First, the Aftermath

Chapter Three of Dragon Heist basically ends when Yalah Gralhund’s nimblewright makes off out the window with the Stone of Golorr, but I felt like a little follow-up was worth providing. So, the heroes leave the Gralhund Villa victorious in battle, and presumably without killing any of the family members; the fact that killing a noble carries a death sentence in Waterdeep should probably dissuade them from killing Orond, Yalah, or either of their children. Urstul Floxin and a number of his Zhent minions are probably dead, as well as some of the Gralhund family guards. And, if things went really well, the PC’s were able to meet Orond Gralhund and get some indication from him that Yalah and the Zhents have been plotting together.

So they all head back to the tavern (have a Thanksgiving dinner that can’t be beat) and don’t wake up until the next morning… when they discover that they are in the morning news. Apparently, Yalah has been busy talking with the publishers of the morning broadsheets and spinning her own story of what happened in the villa the previous day. According to her story, the Gralhunds were assaulted in their own home by an invasion of Zhentarim assassins, led by the notorious Urstul Floxin, but were defended by valiant adventurers who showed up in the nick of time to rescue her and her children. Sadly, Floxin and his goons killed many loyal Gralhund guards, as well as Lord Orond Gralhund, who will be sorely missed, so please send flowers.

Of course, the players know that most of this is outright fabrication, because the Zhents were in the house by Yalah’s invitation, and she was in collusion with Urstul Floxin. They might even know that she was responsible for the fireball attack in Trollskull Alley, and they might have figured out that she double-crossed Floxin as well. They also know that Orond Gralhund was alive when they last saw him, and that he wasn’t killed by Floxin because they killed Floxin before he could get to Orond in the first place. If they’re really on top of the situation, they’ll suspect that Yalah had her husband killed to shut him up, and used the “Zhentarim Massacre” as a convenient cover for the murder.

Even if most of these deductions don’t get made, it should be pretty clear to the players at this point that Yalah Gralhund is a liar at the very least, and possibly also a terrorist, murderer, thief, and traitor. That’s the point of the whole “morning news” bit: make sure that everyone knows that Yalah is bad people, because you may need an extra villain at some point, and this is an easy way to create one. Also, with Orond dead, the party’s informant in the Gralhund family won’t be providing any additional information to them, which neatly ties up that loose end.

The City Watch will probably show up later that morning to take the PC’s statements, and the players will have to decide how much of Yalah’s story they’re prepared to support. It’s a tricky question, because her tall tale casts them as heroes and rescuers, even though they weren’t really invited into the villa either. Yalah knows that they can reveal to the authorities that there was no Zhent break-in, and that the Zhents were there by invitation, but she could also tell the authorities that the party was there to aid the Zhents and were driven off with the rest. In other words, her morning news gambit sends the message to the PC’s that she won’t implicate them in the attack as long as they don’t implicate her, but that if they rat her out to the Watch then all bets are off. I really love giving the players difficult decisions, and so should you.

Chapter Four: The Chase

This should be a very familiar statement by now for those of you who have read the previous Dragon Heist articles: you can play Chapter Four according to the book, but I had problems with how the story went, and so I changed the whole chapter significantly. You don’t have to change anything if you don’t want to, but I’m going to go over what I changed and why, because I (of course) think my way is better, for all of the usual this-is-a-detective-story reasons.

The way the chapter is intended to work is rather innovative conceptually, but can somewhat complex to implement, because the way that the material is presented is not at all intuitive. Remember back when you were supposed to choose a main villain, but it didn’t really matter so much until a later time? This is that time. Each of the four villains are assigned to a season of the year, and Chapter Four presents maps and information on ten locations, with encounters and activities for each location that are different depending on who you chose as the main villain. Bear in mind that each encounter chain only uses eight of the ten locations, for some reason.

But, of course, I didn’t do it the way the book says, for the usual reason that I didn’t think it made enough sense to create a coherent detective story. I also decided at the last minute to completely scrap having Jarlaxle as my villain, and to choose a completely different NPC to use instead. Also, that other NPC wasn’t actually one of the four villains, so I felt pretty justified in just rewriting the chase completely. I’m not going to give you all of the details of how I did my chase, but I’ll give some general advice as well as the broad strokes of how I changed things around and why. If you’ve been developing your own story so far, my specific chase items won’t be much good to you anyway.

Homebrew Tip #1: The Seasons

First off, don’t feel locked into a particular season because it matches with the villain you like. At the end of the summaries of each season’s encounter chain, there’s a section on “weather effects”, which gives some mechanical changes based on the commonly encountered weather in Waterdeep during any particular season. Choose your season based on the weather complications you like the best. If you have a party of heavy-armor wearers, summer creates dehydration dangers if they don’t keep drinking water. If you have a lot of PC’s who prefer ranged weapons, you can choose autumn, with winds that blow arrows off target. Or, on the flip side, you can choose weather that isn’t going to be a problem for the party. Or you can ignore the weather entirely. The point here is that choosing the Cassalanters, for example, does not necessarily bind you to using the summer season.

While we’re at it, you can integrate any of the holidays or celebrations from pages 184 to 188 into your story… and it doesn’t really matter what time of the year you chose for your adventure. If something seems like it would make things interesting, put it in there. For example, Simril is a holiday with lots of folks staying up all night feasting and stargazing, and maybe you want streets full of revelers to make things a little more complicated for chases or moving around. According to page 188, it’s a winter holiday, but that makes no difference; if you want summer weather as well as full streets at night-time, you can have both. Sure, maybe you have a player who really studies Waterdeep lore, and would know you made a twist with the calendar… but if you have a player like that who would actually choose to make a fuss about it, you probably have internal group problems that are beyond the scope of this article to address.

Homebrew Tip #2: Things to Avoid

First, I advise against allowing your main villain to get hold of the Stone of Golorr, even though there’s a sidebar on page 60 that talks about how to handle that. Unfortunately, the two suggestions on how to handle it both involve a raid into the main villain’s stronghold (as described in Chapters 5 through 8), either to steal the Stone back, or to steal the gold if you determine by rolling dice that the villain has had enough time to recover it. Surviving such a raid is highly uncertain; the information on villain lairs on page 10 specifies that the lairs are not really meant for characters below fifth level, which is problematic considering that the PC’s don’t reach that level milestone until they complete Chapter Four, which of course they haven’t if they’re still trying to acquire the Stone. Also, incursions into a villain’s lair are unlikely to be accomplished quickly or easily when you’re looking high and low in uncertain territory for a small object like the Stone. Stealing the treasure, which weighs about ten thousand pounds, is such a ridiculous plan that I don’t know why they even suggested it as a possible course of action. It’s unlikely that the main villain will get the Stone personally, but there are opportunities (like the chase scenes discussed below) for the villain’s henchmen to seize the Stone and escape with it.

Second, I advise against letting your party get hold of the Stone of Golorr, at least the genuine one, until just before it’s time for them to use it to find the Vault. There’s a sidebar on page 58 that talks about how to arbitrarily change the requirements to use the Stone based on whether they’ve “earned it” or not. The sidebar basically tells you to arbitrarily make the Stone unresponsive and uninformative, until the party reaches the point in the story when they are supposed to obtain it, at which point it will begin to work. How exactly the players will figure out that they need to try again to use an object which didn’t work previously, just because maybe it might be different now, is a question that really needs to be answered. What it comes down to here is that you’ll have to come up with a good reason for the Stone not to work if they get it early, or else just make sure they don’t get the opportunity to get it early in the first place.

Third, don’t let the second nimblewright search take too much time or trouble. The party has already searched Waterdeep once looking for Yalah’s nimblewright, and arrived just in time to not catch it at the Gralhund Villa. And, frankly, they never had a chance of catching it, because Chapter Three pretty much has to end with the nimblewright getting away with the Stone of Golorr, because otherwise there would be no need for the chase portion of Chapter Four. Now the PC’s have to search for the nimblewright and find it again, and their reaction to this will range from mild annoyance to outright irritation, depending on how much trouble they had finding the nimblewright the first time. They will never be pleased to hear they need to repeat their performance, and that’s a guarantee. Squandering your players’ goodwill on this task is not worth it; just let them find it in a few minutes of table time, even if you have to narrate “you spend the next five days searching the streets of Waterdeep” or somesuch.

Homebrew Tip #3: Chases Are Tricky

There are two chases available in the chapter, one through the streets and one across rooftops. Chases can be fun, but they also put the story in fairly serious jeopardy, and these chases are no exception. You don’t have a motivation for a chase unless you put the Stone of Golorr in the hands of the villain’s henchmen, and there’s no point in having a chase in the first place if there’s no possibility that the badguys might escape with it.

There are two ways for the chase to derail the story, and they also play off one another to create a sort of hybrid derailment. If the badguys get away, which might happen based on bad luck for characters or bad decisions by players, then the Stone ends up in the hands of the villain, which we’ve already mentioned is something you really want to avoid. It’s also possible that the badguys reach the endpoint of the chase and then turn and fight the party, and the party loses, and the Stone ends up in the hands of the villain. The hybrid problem is when only some of the PC’s manage to keep up (maybe one of them falls off a roof and is left behind), and arrive at the end of the chase unable to hold their own in a fight.

I’ll also comment that if you plan on using the encounter chains as presented in the book, the chases have to meet some fairly strict requirements for success. They essentially need to proceed from a fixed starting point to a fixed endpoint, and in the middle the party needs to keep up with the badguys in order to arrive at the endpoint, but also needs not to actually catch them and stop them from reaching the endpoint. In other words, the party needs to neither succeed nor fail in the chase, but have a sort of partial success in order for the encounter chain to not be broken.

Ask yourself if you really trust WotC’s designers to come up with a chase so perfectly balanced that you can be sure that the party will have just the right level of success and not find themselves in a street or on a rooftop wondering what to do next. It’s kind of a big chance to take, and it’s obviously your decision as the DM, but please don’t walk into this one blindly and just run it as written without considering the risks for the continuation of the story.

Homebrew Tip #4: Choose Better Keys

Okay, let’s get this out of the way: the “keys” needed to open the Vault of Dragons are stupid. Really, really stupid. We’re talking about items needed to open a very imposing magical door that leads into a very serious ancient dwarven temple, and drunken elves and music-hall performances are possible choices. It’s idiotic, and I’m using that word with all of its implications, including the one that says that whomever came up with the possible vault keys is an idiot, or was at the very least having a moment of staggering idiocy and then never looking back soberly and realizing the full stupidity of that moment’s decisions.

And yes, that includes Perkins, Crawford, and even Matt Mercer (All Praise To His Holy Name). In fact, it might include all of them, because someone among the authors of this adventure should have stood up and said “are they seriously going to need a shaved dwarf to open this thing?” and avoided the whole ridiculous situation.

You can invent your own keys. You should invent your own keys. The only ones worth anything on the list in the book are the warhammer and the sunlight. The warhammer at least has some significance for a dwarven temple, and figuring out how to get sunlight onto an underground door offers some opportunities for clever problem solving.

I decided that the keys to open the vault would take the form of three weapons to correspond with the three dwarven deities represented in the vault: a gold warhammer for Moradin, an adamantine battleaxe for Gorm Gulthyn, and a dark oak greatclub for Dumathoin. Although I just made them up, a Religion check would reveal that these were weapons known to be symbolic and sacred to those deities. Furthermore, they could be procured in Waterdeep, perhaps with the help of the smiths and the woodworker who have shops near the PC’s tavern.

The fact is that you don’t need three keys, of any sort, to open the door. Maybe it opens with a riddle. Maybe it opens with an actual, non-metaphorical key. Maybe you can just knock politely and then push it open. It doesn’t matter how it opens… but please, don’t make unicorns have anything to do with getting into an ancient dwarven temple.

Homebrew Tip #5: Pick a Frienemy

I mentioned earlier that I had originally intended to use Jarlaxle as my main villain, but that I decided not to use him after all, and in fact decided not to use any of the main villains from the list. My main villain turned out to be Vincent Trench, the neighbor with the detective agency, and that decision was made very late in the adventure. I still think Trench was a better choice than any of the official main villains, especially for my players, because defeat combined with betrayal is much worse (and paradoxically more satisfying) than just plain defeat.

How it happened was that I was going through the autumn encounter chain, which involves Jarlaxle impersonating various Waterdavian citizens to misdirect the party into believing that they’re working on the side of good, when in fact they’re doing Jarlaxle’s dirty work for him. Eventually, when they’ve found the Vault of Dragons, Jarlaxle is supposed to show up and reveal himself as the deceiver, before going after the treasure himself. I was writing out a little script for myself to work from, having Jarlaxle shift appearances to show the players that the different NPC’s had been him all along, making snide remarks all the while, and then finally changing into his normal drow appearance before insulting them once more and sending them on their way emptyhanded.

Then I realized that Jarlaxle meant nothing to them, because they had never met him or interacted with him in any way: he was just a villain who swooped in at the very end to say “hi, you don’t know me, but I’ve been your enemy all along.” That seemed very unsatisfying, and I realized that almost anyone from the entire story could be the villain, as long as they had Jarlaxle’s same ability to disguise themselves, which can be provided just by giving the villain a hat of disguise like Jarlaxle’s. That’s when I decided that Vincent Trench would be the best villain, because they had been talking with him and trading information all through the previous parts of the story, and now he shows up to sarcastically thank them for all of their help and shoo them away like wayward children.

That’s the tip here: choose someone the PC’s know already, and then make that NPC into the villain. Choose Yalah Gralhund, or Valetta from the Temple of Gond. Choose gentle Fala the apothecary neighbor, who we empathize with because of the way people discriminate against those without a binary gender. Hell, choose the halflings who were killed in the fireball attack and are now demonic zombie buskers, but choose someone that the party and the players already know. Beef up their stat blocks, or have them demonstrate powers that the PC’s never suspected. In the case of Vincent Trench, I actually had to make him into a less powerful enemy than his original version, because a rakshasa would have been too much to handle at fourth level. Basically, create formidable enemies, but create them with familiar faces.

Still Chapter Four: The Vault

For some reason, the Vault of Dragons itself doesn’t get its own chapter; they just sort of tacked it onto the end of the chase. I suppose it doesn’t really matter how they organize the book, but I would definitely recommend that you create some punctuation when you move from the chase part of the chapter to the dungeon-crawl part. Make sure that everyone realizes that the chase is over, the vault entrance has been found, and we are now moving to a completely different adventure style than we were just using. With my Neverember-specific Stone of Golorr, setting up a meeting with Renaer Neverember to get the Stone’s information was a good way to bring the chase to a full stop and transition into the much more dungeony Vault of Dragons.

The Vault itself is fairly unremarkable. There are some traps, false leads, and difficult terrain, but nothing that I think needs to be really discussed. I will point out that it should be possible for characters to climb the support posts for the overhead bridges if they want to get to the far side of the crumbled bridges safely, but that’s the only part of the setup of the vault that I found less than clear.

What matters is what happens in the Vault, after the characters find their way down to where the treasure hoard is piled up. I’ve already gone into the problems with this situation in a previous Dragon Heist article, but I’ll give an overview here so we can get into how to handle them.

The biggest problem is that the treasure is guarded by a full-grown dragon, an enemy that is far and away beyond the capabilities of a party of fourth-level characters to deal with. I’m including non-combat options here too, because Legendary Resistance is going to make spells to deceive or charm the dragon ineffective, and the dragon’s high wisdom and charisma will make it very unlikely that the PC’s will somehow convince the dragon to just hand over the treasure and walk away.

Nevermind that the treasure itself weighs at least ten thousand pounds: that’s half a million coins, at the usual rate of 50 coins to the pound. Even if you go with the extremely generous carrying capacity offered by the PHB, you’re still talking about a couple dozen or so trips per party member to carry it all out, and that’s ignoring the serious inconvenience of carrying it across crumbling bridges or lowering it down on ropes. I decided not to wonder too much about how Dagult Neverember got it all down there in the first place, actually, because presumably he would have had the same problems in reverse.

The solution that seems to be implied (not really spelled out) by the hardcover is that powerful NPC’s will become involved: some on the side of the PC’s and some on the side of the main villain. How these NPC’s will become aware of the location of the secret vault, and how they will know that their help is needed there, is not very clear. The relationship between these friendly NPC’s and the dragon guarding the treasure is rather vague as well. However this plays out, the intention seems to be that the intervention of friendly NPC’s will prevent the party being killed by the enemy NPC’s, and that the friendly NPC’s will also help them either defeat the dragon or convince him to hand over the treasure. Of course, maybe everyone will just be killed off in some kind of battle royale. Or maybe the PC’s will be the ones who get killed in said battle, in which case it might not matter much which NPC’s are eventually triumphant.

My players suggested that if I had decided to bring in NPC’s to save the day, a nice thing would be for me to bring them some popcorn to eat while they watched me battle enemy NPC’s against friendly NPC’s, who were there to solve the party’s problems for them.

Don’t worry, though, because there’s a better way to do this. It’s been lurking in the story all along, but for some reason it hasn’t been called out as probably the most valuable thing in the Vault of Dragons. More valuable even than half a million gold pieces, even. The legendary Dragonstaff of Ahghairon is down there, and that’s what the treasure hunt needs to be about.

The Dragonstaff is the real treasure here, because whoever has it can blackmail the City of Waterdeep with the threat of letting a horde of evil dragons through the ward and letting them wreak havoc. The lords of the city will pay whatever they have to, in money or power, in order to prevent that. And let’s not forget that many evil dragons are quite intelligent and very rich as well, and that anyone with the Dragonstaff can offer such dragons the opportunity to ravage and plunder Waterdeep for fun and profit… for a price. The possibilities for an unscrupulous individual to gain wealth and power are immense.

So… forget the money. I actually decided that the coins would be gold-plated lead meant to be a distraction. Really, would someone clever enough to embezzle that kind of money from a major city government actually be foolish enough to essentially hide it under a mattress? But, when people “know” that a legendary treasure is buried in a secret vault deep under the city streets, they tend not to look for that treasure in more reasonable locations, like banks and brokerages. As far as I was concerned, most of Dagult’s ill-gotten gains had probably already been laundered through reputable institutions and moved out of the city.

That made things a little easier, because now there was no reason for the party to have to remove the treasure from the vault. In fact, because the treasure isn’t really worth anything, the DM doesn’t have to figure out how to not give the PC’s insane amounts of gold to spend on whatever they want. If you’re planning on continuing into Dungeon of the Mad Mage, which is intended as a follow-on to Dragon Heist, too much spending money for the characters is a big problem: either they have so much money that they can buy far more high-end equipment and magic items than 5th-level characters should have, or else their already vast wealth makes the opportunities to profit through exploring Undermountain less than compelling. If you go with the numbers given in the adventure conclusion on page 98, the party ends up with fifty thousand GP, and assuming that they pay all of the various expenses suggested (many of which are optional), they still have more than half of that left over. If the treasure is fake, a shill, then there’s no money for them to claim.

The other thing that is made easier when we focus on the Dragonstaff instead of the money is that now the villain is interested in stealing something relatively small and portable, instead of a giant pile of metal. The Dragonstaff is something that you can carry around. In fact, it’s something that you can grab and run away with, requiring someone to chase you down in order to recover it.

So now we have a way to resolve the Vault of Dragons in a suitably heroic way. It works something like this: the villain shows up, insults the PC’s, and heads into the vault to claim the treasure, perhaps with a threat that the PC’s are next on the list to die. A battle between the dragon and the villain ensues, and this battle can be fairly evenly matched, resulting in a stalemate in which both the villain and the dragon are seriously injured. Maybe the party gets involved in this battle, but whether they do or not, the villain grabs the Dragonstaff and makes a run for it. The dragon, too injured to give chase, implores the heroes to stop the villain and recover the Dragonstaff… no time to explain why, just go, and all will be explained later. This is a lawful good gold dragon, and can presumably be trusted, so off they go, trusting that getting the staff back is really as important as the dragon says.

You can work this pursuit however you like; a short chase might end in just another part of the vault, but you could also have a longer chase that moves out into the city streets. Either way, the consequences of the battle with the dragon are that the villain is considerably weakened, and thus possible (but not necessarily easy) for the party to defeat even at their current level. With the Dragonstaff recovered, the dragon can now leave the vault, and shows up to the PC’s tavern to clear up the remaining mystery, and make it clear just why it was so important to chase down and recover the Dragonstaff. His advice is to turn over the Dragonstaff to the Open Lord, no strings attached, and trust that she will reward them fairly for their heroism.

I advise against giving the party any portion of the actual treasure, even if you decide that it’s a real treasure and not a fake. If you want to provide gold, a good solution is to give them the remainder of the gems that were in the vault for the dragon to eat; now that he’s leaving the vault, he won’t need them anymore, and they make a fitting gift to the heroes who have made it unnecessary for him to remain cooped up in an underground temple.

The Open Lord can provide rewards other than currency: magic items for use in Undermountain, land grants with noble titles, or medals and letters of commendation that provide special treatment in Waterdeep. In other words, she provides impressive rewards, but not in cash. One of the rewards I used was ensuring the success of the party’s tavern by providing government contracts and tax exemptions and such, which had the added benefit of not having to actually manage the tavern while adventuring in Undermountain.

Exeunt Omnes

And that, dear DM’s, brings the Dragon Heist adventure to a close. The upshot of this entire series of articles is that you can run the adventure more or less as it’s presented in the book, but that you might want or need to change certain things in order to make the mystery-story aspect work adequately.

My group of players needed to have a coherent story, because figuring out the backstory and understanding the motivations of the various people involved in the situation was something that fit with their playstyle inclinations. To make it work for them, I had to make some significant changes across the board, but you may not need to make extensive alterations to the adventure as written; every group has different priorities, and will find different types of gameplay to be the most fun and rewarding. Figure out what your players want, and go for it.

One of the nice things about the progression of the adventure is that you don’t have to immediately figure out what you want to change and how you want to change it. I mostly decided what I wanted to fix before running each chapter, and I definitely changed my mind on a number of things as the story progressed. As mentioned, I even decided completely at the last minute who the ultimate villain would be, so you shouldn’t have excessive difficulties changing what you need to change.

So, to recap my assessment of Dragon Heist: if you’re looking to use it as the urban-mystery-adventure campaign setting, and you’re planning on taking what’s in the hardcover and rewriting and revising it, and making it your own (while getting the benefit of maps, and NPC’s with names and stats, and a local setting with flavor and lore) I give it four and a half stars. If you want to run it straight out of the book as written, I give it two and a half stars. It’s not that it’s a bad adventure on its own, but there’s a lot of potential that goes untapped here.

Final verdict? If you want a short adventure with a different dynamic than the usual dungeon crawl, MacGuffin hunt, or daring rescue, this is a good one. If you want an adventure to lead into Dungeon of the Mad Mage, this is a good one because it also takes place in Waterdeep and gives some character continuity. If you’re looking for adventures for low-level characters, and you don’t have any stake in Waterdeep or a desire for something other than a more classic dungeon setting, go with something from Tales from the Yawning Portal instead.

But, if you’re going to run Dragon Heist, you now have the best advice I can offer to make it work, whether you’re sticking with the original as written or homebrewing your own content on top of what’s already there. Have fun with it, and as always, remember not to let hardcover authors and designers push you around… you’re the DM, after all, so be bold and creative and run your best game.