Patreon DevBlog | These Boots Were Made For Walkin'

Welcome back to the DevBlog! This week, I kept working on Curse of Strahd: Reloaded, Dungeon Mentor, and this month’s DM’s Toolkit/homebrew content pack updates. Let’s talk shop.

Problem 1: How do I fix Strahd’s introductory encounter (again)?

You might remember that last week, I published a DevBlog discussing my efforts to keep Strahd’s introductory encounter in Re-Reloaded from falling flat. In it, I went through my realization that the encounter lacked a strong dramatic question, and my process for giving it a proper one to install some proper tension.

Except there’s a problem—I went too far in the other direction. I added so much tension that the scene became utterly suffocating. As my friend (and constantly appreciated critic) Twi accurately put it, “Motivation-wise, it feels like you've got a Quantum Ogre here. There's no right choice.”

So here’s the dilemma I was facing:

1. I needed a challenge to provide meaningful tension. (By “challenge,” I mean a challenge that the players could interact with through one of D&D’s four pillars of gameplay: combat encounters, social encounters, obstacle courses, or puzzle encounters.)
2. I needed that challenge to neither be opt-in nor opt-out, such that the players couldn’t avoid it merely by ignoring Strahd or playing nice with him. (I called this the “Geralt” problem—Geralt, of the Witcher series, would neither look a vampire in the eyes nor allow himself to be provoked into endangering himself. Too many players play like Geralt for this challenge to be optional.)
3. I needed that challenge to allow the players to make meaningful decisions, such that there was a genuine possible “win condition” and a genuine possible “lose condition.”
4. I needed that challenge to be structured in a way such that even if the players won, Strahd also won. (After all, this is Strahd’s introductory encounter as the villain—I can’t allow him to lose face here!)
5. I needed the challenge to establish and strengthen the campaign’s core dramatic question (“Kill Strahd before he kills you”)—so that the players would feel strongly incentivized to meet with Madam Eva to have their fortunes told—but without making the players feel immediately threatened. (Why? In my experience, the “dinner with Strahd” event later in the campaign is basically impossible to pull off if the PCs feel actively unsafe in Strahd’s presence.)
6. I needed the challenge to establish Strahd’s major character traits—cunning, cruel, and arrogant—without making him out to be brutish, blunt, or crude.

My first attempt—in which Strahd merely tries to charm the players into revealing information to him—failed requirements #1 and #2: there was no actual “gameplay” for the players to interact with, and the players could easily avoid it by Not Meeting Strahd’s Eyes.

My second attempt—in which Strahd merely has a charmed Gertruda with him—failed requirements #2 and #3: the players could easily just Not Care about Gertruda, and even if they did care about her, there was no meaningful way for them to “win” anything.

My third attempt—in which Strahd pushes the players to admit they want to leave Barovia and then demands that they fork over Ireena in exchange—failed requirement #3 hard: no matter what the players did, they were going to inevitably have to fight their way through (and be defeated by) a bunch of dire wolves.

To see where I went from here, let’s go back to Requirement #1. I wanted meaningful tension, which meant I had to choose a pillar of gameplay.

• Puzzles were out, since “Strahd as a riddle-giver” just feels bizarre.
• Obstacle courses were also out, since I wanted to limit this encounter solely to the River Ivlis Crossroads.

That left two options: a combat encounter or a social encounter. Due to Requirement #4, I felt like I couldn’t go with a combat encounter; no matter which way I set it up, it felt like a win for the PCs represented a loss—or at least a loss of face—for Strahd. (It’s kind of awkward for Strahd to say, “okay, now fight my dire wolves,” and if/when the PCs win, say, “wow, I meant for that to happen the whole time” and then immediately slink away.)

That meant my only option was a social encounter. To get my mental gears turning, I went back to the dramatic questions reference sheet in the DM’s Toolkit and started browsing the section that listed possible encounter verbs. The verbs that immediately leapt out at me for a social encounter were: “antagonize,” “befriend,” “dissuade,” and “persuade.”

“Anagonize” and “befriend” didn’t quite make sense—obviously, the players wouldn’t want to antagonize Strahd, and there’s no way that they’d be able to befriend him. “Dissuade,” however, seemed like a much stronger starting point—largely due to a single, looming question that dominates the original module.

Why doesn’t Strahd just take Ireena once she leaves Barovia with the PCs? She’s out in the open, unprotected. If he wanted her, he could just, y’know, take her.

…unless he gives the PCs a chance to dissuade him from doing so.

Remember: this version of Strahd doesn’t want to torment the PCs—or, at least, doesn’t want to yet. He sees them as possible successors and/or consorts. And that means that he wants to test them.

And what qualities is he looking for in a potential successor or consort? Intelligence, cunning, and arrogance—all excellent attributes that a PC can show off in a “debate”-oriented social encounter.

This wasn’t a torture session, I realized. This encounter was a job interview—which meant that it was time for Strahd to ask questions.

From there, the rest of the encounter fell into place. Where does the tension come from? Well, Strahd is suggesting horrible, terrible things—killing Ismark, kidnapping Ireena, killing the PCs—as if they were as innocuous as walking down the road to the supermarket. This clearly establishes requirement #5; while the players don’t feel immediately in danger, they are now clearly On Notice that Strahd won’t hesitate to kill them, and is in fact happy to find excuses to do so.

On top of that, Strahd “the Devil” von Zarovich is also a perfect candidate to play Devil’s advocate, pushing back on the player’s positions and forcing them to strengthen their arguments. This does a great job of meeting requirements #3 and #6.

Throughout it all, though Strahd will always inevitably concede and accept the players’ well-reasoned arguments, it’ll be clear that he’s merely humoring them because he finds the entire exercise amusing. Even if the PCs win, Strahd wins—they keep their lives, while he gets his entertainment. Requirement #4 satisfied.

Of course, this is all still theoretical—I haven’t had this encounter playtested yet, and I’m still workshopping it around. Still, I’m a lot more confident in this version than I’ve been with any of its predecessors, so hopefully that’s a good sign!

Problem 2: How do I organize the DM’s Toolkit resources for prepping?

My big project this month in updating the DM’s Toolkit was originally going to be focused on revising the session-prep materials (reference sheet and template), since some patrons have noted that they feel over-complicated or otherwise intimidating.

Now, I’ll concede upfront that I’ve felt a bit weird about these for a while—they have felt more complex than my usual approach to preparing sessions. For a while, I just kind of mentally wrote that off as “oh, I must be doing a lot of these steps in my head,” but then I looked back over the reference sheet and realized, no, there’s actually a lot of stuff in here that I literally never do.

The problem? I’d conflated “session prep” with “adventure planning.” This is a hallmark technique of the Sly Flourish “Lazy DM” method (which tends to fly through campaigns by the seat of its pants), but that’s not my method—instead, I prefer to outline the broad strokes of my adventures first, and then fill in the more granular beats session-by-session.

So here’s what I ultimately decided to do: session-prepping is about actually organizing the adventure notes you’ve already got and then filling in some of the details, like “how will this combat be balanced?” and “how should I make this NPC more memorable?” Meanwhile, adventure-planning is about outlining the adventure’s arc and chapters.

This has the advantage of keeping “session-prepping” relatively simple, and actually quite straightforward most weeks. The main disadvantage here is that this approach offloads a bit more work onto the “chapter-planning weeks”—weeks when your players have just finished an adventure chapter and you need to start scaffolding the next one.

Still, in my experience, that’s an unavoidable consequence of any method that doesn’t rely on week-to-week improvisation; sometimes, you just need to buckle down and get crackin'! Hopefully, though, the revised materials I’ll be coming out with will make the whole process easier, faster, and as pain-free as possible.

Problem 3: How do I implement the phantom hitchhiker in D&D?

This month’s homebrew content pack is Boo! A Ghost-Themed Content Pack. Given the subject matter, I felt like I just had to incorporate some traditional ghost stories into the content pack—and I wound up settling on the phantom hitchhiker as a random encounter.

But I ran into a bit of a problem! The phantom hitchhiker usually relies on the existence of, well, hitchhikers—and, by extension, cars. In a fantasy D&D world, that’s not going to be a thing!

In order to get over this creative hump, I turned to my new DMing best friend—ChatGPT. Here’s how I prompted it:

“Create a D&D random encounter in which the players meet an NPC on the road who is secretly a ghost seeking to lead travelers to their desecrated burial site.

The narrative of the encounter should strongly resemble the "phantom hitchhiker" urban legend, in which people meet with or are accompanied by a hitchhiker who subsequently vanishes without explanation, having borrowed a garment that the players later find draped over a nearby gravestone.

The encounter should have a strong dramatic question resembling the format: "When [an event happens], can the players [accomplish a goal] before [a bad thing happens]?" (This dramatic question should not obviously relate to the ghostly traveler's ghostly nature.) However, this is a false dramatic question; the true purpose of this encounter is to lead the players to the ghostly traveler's desecrated grave.”

Here’s how it replied:

“The players are traveling on a remote road when they come across a lone figure on the side of the road. The figure, a young woman, is dressed in tattered clothes and seems to be in distress. She explains that she was on her way to visit her grandfather's grave and has lost her way. She asks if the players would be willing to escort her to the graveyard and they agree.

As they travel together, the woman tells them stories of her grandfather's life and how he was a respected member of the community. The players may notice that the woman is acting strange and that the weather seems to change whenever she is around.

As they near the graveyard, the woman disappears, and the players find her scarf draped over a nearby gravestone. The gravestone bears the name of the woman and the date of her death, revealing that she was a ghost all along. The players discover that her grandfather's grave has been desecrated, and the ghost was seeking their help to restore it.

Dramatic question: "When the players come across a lost traveler, can they find her grandfather's grave before they discover the truth about her ghostly nature?"

The true purpose of the encounter is to lead the players to the ghostly traveler's desecrated grave and help her restore it.”

Perfect. With a few tweaks, I wound up with an encounter that I was more than happy to add to the content pack.

Now, I’ll say this—one of several reasons why I’m relieved to start doing these DevBlogs instead of the content packs is because of my increasing reliance on ChatGPT for ideas; it just doesn’t feel right to publish things under my own name when 75% of the creative concepts come from (an, admittedly, well-prompted) ChatGPT.

In general, though, I think that this is an important opportunity to reflect: With new AI tools offering revolutionary creative output, I think the time is ripe for to refocus our DMing efforts on the harder, but important work of conscious game design and narrative design. In a world where ideas truly are cheap, execution—now, more than ever—is how we can stand out to make our games truly memorable for our players.

Problem 4: What’s the best way to teach new DMs how to avoid escalating situations?

I was browsing Reddit last week when I found this post about a group of players who had escalated an in-game situation with the DM in a classic murderhobo way, winding up with civilians dead, buildings on fire, and the plot generally falling apart.

Now, the DM in that post seemed pretty happy to adapt and continue on, which is of course their prerogative. But this kind of escalation—in which the players try to do something cruel or stupid, fail, get “punished” by the DM, and fight back—is an incredibly common issue, especially for new DMs, and I’m not sure that “punishment” (as the post put it) is necessarily the right way to go.

Previously, I’d planned to address the question of resolving player problems and running Session Zero in a Dungeon Mentor course titled “Managing Your Table.” As I was reading this post, though, I realized that I didn’t really have any lessons that would help a DM to proactively recognize this loop of escalation and shut it down before things went off the rails in the first place.

After contemplating the question and chatting with Brytenwalda in the Patreon Discord, here’s how I settled on it: this was a question of mismatched expectations. The players had expected that they’d be able to get away with robbing random shopkeepers because, well, they’re adventurers! Batman would get away with it. The Punisher would get away with it. Why can’t they?

But the players fail to realize that they’re not Batman or the Punisher. In a system like FATE, a player playing a Batman-esque character would probably Invoke an Aspect and have a really good chance of intimidating the shopkeeper into giving the item up. A 3rd-level PC in D&D just doesn’t have that; even in the best-case scenario (+5 to Intimidation), a DC of 15 means you’re looking at a 50% chance of complete and total failure—which is something that the players probably didn’t expect!

Here, there was a further complication: the random shopkeeper somehow had a CR 5 NPC backing him up! This is absolutely crazy! Batman and the Punisher don’t run into NPCs that strong except in really climactic scenes. The players, again, didn’t expect this to happen.

How could the DM have prevented this from happening in the first place? They could have interrupted the players and said: “Just to make sure we’re all on the same page—it’s entirely possible for NPCs to refuse to give into bullying in this campaign setting, and it’s also possible for random NPCs to have really strong hired muscle that you can’t easily defeat. Also, this campaign assumes that you’re playing heroes, not bullies. Do you want to reconsider what you’re doing right now, or should we discuss further?”

This is all about establishing expectations—either because the players forgot them, or because you never set those expectations in the first place earlier on. Importantly, this isn’t “railroading!” To quote the lesson I wound up writing about this:

When a playgroup sits down with conflicting expectations, it’s far more likely that a Dungeon Master’s response to a player’s “misbehavior” will make that player feel “railroaded.” Instead, by communicating and agreeing to a set of clear expectations, the players agree to constrain themselves in a way that promotes the fun of the whole table, and allows players far more freedom to act in ways that might otherwise be misinterpreted.

TL;DR: Expectations matter. Set them!

Designer Commentary | Puss in Boots: The Last Wish

I have no idea if this is going to be a one-off thing or a more frequent installment, but I want to take a moment to make some observations about the lessons that DMs can take from the new Dreamworks movie: Puss in Boots: The Last Wish (which I just saw on Saturday night).

(Warning: spoilers below.)

Specifically, this movie is a fantastic example of the reasons why resonance matters for NPCs—and especially villains. By “resonance,” I mean the NPC’s relationship with the players, rather than their characters—how the NPC makes the players feel OOC.

For those who haven’t seen it, The Last Wish has three sets of antagonists: “Big” Jack Horner, Goldilocks and the Three Bears, and the personification of Death. Let’s go through them in order.

First, Jack Horner is perfectly built for the protagonists to find insulting, disgusting, and altogether unpleasant—he’s arrogant, rude, and repulsively unsympathetic due to his goals and backstory. Beyond that, though, he’s also built for the viewers to find funny—his sheer evil and cruelty are so cartoonishly over-the-top that they loop around to being actively amusing.

Second, Goldilocks and the Three Bears are fantastic examples of likeable NPCs. They’re sympathetic (they’re members of a close-knit family who truly care for one another), endearing (they’re far more bumbling than intimidating at most points), and genuinely amusing (each member of the family has their own idiosyncrasies that make for solid humor).

Third—and most importantly—Death highlights an incredibly important lesson for DMs to remember: You can’t rely on fear alone to make a character memorable. That’s not to say that Death isn’t intimidating as all hell. But what makes him truly stick out are the things he says and does that aren’t scary or spooky. When he formally introduces himself, he says:

“[I’m] Death. And I don't mean it metaphorically or rhetorically or poetically or theoretically or any other fancy way. I'm Death. Straight. Up.”

A more traditional “Grim Reaper” approach to this dialogue might have tried to be spookier, more foreboding, and genuinely vague. But instead of trying to scare the protagonists further, the movie let its villain breathe in a way that is unmistakably more human. In The Last Wish, Death isn’t one to put on airs; because he has a distaste for “fancy” speech, his resonance becomes almost endearing—he’s an everyman, not a scholar or sage.

Beyond this, Death builds a sympathetic resonance with the protagonists: his whole reason for being in the movie is that he feels that Puss in Boots has disrespected the opportunities he’s been given, and he’s here to exact justice. That’s a far more powerful and memorable motivation than a version of Death who merely wants to kill a long-lived character because he likes it when people die.

Through these examples of resonance, the film turns a set of potentially forgettable antagonists into a vivid gallery of absolutely memorable characters. So next time you’re designing an NPC, make sure to spend at least a few moments considering why your players should remember them—and how you’ll use resonance to accomplish that goal.

Dragna’s Campaign Advice Roundup | January 29, 2023

• Good game design requires more than just cool ideas. Make sure you always start by identifying your dramatic question. Then, make sure to consciously and intentionally apply your skills as a designer to find ways to accentuate, develop, or otherwise explore that dramatic question. If you find a way to use that cool idea, great! If not, there’s no shame in discarding it and going back to the drawing board—it’s better to have a mundane, but effective execution than a radical, but ineffective implementation.
• My house rule for monster knowledge checks looks like this: DC 10: You recognize the monster / DC 15: You know the monster's unique features and tactics, as well as any unique resistances, weaknesses, or immunities / DC 20: You know the monster's resistances, weaknesses, and immunities. Note, however, that these DCs are for common monsters, like werewolves or ghosts or goblins For uncommon monsters like demons, all the DCs go up +5; for rare monsters like aberrations, all the DCs go up +10.
• When playing a companion NPC, it’s okay to make them an active character in party conversations so long as you remember that, at the end of the day, they’re only resources to the players and can’t actually make choices on behalf of the party themselves. (To avoid the “Mary Sue” problem, it can help to play into that NPC’s flaws as well, especially if you can do so in a way that extends their resonance.)
• Don’t forget that you can find a full timeline of RAW Barovia in my Comprehensive History of 5th Edition Barovia!