repaid with endless torment. I don’t think I’ve ever seen such intense hatred focused towards an NPC before. It reshaped the entire course of the campaign.
Arveth was a mook no more.
When she finally died, the cheers of the players rocked the house. They literally took her miniature as a trophy so that it could never be used in a game again.
WUNTAD: Finally, here’s an example of me breaking (or at least bending) my own rules. Following the scenario laid out in the excellent Night of Dissolution campaign by Monte Cook, Wuntad and a gang of other chaos cultists show up just as the PCs finish clearing out a dungeon. The intention of the scenario is really clear: The PCs have been beat up. Wuntad and his cultists should 
have a really easy time of beating them into unconsciousness and then stealing several key items that the PCs had taken from the dungeon.
Stealing the PCs’ stuff? That’s pretty much guaranteed to piss them off for the rest of eternity.
Deliberately designing an interaction to create a major villain is against my “rules”. But it worked.
What I consider the key thing here, however, is that I still wasn’t invested in a particular outcome: Common sense showed that the outcome was likely, but I still wasn’t predetermining it.
What if Wuntad had died? Well, I had reinforced the scenario by following my other design principles: He was supported by lots of bad guys (including Silion from the example above). If he had died, somebody else would have stepped in and taken control of the cults. (Which is not to say that these characters are interchangeable: Killing Wuntad would have made the PCs a major target in the campaign a lot sooner. Factions within the cults probably would have broken away from the new leadership. And so forth.)
Similarly, returning to one of our faux-examples for a moment, common sense tells you that the leader of the terrorists who have taken over the Nakatomi Plaza is more likely to become John McClane’s nemesis than one of his mooks. You don’t have to abandon that common sense in order to follow the principles of RPG villainy.
FURTHER READING
The Railroading Manifesto
Node-Based Scenario Design
Gamemastery 101
There’s been an odd sort-of-inversion of this in my current D&D5 game. The campaign has lots of minor villains running about, most of whom have some sort of connection to another villain called the Black Spider.
They haven’t met the Black Spider yet but keep hearing about him and this is where it starts to go a bit odd, because they think that every minor villain they meet is the Black Spider despite having no evidence for such a conclusion. It doesn’t seem to matter that they’ve got it wrong two or three times already; the next villain has to be the Spider.
The other odd bit — and the one that’s more relevant to your post — is that without exception each of these minor villains has got away to torment the players again, but I’ve not been pushing for it in the slightest; instead they keep leaving vast gaps in their defences that allow the bad guys to escape. One time they found a wizard’s laboratory, where he’d been creating invisibility potions, but couldn’t find the wizard. Of course, he was in the room with them, staying quiet, but they didn’t put two and tow together and so he escaped.
What’s happened is that as we near the climax of the campaign they’re faced with a gang of villains who have returned not because of clever plans and audacious tricks, but because of recurring cases of mistaken identity and baffling decisions by the players.
I’m going to play the villains as having quite an ego boost as a result of surviving their encounters with these bufoons. Let’s see how it goes.
I never really thought of Darth Maul, Count Dooku and General Grievous as major villains. I thought they were just henchmen. As an audience member, I knew who the real villain was.
In an RPG, the players might have had more of a chance to figure out who the real villain was. What if they did catch and kill the Emperor at the end of episode 3? I suppose there could be a power behind him, but that also could seem contrived.
So, there are times when it is fine to put your thumb on the scale. Sometimes your major villain could have more up his sleeve than you might have thought of beforehand. Maybe he had a secret exit from his storage room. Maybe more guards showed up in time to let him escape. Maybe he was just a clone.
The distinction between “audacious tricks” and “baffling decisions” is often a slim one when it comes to the PCs.
Oh. Wait. You meant audacious tricks by the NPCs. 😉
I found this post very useful. For a recent video game example of this, See Shadow of Mordor. Its USP, the ‘Nemesis’ system works exactly in this way – even if an ordinary orc gets lucky and kills you he could become a captain and start working his way up the uruk hierarchy. Captains remember you and are shaped by their battles with you (if they survive). It’s a lot of fun.
I take issue with your Dracula example. In Bram Stoker’s Dracula, if Dracula dies, the entire literary background to the work is invalidated. Dracula’s brides can’t be the major villain because them being so changes the meaning of the work fundamentally. Even a cursory look at the literary analysis published on the work indicates that there are several fundamental and insurmountable problems with that plan. The issue here is that Dracula isn’t just a random mook, Dracula is a symbol. His symbolism relies upon aspects of his character (masculinity, reserve, ethnic stereotyping) that the other vampires *just don’t have*. This is clear almost from the beginning of the novel. Jonathon can’t be allowed to stake Dracula without also metaphorically staking the things Dracula represents, which, regardless of when/where it happens *is the end of the campaign*. Perhaps that means Dracula isn’t something that can be done in the RPG medium, or is something which should be avoided, or is something to which your advice does not apply. Regardless you can’t just pull out some random other vampire and be like “Ok, you conquered your sexuality and are now at home in England, but here’s another symbolic representation of repressed sexuality and desirable evil, only it’s less good than the other one you already killed.” If you kill Dracula, start a new campaign.
Okay, so Dracula is a symbol of masculinity, reserve, and ethnic stereotyping. What, symbolically, does it mean when Jonathan Harker stakes him and releases his wives to terrorize England?
Is that a different message than the one you want to railroad your players into? Probably. Does that mean you have no choice except to railroad you players? Nope.
I was primarily discussing things in terms of plot. You’re discussing it in terms of symbolism and theme, but the conclusion is the same: The primary strength of an RPG is in exploring the choices made by players. It’s a fairly terrible medium for delivering predetermined outcomes.
“Dracula’s brides can’t be the major villain because them being so changes the meaning of the work fundamentally.”
So?
I think that’s the point. Don’t be so stuck on it meaning X and allow it to mean Y if circumstances (read: the players) make that happen.
Dracula’s brides can be the major villain, and it’s a different story. The idea that if you kill Dracula you need to start a new campaign only exists because you know how the Dracula story ends as it was written (because the plot was prepped). It could have been written different, and been a completely different story…and that is fine.
[…] Alexandrian posted a great article a few weeks back with a great piece of GM advice: Create lots of villains – develop the ones […]
“What if they did catch and kill the Emperor at the end of episode 3? I suppose there could be a power behind him, but that also could seem contrived.”
If they kill Senator Palpatine before he makes himself emperor, now they have to deal with all the other Senators who were going along with him, which allowed him to nearly become Emperor. Half those Senator think theu should be running things now, the other half are out to create personal fiefdoms, and the PCs find themselves running all over the galaxy putting out fires while the Republic comes apart at the seams.
“When you push your thumb onto the scales of fate in order to predetermine the outcome of your game, you deflate the value of that outcome.”
In 5E D&D it has a section on awarding experience that includes just leveling up players so that they are ready to meet the next part of the “story”. This seems to be becoming a common practice but I have asked my DM not to do it. It seems to be a railroad technique and it takes away any sense that I have earned the new level by my actions.
Love your blog, and this post especially. I think sometimes the main obstacle is us GMs thinking we’re Tolkien or Lucas or basically great writers and become too infatuated with our own plots, when in reality it’s never nearly as good as we think it is. And even if it is, our players aren’t a TV Audience/Book reader you need to move, they are the goddamn Main Characters and they need to enjoy themselves, and their freedom of choice is basically the reason it’s called a GAME, not a literary or theater exercise.
If the Big Bad is killed, there are consequences to that too, there’s no plan that can be carried out by one villain alone, there’s always lieutenants and henchmen, and maybe the death of the Big Bad is what kept them civilized, maybe by killing him/her early they unleash complete chaos or something worse, like when a dictator gets killed and the country goes into full civil war between the factions that were once somehow being controlled by that evil leader. The GM is a player too, and need to roleplay and react accordingly with his characters just like the PCs do with their own, that essential to Gamemastery and once you accept that there isn’t a fixed plot, it’s amazing what you can come up once you remove your shackles of railroading habits.
The Dracula example isn’t a bad example. Yes, if Dracula gets killed early it’s bad for the story as it was intended, but a RPG is a GAME, not a novel or play or movie, just because something is a bad movie plot doesn’t mean it’s a bad RPG one. A movie/novel/play/etc. plot is written for a passive audience, a RPG story has to be, first and foremost, fun to PLAY TO, even if the story is bad for a passive audience to watch or read, as long as the people involved enjoy it, that’s the whole purpose.
[…] guys go out of their way to help beneficiaries. This is mentioned indirectly in a great post about RPG Villainy at The […]
I came across your web recently and I’ve been devouring the content ever since. The reason I comment in this particular entry is because I’ve seen this particular topic discuss in several other websites, blogs, youtube channels and articles, and I think yours is the best one I found on describing this concept. Great examples, great explanation. Thanks a lot.
It just so happened to me that the Nemesis system (from Shadow of Mordor and the upcoming Shadow of War videogames) is a great example of your principle. The mook that happen to land the killing blow on you gets promoted, so becomes a “big bad”. You mowed down hundreds of mooks just like him before, but this one becomes special because you did not. And you may have killed tens of different such big bad (captains in the game), but the one that killed you became an even bigger bad (a warlord, if I remember correctly). And the ones you did maim without killing? They can also come back as captains, with all the scars and memories you’d expect.
Here, this article sums it very well : http://www.gamesradar.com/shadow-mordor-nemesis-system-amazing-how-works/
(Of course, I read the previous comments after posting, and DanJW already said it. Oh well, sorry for repeating it!)
Hey, man. Great minds think alike. 😉
[…] complicates the characters’ lives. Consider promoting that enemy to the big bad. Justin Alexander advises, “The real key here is to simply refrain from pre-investing one of these guys as the ‘big […]
Wow. Just found this amazing site via a Reddit post. How is it that I did’t know it long ago??
[…] By letting go of the idea of one central keystone villain, the narrative breadth and responsiveness of your scenario to the player’s actions increases dramatically. I’ll point at Justin Alexander’s nice Article “The Principles of RPG Villainy”[…]
[…] Alexandrian Illustrates this dilemma in his “Don’t Prep Plots” series. He uses the example of Dracula by Bram Stoker. What if Dracula dies early? The story, as written, […]
Thanks for the article and this site. I was directed to recently it via the Gumshoe Facebook Group and I am really getting value out of the content.
Do these rules get modified for one-shots? I have played a number of convention one-shots were the BBEG is locked behind a door or hidden until he is discovered. This is typical of the investigative games I have been diving into. In the horror genre specifically which typically ends in a climactic confrontation with the evil that is the subject of the story, giving that power to others might be a challenge.
In investigative game-planning where clues are carefully laid out (still following the situation design) switching the BBEG could be problematic.
Any advice on one-shots or investigative/horror style games were one of the main designs is the BBE(?G) that would still follow the rules laid out here?
I just watched this video on YouTube. Great stuff, as always. Please come back to the tube soon. 😁
There are many great tropes a cunning Dm could hang the Weird Sisters on. Feminism opposing masculinity, the three magical females (Fates, Hecate Sisters, etc…) banding together, so on and so forth. They might have to prove themselves to Dracula’s minions in order to take them over but certainly not insurmountable.