Water Takes You Home — A Rambling Postmortem of Dragon Heist

TL;DR – If you’re a DM thinking about running Waterdeep – Dragon Heist and are looking for tips from someone who barely knows what he’s doing, feel free to skip to the bulleted list at the end of this harangue.

Recently, I finished up the most recent D&D campaign I was running that began back in January of 2019 (diligent readers may recall a short-lived series of campaign diaries I attempted to write — these may return!). Ostensibly based upon the two recent published D&D 5th edition adventures, Waterdeep – Dragon Heist and Waterdeep – Dungeon of the Mad Mage. That said, as anyone who’s played with me can tell you, I have a tendency to go off the rails, speed up toward the finish line, and generally rework the material so I can keep what I think is cool and scrap what isn’t.

What this is intended to be is a way for me to gather thoughts about how I thought the last 21 months of game-mastering went in an attempt not only to make my next game better, but also to provide some theoretically useful insight into the process for prospective DMs, especially those interested in running one or both of the aforementioned adventures. A lot of this is going to be based on several long Twitter threads wherein I’ve discussed this experience at length. I’ll try to keep it as interesting and/or engaging as possible, but as the title of this post suggests there’s going to be a lot of mildly-incoherent rambling and stream-of-consciousness trains of thought that might not go anywhere. I don’t have an outline I’m working off of; I’ll try and keep it roughly chronological but I’ll undoubtedly jump back and forth occasionally. If you’re interested in what passes for “usual fare” on this blog, namely video games and music, I’d advise you basically not to read this, since you’ll probably be bored to tears. As I often say, nothing is more interesting to someone than their own D&D campaign, and nothing is less interesting than someone else’s. Oh well. Let’s get to it.

I decided I was going to run one or both of these adventures shortly after they were announced in the summer of 2018; at the time I was in the midst of a run of the 2015 adventure Out of the Abyss. The Waterdeep adventures appealed to me for a few reasons. Primarily, OotA was the first time I’d run a published 5e adventure and, simply put, I wasn’t incredibly impressed with it. It was, in essence, a collection of encounters, locations, and scenarios, running the gamut from fascinating spectacles to boring trudges through mires of spiders, drow, and mushroom people. I wanted to run another published adventure to get a better idea of how it worked, and my options were limited by my group’s experiences, given that all of us had experienced most of the published work already, and I’d heard unenthusiastic reviews of the one’s we hadn’t touched.

I’d be lying if I said Matt Mercer’s involvement in the story development if Dragon Heist wasn’t a factor as well, not to mention my general interest in Waterdeep as a setting. I’d gotten a small taste of the Forgotten Realms from OotA, and I was eager to explore it, and my version of it, further with the same group. I also liked the urban aspect of WDH. OotA begins with your players trapped far in the Underdark, and essentially tries (and largely fails) to be a survival-horror type of game where you’re relentlessly pursued by insane drow and facing off against demon lords while you try to escape. And then, after you do escape, you … have to go back down. Kind of a bummer. I didn’t like that whole “your players are trapped in an interesting but incredibly repetitive location for their entire tenure as adventurers” thing, and I thought setting it in a deep, but contained sandbox like Waterdeep would be great.

And it was! I’ve said this on Twitter a lot but I think WDH is my favorite published adventure from 5e (Red Hand of Doom is probably my favorite overall from any edition). It makes a lot of smart changes to the formula, many of which are unfortunately overturned by the adventure that ostensibly follows it (more on that later).

I brought over 3/4 of my players from 2018’s OotA game and started in January of 2019 with one additional player. We started out with a solid group — an elf druid, gnome wizard, human cleric, and warforged craftsman (a homebrew class from Mage Hand Press). The only thing we were maybe missing was a rogue, but between the rest of the group they found they could fill in whatever gaps were left there with clever Wild Shapes, the use of invisibility and knock, and whatever dumb shit the craftsman class could make.

The craftsman was, like a lot of homebrew classes, an interesting idea that was implemented with mixed results. Unfortunately, most homebrew designers seem to kind of miss the point of 5e’s design philosophy — trying to make robust rules about crafting just plain doesn’t work with the limited skills and tools available to players, and without a substantial amount of work on the part of the DM (which I tried to do), it ends up feeling either overpowered or worthless. No one wants to feel worthless, so I opted to err on the side of giving the players too much, as I always tend to. As I often say, I’d rather the players feel like they’re more powerful than they should be than less, and it gives me an excuse to use the cooler monsters against them, anyway. But, ultimately, the craftsman’s costs exceeded its benefits and he began to multiclass a few different times, eventually transferring some levels over to artificer before finally scrapping them altogether and taking fighter levels instead.

In keeping with my philosophy of trying not to make my players hate playing their characters, I’ve always been flexible about letting them transfer class levels around (usually on level up) as they see fit to try and tweak something that isn’t working. One of my players from the previous homebrew game I ran went through, like, four different iterations of her character that just didn’t work at all before finally settling on ranger. Believe me, the irony of going from a litany of shitty homebrew classes to the unequivocal worst class in official 5e is not lost on me. Luckily my warforged PC’s player was more than happy to experiment and I think he was generally grateful at the flexibility I gave him when stuff just wasn’t jibing the way it ought to have. In retrospect, as I should’ve already known, homebrew classes and subclasses are almost never a good idea. We can’t really tell if something is balanced until we play it, though, so I have no regrets about trying it out and I hope he doesn’t either.

Anyway, Dragon Heist. As a general overview, it does a fantastic job of packing a lot of fun into a small package. For starters, it’s interesting off the bat by allowing you to pick one of four villains. I like to suck all the interesting content out of every source book I feasibly can, and all of the villain’s lairs were too cool to pass up (and indeed basically the only real dungeons in the campaign), so I opted to utilize all four villains in some capacity or another, in ways that often managed to worm their ways into player backstories and into the endgame content in Dungeon of the Mad Mage.

At the start of the campaign I gave the party short blurbs about each of the villains — Jarlaxle Baenre, the Casslanters, Manshoon, and Xanathar — with just enough to give them an idea of who they were and why they were after the money without spoiling too much. They decided pretty much unanimously on Jarlaxle — I think in part because they’d met him briefly in OotA and were interested in seeing him again, with a close second in Xanathar. With that in mind, I decided to have Jarlaxle be the primary “villain” (villain in scare quotes because I have a tendency to make 99% of my NPCs insanely reasonable and Jarlaxle’s goals just didn’t seem all that nefarious, as it turned out) with Xanathar being a known entity and consistent threat.

Background-wise, I opted to have our gnome wizard Wobbles be formerly in debt to the Zhentarim, due to a protection racket his father had bought into, and also had him be a semi-recent temporary victim of Xanathar’s wherein he’d been temporarily petrified by him. He was also a member of the Lords’ Alliance. A note for prospective DMs — factions are super important in WDH, so get to know who they are and try and encourage your players to be a part of one of them, because it makes up a large part of the actual questing content in the second chapter of the book and beyond.

Our warforged Copper was an adoptive son of Wobbles and tentatively interested in the Gray Hands. There wasn’t a ton else to do with him, faction or backstory-wise given the fact that warforged aren’t technically people and are generally looked down on. I cleared this with the player before starting, obviously. No one wants it sprung on them that everyone is going to be racist to them. The druid Gardenia was an orphan who had drawn the attention of the Emerald Enclave — she was also a teenager, which is kind of weird as far as elf-human relations go since elves don’t consider you an adult until you’re a hundred.

Our cleric Jo had some background with the Order of the Gauntlet and Force Grey, but she ultimately didn’t end up sticking around in the game for personal reasons. She dipped out basically around the climax of the first act of WDH, so it served well as a time to bring in two new players. I asked two players from another game of mine that was nearing its end by this point if they would be interested — both said yes. The first one came back almost immediately with a hexblade warlock. “Yes, perfect,” I thought. See, I had been reading ahead and — even though we were only in chapter 2 — I was concerned about the internal logic, or lack thereof, that bound Dragon Heist to Dungeon of the Mad Mage. It’s basically a geographical connection, and that’s it. Hm.

With that in mind, I began to come up with an elaborate backstory for Halaster Blackcloak (the titular mad mage) involving the “knot in the Weave” that he’s evidently got … some sort of connection to. I decided that he had gone basically insane and that his mind had fractured into two — a sane version and an insane version, who swapped like Jason Isaacs in that show Awake that got cancelled way too early, every time he went to sleep. The sane version would be fed misinformation by the insane version about how to “untie the knot”, believing that it would free him from his insane self and bring peace back to Waterdeep — in reality it would allow the insane version to regain full control.

I made some bullshit up about a ritual involving seven apprentices that would accomplish this, and decided that Nia, our warlock, would have her patron be sane!Halaster, unbeknownst to her. She gave me some great background of herself as a member of a disgraced noble family in Waterdeep (which coincidentally finally gave me a way to involve the Cassalanters), and joined the Harpers — this gave me an opportunity to make more use of Mirt, basically my favorite NPC from WDH, as a sort of adoptive father. I played her patron as a sort of distant, vaguely confused figure who, as the campaign wore on, got more and more aware and more easily able to communicate with her. From the get-go, I decided he would call her “dear one” in their conversations, as a way to twist the knife in the distant future when they finally realized it was Halaster all along. Great! Now they’ll eventually have a reason to venture below, since they’ll necessarily be among her seven “apprentices”. Who are the others? No idea. What’s the ritual? Who knows? Whatever.

Then, our second new player gets back to me, saying she’s going to play … a warlock. Shit. I had a great idea for the first one, but one idea was all I had. Oh well, she’s going warlock of the fiend, anyway … I guess Asmodeus? He’s got that Cassalanter connection, maybe I can use that? And her character (Hihro “the dragonborne”, a kobold) is part of the Zhentarim? Okay. I can do something with this. I decided that Hihro thought his patron was actually Manshoon — Manshoon did some NPC-only magic mumbo jumbo to sort of intercede between Asmodeus and Hihro and trick him into doing his bidding. I put a glyph on his pendant that served as an arcane focus to Feeblemind him if he ever said Manshoon’s name aloud, which I thought was neat. I still had no idea what to do with Asmodeus, though.

Anyway, I digress. That’s the party. Our first 7 or 8 sessions were glacially slow.

I pulled a lot of extra content in for Dragon Heist. I’m not sure why. I think I was just excited to be doing something so different from OotA and wanted to milk as much out of the urban aspect of this adventure as I could. In retrospect, I’m glad I did. As it’s written, WDH is a lean adventure — it’s most likely meant to take somewhere between 5 and 10 sessions and only runs from level 1 to 5. It’s structured really, really well, though, and I love that. Chapter 1 is your introductory adventure to introduce the plot, chapter 2 is a sandbox where you get settled in Trollskull Manor and run missions for your factions, chapter 3 is where the shit hits the fan, chapter 4 is a long chase for the MacGuffin that’s different depending on your villain, and finally the actual vault and the acquisition of the titular dragons (gold coins). Chapters 5, 6, 7, and 8 are a bit … odd. They’re basically just dungeons, the lairs of each of our villains, but it’s not super clear when exactly you’re supposed to use them. I had an easy workaround for this, which we’ll get to later.

As far as all the additional content I pulled, you can comb through the campaign diaries I did get around to writing, as I call out most of it there. Mostly stuff from dmsguild.com, which I highly recommend as a resource for extra stuff. There’s a lot of trash there, but there are a few diamonds in the rough that will add a lot to your games. The first chapter, apart from the first session, was pretty smooth. I added the Rats of Waterdeep adventure from the DM’s Guild which was a lot of fun, and added some fun recurring NPCs for later. The party was instantly attached to Waterdeep City Watch captain Hysutus Staget, who became a recurring character, as well.

Chapter two was also smooth, and we spent a lot of time just renovating the manor into a functional tavern. This served as an opportunity to introduce Mirt (Nia hadn’t come into the party yet) and get them acquainted with the bureaucracy — we had a lot of fun just talking to the different guilds involved, doing favors for people to lower costs, and exploring the garbage leftover in the manor. I had a ton of fun with Emmek Frewn, the slovenly tavernkeep who moves in across the street, and they loved thwarting his plans and dealing with his poorly-enacted ideas to run them out of business.

By the time we got to the inciting incident of the adventure proper we were already 8 or 9 sessions in. The fireball was a great time to introduce Nia to the group — assigned by the Harpers to keep an eye on the place since Mirt suspected they were soon to be involved in something nefarious. The investigation afterward was a delight, and a good opportunity to have them make use of Vincent Trench’s services — they had gone to him before a number of times and were quite attached to him already. He’s secretly a rakshasa. It’s nuts. The book gives you exactly nothing to do with that, so I had to elaborate on that much later in the story.

They eventually found out that the fireball was thrown by a nimblewright, which gave Copper plenty of fun interactions with Valetta at the House of Inspired Hands, as he’d always been curious about constructs in an attempt to discover his origins. When the investigation led them to Gralhund Villa, I had a perfect chance to bring Hihro in, our fifth PC, as a Zhentarim contact assigned to implant himself in the group.

Things got a bit dodgy with him since the party didn’t necessarily trust him from the outset despite his attempts to ingratiate himself (even though he was a warlock he was dreadful at persuading or deceiving anyone) and even going into the encounter chain of chapter 4 he still wasn’t a really solid member of the group. Before the encounter chain got set off in full, though, they had breakfast with a certain Zardoz Zord … Jarlaxle in disguise, of course.

It was an extremely fun roleplaying encounter, with Jarlaxle using the opportunity to assess their strengths and determine if they were a threat, and them not any the wiser about who he actually was. (He had come into the tavern a few days before, as well).

The encounter chain began, with the deceptive Jarlaxle proving an increasingly entertaining villain — they totally bought every single one of his disguises, even when he pretended to be the Open Lord, of all people, so their confusion at finding out it had been Jarlaxle all along was delicious. I was especially proud of one of my machinations — the encounter at Mistshore with the mechanical dragon turtle nearly wiped everyone out, and they were arrested by the Watch for the ruckus they’d caused. Captain Staget took them in and locked them in a cell, where they attempted to contact Laeral Silverhand with a Paper Bird to exonerate themselves. When their email got bounced back to them, they resorted to having Nia contact her benefactor (Mirt, though the rest of them didn’t know it yet). He told them the constabulary they were in was out of operation — it was Jarlaxle pretending to be Captain Staget all along! Aha!

They found the real captain tied up, and then went to infiltrate the submarine … which went poorly. They learned the truth of Zardoz’s identity, and were unable to avoid his detection. This is where he hired them to take the Stone of Golorr, and find the money for him, offering them 10% of the reward … this was where I decided to embellish even further.

See, the Stone of Golorr gives you three clues to three keys you need to unlock the Vault of Dragons, with a list of random ones to choose from. Some are okay, some are pretty boring. I decided to keep the “you need a celestial” and “you need a shapechanger” keys. The first, they had some discussion about — they thought about trying to use Primara, the unicorn rescued earlier in Blue Alley (another side adventure), and they talked the Blackstaff (to whom Copper had been getting closer) into summoning a couatl for them, but then Wobbles realized that his ever-present familiar Ichabod was actually a celestial all along. Go figure. As for the shapechanger … they had a druid.

I decided to add two extra keys for a total of five, so that I could provide an excuse to have them explore the other three villainous lairs. As such, they had to steal one of Manshoon’s masks, the Mace of Disruption from beneath Cassalanter Villa, and Sylgar, Xanathar’s treasured goldfish.

I don’t recall the exact orders everything went in, but all the while I was peppering in various faction missions — Harper missions for Nia, Gray Hands missions for Copper, Lords’ Alliance missions for Wobbles, and Emerald Enclave missions for Gardenia. I think there might have been a Zhent mission in there for Hihro, too, but those tended to be a bit harder to work into a typically Lawful Good party. Many of the faction missions I ended up expounding on in one way or another. The poop-sweeping mission from the Lords’ Alliance was a particular highlight, as was the Harper mission to find a missing talking horse and the mission involving finding the Intellect Devourer that had taken over Meloon Wardragon — this one led to Copper being granted Azuredge, a powerful magic axe.

Throughout, they also had some great NPC interactions, as well. Wobbles took a liking to using dust of dryness to create small beads of sewage and throwing them into Frewn’s tavern, Gardenia made friends in Fala, the herbalist neighbor, as well as Cat and Maladie from Rats of Waterdeep. Mirt was always a favorite, as well as Esvele Rosznar, the Black Viper, who was a close friend of Nia’s from youth. Copper drew the attention of not only the Blackstaff, but also a jeweler shopkeep I came up with off the cuff, and the gang also became tight with another invented shopkeeper Merric Fastfoot, who was pretty much based on Toehider’s How Much for that Dragon Tooth? of all things. Wobbles and Copper had a lot of interactions with the Doom Raiders, especially Davil Starsong and Ziraj the Hunter, thanks to Wobbles’s history with the Zhentarim, and those were a lot of fun, too. And who can forget Durnan, the gruff proprietor of the Yawning Portal, or the flamboyant Volo, writer extraordinaire?

All that said, nobody was a better NPC companion than their foe-turned-tentative-ally, Jarlaxle Baenre. The guy is just too damn charismatic not to be likable, and though their work for him to retrieve the other keys began under threat of death, they eventually grew to sympathize with him. His reasoning for wanting the money is, after all, far less sinister than the other three potential villains. So by the time they found the vault, they were advocating for him to the Open Lord. Fun twists.

At any rate, they found out that the Vault itself was located under a windmill recently purchased by Esvele (I pilfered this from a different encounter chain since I liked it better), who wanted a cut of the gold, as well. First up of their keys to abscond with was the one in Cassalanter Villa. They managed to sneak in through the mud tunnels (the book doesn’t actually tell you where these go so I said they spat out along the coast) and explored a bit before finding the Mace and then learning that a whole bunch of cultists of Asmodeus had come in behind them. Uh-oh. They snuck out through the house in a hilarious bit of farce, involving Ichabod, Gardenia turning into a cat, and sneaking around through secret passages behind paintings like some sort of maniacal game of Clue, in order to avoid the Cassalanter children.

Next up was Xanathar, and that was a real doozy. Jarlaxle had informed them of Nar’l Xibrindas, a drow planning to blow the place up, and given that they had long since run Emmek out of town I opted to have him here as a statue, evidently a victim of Xanathar’s rage. They blew some of the place up, pooped on Xanny boy, and stole the fish, but not before Hihro met an unfortunate demise at the hands (paws?) of an Intellect Devourer. This was a bummer, but ultimately for the best since, admittedly, I don’t think either of us had any clue where his character was going to go. It was unfortunate that we couldn’t have his showdown with Manshoon, but it was what it was. We also got to meet the best-named NPC ever, Bepis Honeymaker. So that was a win.

After retrieving 4/5 keys, they decided to come clean to Mirt about everything, including their working for Jarlaxle, and he was pretty much understanding of their choices. It was a rough situation, after all. I decided to introduce some much-needed fanciness to our game, here, by having them receive a few different missions at a ball held by Remallia Haventree. I threw a couple of faction missions together and also noted that it was likely that Jarlaxle would be there in disguise along with an old Harper associate of Mattrim “Threestrings” Mereg, namely Humphrey B. Bear (no relation), Hihro’s player’s new character.

The ball was a blast, and it gave them an opportunity to explain how their characters all dressed up for the occasion. I had Wobbles run into his ex-wife, Copper ran into Celniana, the coy jeweler with a crush on him, Gardenia and Nia thwarted some nasty guys from taking advantage of Remallia’s drunken daughter (they ended up being drow in disguise), Humphrey helped root out some doppelgangers, and ultimately Jarlaxle escaped before they could learn who he was … and then came Gardenia’s crazy uncle.

Gardenia had it as part of her backstory that her parents were part of a bioterrorist sect of druids who had staged a siege against Waterdeep years earlier when she was a kid — that’s how she ended up as an orphan. Her crazy uncle worked for Manshoon and had a simulacrum made of him in an effort to capture her, make a clone of her, and implant it back into the group. That would’ve been cool, but … they managed to defeat him. Life is full of surprises.

With their new member in tow, they finally made their way to Kolat Towers and infiltrated it pretty quickly. Jarlaxle waited outside with Humphrey as collateral (his player couldn’t make it), and they got into Manshoon’s sanctum and fought a simulacrum. Bear in mind, they were level 5 for this, so that was no mean feat. The got the mask and got out of dodge, and headed over to the Vault at Jarlaxle’s behest.

The actual Vault is pretty underwhelming, to be honest. It took about half an hour to make it through with Jarlaxle, and when they came out I staged another fight with another Manshoon simulacrum (this is a recurring theme), Skeemo, the traitorous Doom Raider, and some other lackeys. Davil and Ziraj showed up to help, and before long he was vanquished and the Open Lord, the Blackstaff, and Mirt showed up to negotiate with Jarlaxle. By now, the party had taken something of a liking to the drow rapscallion, and they put in a good word for him with the Open Lord. Because of this, she agreed to let Luskan into the Lords’ Alliance (his goal from the start) and to give the party their agreed-upon 10%. How nice!

That’s pretty much how Dragon Heist went. I added a lot of extra shit to it. There was lots of roleplay opportunities, city exploration, development of their tavern business, and advancement within their various factions. I’m much more proud of how this part of the campaign went than with how Dungeon of the Mad Mage turned out. All told, Dragon Heist took us a whopping 29 sessions to complete. 5-10 my ass. With that, we were on to the much, much worse, in DotMM.

https://mobile.twitter.com/Cheddarface_/status/1184677951884840962

WDH is, once again, a delightful urban adventure. It’s got 50 or so pages of campaign guide at the back in the form of Volo’s Waterdeep Enchiridion, which is a wonderful supplement describing the city. I appreciate this setup greatly — having a campaign setting separate from the actual adventure is a much better way to organize things, and I wish OotA had done this as well. DotMM is a much less impressive romp, despite being three times as long.

Lessons learned for prospective DMs of Waterdeep – Dragon Heist

  • Factions – as I mentioned above, factions are a huge part of the middle part of this adventure — between the mission to retrieve Floon Blagmar in chapter one and the fireball incident that sparks the chase for the Stone of Golorr, there’s not much else to do other than renovate the tavern (which doesn’t provide much in the way of experiential gains as written) and deal with Frewn (which could include a couple encounters if you expand on the provided material) unless your players are affiliated with factions. The faction missions are delightful little blurbs that are easy to expand on and provide some really great creative sparks — in fact this applies pretty much to all of chapter two, the Frewn plans and the renovation of the tavern included.
    • If you’re a DM make sure you read up on the 6 factions (the 5 basic factions of Faerun are included, along with Bregan D’aerthe). It’s fairly easy to divide your players into ones that make sense — rogues and warlocks can fit into the Zhentarim easily, bards, artificers, and wizards can fit in with the Harpers, clerics and paladins line up with the Order of the Gauntlet, druids and rangers will like the Emerald Enclave, and basically any class can join the Lords’ Alliance.
    • It’s a little harder to incorporate Bregan D’aerthe — they’re a bit more morally grey and they strictly only allow male drow among their ranks so the likelihood of being able to utilize them is slim.
    • In any case, it’s fairly easy to transplant missions between the factions with a little effort. Harper, Lords’ Alliance, and Order of the Gauntlet Missions can all rotate pretty freely — the Zhent missions are a bit less cut-and-dry and the Emerald Enclave missions are more nature-themed, but honestly you can swap in whatever you think is cool.
    • If your players aren’t receptive to the faction idea you can still introduce the NPCs affiliated with them and give them the missions that way. It’s recommended that they take out loans to build the tavern, either from Mirt or the Doom Raiders’ Istrid Horn. Mirt could sweeten the deal a bit if they do some Harper missions for him, and Istrid can likewise do that with Zhent missions. The Lords’ Alliance could call upon them as a civic duty, and the Emerald Enclave theoretically do so as well. The Order of the Gauntlet missions could slot in if they ever require help from a temple (to cure a disease or revive a fallen party member, for example). Once again, there are barely any ways to incorporate the Bregan D’aerthe missions, but if your players end up in my situation where Jarlaxle coerces them into procuring the hoard for him, he could also use the opportunity of having them under his thumb to send them on some missions. I suppose you could use those as an example of what the drow are up to, and send your players to thwart them.
  • Waterdeep – It may sound obvious, but if you’re going to run this adventure it pays to know as much about Waterdeep as you reasonably can. Volo’s Waterdeep Enchiridion is an invaluable resource in this regard, and I’d recommend dedicating an afternoon to sitting down and actually reading it. Don’t skim it, read it. Take some notes on the wards, what they’re called, what they’re known for, and who lives there. Some of the important figures around town are worth knowing about too, like the Open Lord, the Gray Hands and their leader the Blackstaff, the City Watch and City Guard (yes, there’s a difference), the underground groups, and well-known business proprietors and religious leaders. It’s worth giving this out to your players, too. It’s pretty fun to use the systems of public transit early on to give a feeling of the scale (hire carriages, drays, etc) and make your players pay for it, but eventually you can just handwave it. Remember to point out the most interesting features — Mount Waterdeep, the Walking Statues, the Yawning Portal, Castle Waterdeep, the Market, etc. Know as much as you can about the city, who’s in charge, and what your players are and aren’t allowed to do. (Wearing weapons is a no-no! That’s going to be a tough sell for adventurers.) If you don’t know, don’t be afraid to pause and look it up. And if it’s not there, make it up, and write it down immediately.
  • NPCs – There are a lot of NPCs in this adventure, a lot of them easily memorable. But Waterdeep is enormous and you’re going to need a good way to pull up more people, quickly. My method for this was to randomly generate 40 NPCs (20 male, 20 female) of varying races, with humans being much more present than half-orcs, dragonborns, and tieflings. I grabbed their name, race, age, some basic personality characteristics, and an easy voice reference and put them on a table. Then, whenever the players interacted with someone for more than a few seconds, I could roll on the table and have a fully-fledged NPC ready to go. Then after the session I could repopulate that cell of the table and add the new NPC to their proper location in my notes. You can use this method to generate a random shop, temple, house, etc. There are lots of resources online to generate them, the trick is to do the legwork beforehand so it seems intentional and you can make sure you only have stuff on your table that you’re excited about.
  • Pacing – Me giving advice on pacing this adventure is a bit like a sloth giving racing tips, but ultimately it’s going to vary heavily depending on your group. If your party isn’t interested in roleplay as much as they are in hacking and slashing, first of all you could just skip straight to Dungeon of the Mad Mage, but if you run this you’ll probably want to breeze through chapter 2 pretty quick to get to the more exciting stuff. The encounter chains all include some fun combat scenarios so you could also try and build your own based on what you think are the most interesting ones. They’ll also probably love the villain lairs so try and incorporate those (you could throw some extra Vault keys in there like I did). If you want to focus on excitement and combat, this adventure is likely to take you a pretty scant amount of time, likely around 5 sessions.
    • Conversely, if you want to do as I did and really milk the hell out of the social aspects of the game like the tavern renovations and interactions with factions, definitely look into the added content available on dmsguild.com — there’s a lot there to expand on all the faction mission blurbs and turn them into full sessions of content in their own right — I went kind of overboard, though, and ended up taking a bit too long. Your players aren’t going to be interested in sitting at levels 1-5 forever, so make sure you either scale up the content of the late game here or get them ready for a long haul. My players spent about 8 sessions at level 2. That’s perhaps 6 too many. Adjust as you go, and don’t be afraid to level your players up when you feel they’ve earned it even when they’re moving past where the book says they ought to be. It’s a lot easier to adjust encounters to be harder than it is to make them easier. Adding more of the same monster tends to be an easy way to do that (CR is a bunk system anyway).
  • Milestones – I’ve never been a fan of XP-based leveling. I think CR and the concept of encounter difficulty are far too vague, and every encounter at early levels is insanely difficult, as every encounter at high level is a cake walk. Only you can know how you DM, and it’s up to you to figure out what’s too hard. In a similar vein I think milestone leveling is always better than XP-based leveling simply because it gives you control over what’s important and worth doing, how long they spend at each level, and how important social encounters are compared to combat ones — especially necessary in this urban adventure that’s far more roleplay-heavy than the others. It’s worth knowing how the adventure is going to play out, on the whole. These estimates are rather vague and definitely not reflective of what my group actually did, but should give you an idea of how much time you need to spend.
    • Chapter one is likely a two-session introduction to the Yawning Portal, our key players, and the city of Waterdeep in general. Expect to get finished with the Zhent hideout by the end of the first, and through the Xanathar Guild hideout before the end of the second. Finish by introducing them to Trollskull Alley and level up to 2. If you think two sessions at level 1 is too many (it might very well be), consider leveling them up after the Zhent hideout, or simply start them at level 2. It’s not going to change much in the way of difficulty.
    • Chapter two is the most fluid of the chapters in the book. Depending on how long you want to make renovating the tavern take, you could just introduce their moneylender of choice right away, handwave the time and money spent, and send them on their way in 10 minutes. You could also spend several sessions sending them to guilds, coercing laborers and doing favors to get the renovations they want done. How descriptive and granular you are largely depends on your group. All that said, this is your time to use the first two (or possibly three, if you need more content) faction missions from each of the factions you can reasonably affiliate your players with. Some of the missions will take half a session, some a full one, but none should take more than one. I’d say a good metric to follow here is (assuming you don’t add any extraneous missions from outside resources) to give them between 8 and 12 faction missions here. That seems like a lot, but you can (and should) break up every 2 or 3 with business with Emmek Frewn, excuses to get to know their neighbors, and interactions with characters like Volo, Renaer, and other characters to establish the plot of the later game. I highly recommend involving your chosen villain in some way, here (spies or visits from the villains themselves are fun), give them an excuse to see a nimblewright, and have Dalakhar show up as well. All that in mind, chapter two should probably take around 5 to 7 sessions total — halfway through, get them to level 3. I think it’s important to have them be done dealing with Frewn by the time chapter three kicks off, and obviously you want them all to be in Trollskull Manor when it begins.
    • Chapter three is where the excitement happens. This is a good time to introduce any new PCs you might have, if you have new players or someone managed to die. Lots of fun characters — read up on the Watchful Order — and a quick investigation. This shouldn’t take too long once they get the nimblewright detector, and characters like the Watchful Order, any of their faction contacts, Vincent Trench, Renaer, and Jarlaxle all serve as good sources of info. I think the investigation shouldn’t take more than a session, two at most, and the infiltration of Gralhund Villa should take one as well. Level up to 4 after this.
    • Chapter four has my favorite part of Dragon Heist — the encounter chains! Note, there are ten locations here, only eight of which are used in each chain. Each chain depends on what villain you’re using and thus what season the adventure takes place in. Each location has a unique description and encounter keyed to the chain it’s in. You can swap stuff around to change things depending on what your players have shown interest in. By this point, maybe they actually found another villain more compelling. Maybe they made friends with others of them. You’re not beholden to the chain, maneuver it around so it makes sense and is the most fun for your group. In general, I found that each chain has only two or three combat-heavy encounters, and as such you can probably knock each chain out in one session at the least, three at the most. It’s up to you if you want to level them to 5 here. If you aren’t adding any extra content, any of the villain lairs, or further investigation, then I’d say get them to level 5 now. If you are adding other stuff, just make sure they’re level 5 before they get to the vault. The vault itself is pretty underwhelming and should definitely only take one session. There’s not even really combat in there.
    • Ultimately we’re looking at 10 sessions at the least, provided your players are receptive to stuff like setting up the tavern and connecting with factions. If not, I could see it being even shorter. On average it’s probably around 15, with no real upper bound other than your players’ patience.
  • Heists and Crawls – Okay, so before I transition into my discussion of Dungeon of the Mad Mage, I want to make one thing very clear. If your party enjoyed the roleplay-heavy, social-encounter-centric, urban romp of Dragon Heist, they (and by extension you) are likely to hate Dungeon of the Mad Mage. It’s absurd that these adventures are considered two halves of the same campaign. Mad Mage is a massive (massive) dungeon crawl, spanning 23 floors with almost no further opportunity for RP. If your group likes that, you likely didn’t spend long with Dragon Heist and can rest assured you’re not going to have much trouble with Mad Mage. On the other hand, if your party enjoyed the social aspects of Dragon Heist, Mad Mage is likely to be a massive disappointment. However — there is hope. My group falls into the latter category, and after about 75 sessions, I’m finished with what I think was a pretty good version of Dungeon of the Mad Mage with some heavy alterations. And I’m here to help you figure out how to make it fun for your group, too.

All that said, this post is already by my estimation about 5,000 words too long, and Dungeon of the Mad Mage is considerably more massive than Dragon Heist, so I’ll be getting to that in another post. I hope this was at least barely readable, and if you made it to the end then I admire your patience and hope at least something was informative or entertaining. Please feel free to let me know if this kind of content is even remotely interesting to you and I may continue with it further on this blog. Additionally I’m sure there’s a ton I left out about my experience with this adventure (since this part of it wrapped up over a year ago), so if you have any questions about any of it, ask away and I’ll respond as immediately as humanly possible.

Until next time.