Patreon DevBlog #4: Design Intent Matters

The infamous DevBlog returns! This week’s projects included Curse of Strahd: Re-Reloaded, Dungeon Mentor, and the final Patreon homebrew content pack—plus some minor commentaries. Let’s dive in!

Problem 1: How can I ensure the PCs experience the Strahd encounter at the River Ivlis crossroads?

Ah, the introductory Strahd encounter of Re-Reloaded—truly, the ride never ends with you.

Still, our problems are slowly getting smaller in scope, which I tend to view as a good thing! In this case, the problem was brought to my attention by a playtester running the guide, who reported back that their players had heard the sound of Strahd’s carriage approaching, and then swiftly ran into the woods.

And kept running.

And didn’t stop.

(What really gets me is that, apparently, the players didn’t even think it was Strahd—they just thought it was a random Vistani wagon and weren’t in the mood to socialize. Wacky.)

This is one of the key things to remember about DMing design: never give your players an option that you aren’t willing to let them take. By allowing the players to overhear Strahd’s carriage before it arrived, I’d unwittingly given them a perfect opportunity to run away and hide, and therefore avoid the entire encounter altogether.

Now, I did have some plans percolating in my head in case they simply hid. Strahd has a massive passive Perception, and his dire wolves have great senses of smell—there’s simply no way the PCs would be able to hide from him or his “pets” in the nearby woods. Once the PCs are found, it’s easy for Strahd to lightly suggest the PCs exit the woods before his “pets” get too hungry.

But this doesn’t account for the (admittedly, absolutely hilarious) possibility that, instead of sticking around, the players will just keep running for the next solid minute. And yet, by allowing the PCs to overhear the carriage before it arrived, I should have expected this!

Let’s take a look at the map of Barovia. The road from the River Ivlis Crossroads basically heads straight toward Tser Pool for around a quarter-mile, then bends left a bit around a hill. As such, it’d be reasonable to expect that the carriage might first come into sight at the same time that the players start hearing it approach.

But that’s kind of lame, and it also gives the players tons of time to run away, making it even more likely that they’ll miss this encounter (since Strahd’s carriage is obviously Bad News).

However, one thing worth noting is that Barovia is full of fog. In fact, a solid number of the original module’s random encounters start off with stuff like, “A creature steps out of the fog” and just throws you immediately into the action.

So how ‘bout this:

• The PCs hear the sound of a carriage approaching from the direction of Tser Pool through the mist.
• If they turn and run, the dire wolf pack is already behind them and half-surrounding them—the fog muffled and concealed their approach, and Strahd sent them ahead of him once they reported smelling non-Barovians on the wind.
• The dire wolves growl menacingly, strongly suggesting to the PCs, “don’t move.”
• By the time the PCs can respond, the black carriage has already arrived, blocking off the only other means of escape. The encounter then proceeds as originally planned.

Is this a very linear encounter? Yes. Is it railroading? No—for the simple reason that the players don’t have the time to make meaningful choices (and therefore can’t feel as though their choices are being nullified). Everything is happening too fast for the players to meaningfully respond. In the language of a combat encounter, this is an ambush—and Strahd and his dire wolves are just taking advantage of their own “surprise round” to get everything set up how they’d like it.

Problem 2: How can I design a good ghost adventure in which the ghost isn’t the ultimate antagonist?

When I was writing the adventure for Boo!, this past month’s homebrew content pack, I knew that I wanted it to be about a ghost-hunt that turned out to be a red herring—the real villain wasn’t the ghost, but the person who’d killed the ghost.

At first, I wasn’t sure how to set this up. My initial thought was “midnight murder mystery” a la Clue, where a bunch of people are trapped in a manor and the PCs have to figure out who killed the ghost before the ghost kills them and before the killer strikes again. But that felt too complicated (why is the ghost suddenly attacking now? why would the killer strike again? why isn’t the ghost attacking its killer?), and, more importantly was basically a reskinned re-run of Murder at Silvercrest Lodge, an adventure I’d already written.

I spent a while banging my head against the wall trying to figure out how to make a workable “twist” in the story, but all of them basically hinged on the kinds of narrative tropes that are more workable in a TV show or movie than a TTRPG—stuff like “the PCs are lured somewhere by an obviously evil killer” or “the PCs split up and Search For Clues, Gang.” When I tried to reframe it as more of a classic “haunted house” story, I realized quickly I was basically just trying to reinvent Van Richten’s Guide to Ravenloft’s House of Lament adventure, and stopped.

My revelation came when I remembered something very simple: D&D isn’t just a storytelling mechanism; it’s a game. So maybe I needed to approach this adventure as a game first, and as a narrative second.

Fans of Magic: the Gathering might be familiar with the concepts of “top-down” and “bottom-up” design. In short, “top-down” design means starting with a setting and narrative theme, and then building mechanical gameplay to fit it. By contrast, “bottom-up” design means starting with a planned gameplay experience, and then shaping the setting and narrative elements to suit it.

It’s very very common for DMs to approach every adventure with a top-down mindset—but what I found in writing this adventure was that it can be even more powerful to approach it as a bottom-up design process.

The basic idea of bottom-up D&D design is that D&D, as a game, is supported by four pillars of core gameplay:

• Combat encounters (in which the PCs have to accomplish some task in initiative).
• Social encounters (in which the PCs have to accomplish some task in conversation with an NPC),
• Obstacle courses (in which the PCs have to accomplish some task by bypassing multiple consecutive obstacles), and
• Puzzle encounters (in which the PCs have to accomplish a number of little tasks in order to figure out a larger puzzle or mystery)

(You can also add in “skill challenges” as well as a fifth pillar, but those aren’t part of the RAW experience, so we’ll leave them out for now.)

(Also—why am I calling these the four pillars of gameplay instead of, say, “combat, roleplay, and exploration”? Because something as simple as “roleplay” or “exploration” doesn’t actually encapsulate the meaningful choices that players are forced to make in the playing of the actual game. To be a proper “pillar of gameplay,” in my definition, a particular type of challenge must take multiple segments of time to “solve.” This might have been the case for exploration in the past—back when AD&D was all about dungeon-crawls and resource management and avoiding random encounters—but it’s not the case anymore.)

Anyways—back to the adventure. I knew this would be a short adventure, so here’s what my “challenge map” looked like:

_____ > _____ > _____ > _____ > _____

Five or six consecutive gameplay “challenges.” This was a good start. From here, I knew that I wanted to end the adventure on a climactic combat encounter (which, after all, is what 5e is best at). I also knew that I wanted to introduce the adventure with some basic problem-solving to get the players warmed up, without necessarily throwing them into the deep end of high-tension combat or roleplay upfront. I was feeling nostalgic for Scooby Doo, so I decided to make a classic “find the hidden doorway/staircase” puzzle. Here’s what my challenge map looked like now:

Puzzle (Hidden Door) > _____ > _____ > _____ > Combat

From here, I knew that I wanted the core of the adventure to be about solving the ghost’s “unfinished business” somehow. I knew that this had to involve some kind of puzzle or social encounter, and that it had to come immediately before the climax. So:

Puzzle (Hidden Door) > _____ > _____ > Puzzle/Social (Ghost) > Combat (Murderer)

Things were starting to take shape! As I pondered more, the beginnings of a narrative began to take shape: the ghost was a victim of the master of the house, who had locked them in the celler and tortured them there.

But narrative flavor isn’t enough; I needed some kind of gameplay to make it feel real. Fortunately, a few months back, I went on a binge of watching assorted Saw clips on YouTube, and realized what how I could communicate that to the players: As they delve deeper into the cellar of the house, they have to go through the same sadistic death-traps that the original victims faced. That meant…another puzzle encounter!

Puzzle (Hidden Door) > Puzzle (Trap Room) > _____ > Puzzle/Social (Ghost) > Combat (Murderer)

Of course, I couldn’t have an adventure that was just entirely puzzles. You need a little bit of combat to spice things up. So I decided to add in a combat encounter with some weaker spirits—other victims of the house—to add a nice bit of action close to the midpoint.

Puzzle (Hidden Door) > Puzzle (Trap Room) > Combat (Specters) > Puzzle/Social (Ghost) > Combat (Murderer)

But how else would the PCs get to the house unless the master of the house invited them there in the first place? This felt like a tough nut to crack…until I started to wonder if the murderer didn’t know what they’d been doing? I know that this is a trope I’ve been relying on a lot lately, but the ol’ “possessed by a demon” trope is a classic one in horror fiction for a reason!

So here’s the critical path I wound up with:

• The PCs get to the house to go ghost-hunting. The owner shows them around and tells them where the ghost has been seen.
• Through investigation, the PCs find the hidden door. Using it, however, traps them in the killer’s death-trap dungeon.
• The PCs have to pass through a death-trap room. After that, they have to fight off a bunch of angry specters who were the killer’s victims before them.
• The PCs find the ghost and its corpse. The ghost can’t talk, so the PCs have to work a bit harder to communicate with it and find out what’s going on. The ghost then shows them a secret door to escape the dungeon.
• The PCs are confronted by the owner (now controlled by the demon) and have to fight him.

All in all, a reasonably satisfying short horror adventure that evokes a bunch of different tropes while still presenting a (hopefully) engaging gameplay experience.

Problem 3: How can I add animations to my videos?

This is more of a pro-tip than a proper developer commentary—while I was making videos for Dungeon Mentor last week, one of my pre-viewers suggested adding animations to spruce ‘em up. This is something I’ve been meaning to learn for a while, so I set off reluctantly, somewhat afraid of the taxing experience I was about to endure.

Instead, I found this video, which taught me how to make top-tier, easy animations using DaVinci Resolve (my video editing software) in about two minutes. Guess I shouldn’t have waited so long to face my fears!

Designer Commentary: Setting the Proper Stakes

As some of you might have heard, I’m currently running a short homebrew campaign for a table of ten newbies to introduce them to the wonderful world of D&D.

The first quest—a short one-shot involving retrieving a stolen magical gemstone—went off (largely) without a hitch. And they’re certainly enjoying the second quest as well. But something about the process of running it has left a bit of a bad taste in my mouth, and I think I finally realized why.

The adventure in question is simple: stop a band of kobolds, led by an evil half-dragon, from reviving an adult dragon that was petrified long ago in the base of a mountain. Along the way, I also tossed in that the dragon had been sealed in the mountain alongside a powerful demon, and that the mountain was actually a volcano—and that, if the ritual went off, the dragon would likely conquer the surrounding area and the demon’s volcano might erupt.

The problem: the players were level 1.

Now, that’s not to say that they were outmatched! Having used the CR2.0 encounter builder to precisely calibrate my encounter difficulties, I felt comfortable that my players weren’t facing anything they couldn’t handle.

The question was more fundamental than that: Why was a group of 1st-level characters expected to accomplish something of this magnitude? And, more to the point, if reviving this dragon were this easy, why hadn’t anybody more powerful than this bumbling band of kobolds tried to do it before?

My players, of course, don’t really notice the dissonance. I doubt they would even if they were somewhat more experienced. But the dissonance persists nonetheless.

D&D’s “tiers of play” divides player levels into four “tiers”:

• Tier 1: Local Heroes (levels 1-4). The PCs undertake small quests with stakes relevant to individual villages, towns, or neighborhoods.
• Tier 2: Heroes of the Realm (levels 5-10). The PCs undertake intermediate quests with stakes relevant to large cities, as well as entire counties or countries.
• Tier 3: Masters of the Realm (levels 11-16). The PCs undertake large quests with stakes relevant to multiple countries or entire continents.
• Tier 4: Masters of the World (levels 17-20). The PCs undertake massive quests with stakes relevant to the entire world—or multiverse.

Why is it important to make the stakes scale with the PCs’ power?

• From a narrative standpoint, it provides an important sense of “fairness”—low-level PCs aren’t going to “cheat” their way to reknown and fame by accidentally and easily doing something with incredibly high stakes, and high-level PCs aren’t going to waste their time by spending it all on stakes that could be done by local town guards.
• From a gameplay standpoint, it ensures that the players feel neither overwhelmed by the prospect of failure (preventing them from panicking or freaking out mid-game), nor underwhelmed by the prospect of success (keeping them motivated to do their best).
• From a standpoint of immersion, it preserves verisimilitude—due to the trappings of popular media, we tend to just assume that stakes will scale with power like this. Captain America shouldn’t spend most of his time beating up schoolyard bullies, and Oliver Twist shouldn’t spend most of his time accidentally saving the world from destruction.

The adventure I’d designed was meant for 1st-level characters, but carried stakes more appropriate for 5th/6th/7th-level characters. In retrospect, that was a misstep, and I think I’d do things somewhat differently in the future.

With that said, it’s worth taking a look at this from a different vantage point. Experienced D&D players often complain that the first few levels of play are boring because it’s basically three levels of shooting goblins and dealing with bandits. More significantly, many new D&D players often want to play D&D because they want to feel like Captain America—they have a vision in their head of Legolas and Boromir fighting to save the world, and they want that! D&D is a game about agency, but it’s the stakes that make that agency feel important. Lowering stakes means devaluing agency.

So what does this mean in the long run? I’m not entirely sure. I do think that it’s important to understand the tiers of play, but I also think it’s important to avoid the trap of making every low-level quest feel like a D-rank mission. I think I plan to ponder this further, but for now, I’ll just leave it at this: No matter what level the players are, their mission should feel important—to the point of life-or-death—to someone. If that “someone” is the players themselves, that’s all the better.

Dragna’s Campaign Advice Roundup | February 5, 2023

• If you can avoid it, never ask your players OOC if they’re going to do X or Y. Rather than suggesting a course of action, just set the scene and stakes and ask what they’d like to do.
• Whenever you’re thinking of adding new content to an ongoing adventure—such as a random encounter, an additional combat, or pretty much anything that the players might experience—use intent-first design. Start by hammering out your dramatic question for the current adventure. Then, try to figure out how this idea progresses, develops, or otherwise accentuates that specific dramatic question. If you can’t think of a way to make this idea do that, it’s time to go back to the drawing board! In general, if adding additional content doesn’t add anything to the core narrative (i.e., the main dramatic question), it might not be a good idea to include it at all!
• You can also do this inversely—start with a particular way in which you want to progress, develop, or accentuate the dramatic question, and then try to reverse-enginer an idea that accomplishes that goal. (Then, your main remaining task is to figure out how to implement that idea as an actual gameplay challenge—combat encounter, social encounter, puzzle encounter, obstacle course, and/or skill challenge.)
• Remember that creatures have disadvantage on Perception checks made in dim light (and -5 on passive Perception). If a creature has darkvision, which makes them treat darkness as dim light, they now have this disadvantage/-5 penalty on darkness instead.
• Except in very specific types of player-driven campaigns, you should always try to make sure that, except for downtime, your players have a clear dramatic question available to them at all times. This doesn’t mean you need a complicated dramatic question, or one that’s particularly challenging—it just means you need to work with and around the players to make sure there’s always something for them to do at all times.
• Chase-style skill challenges work best either when the players have strong reason not to want to fight, or when the players have basically no hope of defeating the monsters and therefore must flee in order to survive. (In CR2.0 terms, this usually means a Devastating encounter or higher.)