You are here

Flames and death in Saltmarsh

The way in which we are playing the game: .

Context: Player characters were chasing some bandits, but lost track of them close to the town of Saltmarsh (from a TSR adventure, U1 The sinister secret of Saltmarsh) and the fort of Morgansfort (from a Basic fantasy adventure of the same name at ). There is some dungeoncrawling in the Old island fortress dungeon near Morgansfort, some characters get to know the local illegal drinking place and the local baronet. One character has criminal connections and finds a weapon smuggler in Saltmarsh.

The Saltmarsh adventure assumes that characters march in for whatever reason, hear about a haunted mansion, investigate, find a smuggler hideout under it, defeat some smugglers, capture their ship, yada yada. It has two follow-up adventures and the series does include quite interesting challenges, which are nullified by the game master advice to have the players succeed and follow the script regardless.

The weapon smuggler? An influential merchant and also together with the smugglers under the haunted mansion. The adventure specifies nothing about this merchant or their purposes, so I determine they are in it not only for the money, but also because the smugglers have the merchant's daughter captured and taken; in truth the daughter is a pirate queen of sorts, but it is left ambivalent if how parents actually suspect this and willingly blind themselves to justify the smuggling, or if they are unknowing.

Anyway, one player decides that instead of going into a dungeon, they'll reveal this smuggler to the baronet and get rewarded for it. A good move in general; avoid the uncertain and dangerous dungeoneering and do some social engineering and intimidation instead, and maybe even do some good on the side. The baronet does offer a non-trivial sum of reward money (and thereby experience) for it. But the baronet does not want to go barging on the home of a wealthy merchant and member of town council without good grounds, so he musters his troops with an excuse and player characters are sent in first, so that if they find nothing, the baronet and the troops will simply go hunting for non-existing hostiles in the swamps as per the excuse. Plausible deniability for the baronet.

The plannin process was so that the baronet simply states that he can't just take a bunch of soldiers and march into the home of a council member; what are we going to do about it? And then the players brainstormed until their plan was sufficient for the baronet to accept, given the risks and rewards. As a referee my role was to tell what the players know about the present situation and what the baronet can tell them; had the baronet been more of a schemer, he might have actively suggested something, too.

Characters go to the smuggler-merchant's house, set up watch (one watchman almost catches a certain assassin from the haunted house; I add active NPCs to the local random encounter tables for exactly these kinds of reasons), break in later and the forcible interrogate the merchant, who under duress agrees to co-operate, given that they let the baronet know of this and promise to help her daughter. He lets the characters know the smugglers are somewhere under the mansion, yet he does not know any details.

Here we see the particulars of the rules chassis we are using (D&D 5); one of the characters is an aarakocra and can fly, while another is an eladrin and can teleport a bit. Flying around revelas the sea entrance in the smugglers' hideout. Three player characters go that way. They have enough rope to climb to within 20' of the sea surface (I first scan the adventure or decide how tall the cliff is and afterwards we check how much rope the charactes have). The wind is rising, but they decide to go in anyway; a surprise attack from rear, what could go wrong? The thing that goes wrong is that there are lots of enemies there and the characters have very hard time escaping, especially after they decide to burn the smugglers' boat so that the smugglers can't get away. Some combat later, the eladrin is the only one to escape by swimming against the wind and the teleporting to the rope, while the others are imprisoned and interrogated.

Meanwhile for the mansion itself; there is investigation, some treasures are found and others missed, some stirges are killed, the rotten floor collapses under someone, and so on. They discover one way down (stairs to a cellar next to the kitchen), but miss a hidden trapdoor that is the other entrance. The assassin is not met; the adventure assumes the assassin is found as a captive and taken along only to betray the characters later, but no such luck here as the smugglers did not have the time to set the assassin up in the manor. The baronet's force arrives. The strike force (couple of player characters and couple NPC knights and some soldiers descend into the cellar, avoid and burn some rot grubs, make noise, alert the smugglers and eventually find the secret door to their hideout.

The smuggler hideout has two viable exists (their boat gone), both from the same big room. They have prepared for confrontation and for escaping with some of their loot ready at the exit, too. The captives are there, knives at their throats, ready to be used for negotiation. Some knights are the first to charge in and quickly deal with the frontline of the smugglers (gnolls, I think). The player of the aarakocra tells that they will try to burst their bounds and shake of the smuggler holding them. Okay, versus test (probably athletics or even initiative; it has been a while); you win and you throw them off, rising to standing position; you lose and they slit your throat, so you are dying; and you do have disadvantage due to being bound and all and the smuggler is explicitly holding you there anyway. The dice are rolled (in the open, of course, since this is an important moment) and a well-loved player character is dying. Some death saves later they are dead.

Meanwhile the fight continues. The smugglers do not do well, so I check morale (a wisdom save; technically not a part of D&D 5, but our point is not so much to play D&D 5, and people being afraid of their lives is something that clearly adds to the game and the tactical and strategic depth). The smugglers start running, the other captive is thrown hard against the ground (the smugglers have not reason to kill them, after all; luckily it was not a gnoll holding them). When the player characters figure this out, there is a moment of confusion until some run back up through the kitchen and others follow the other entryway. The smuggler boss, an illusionist (I use their spells from AD&D; no reason for them to follow the same rules as player characters), manages to escape and is followed by one the eladrin, who much later finds out that a single higher level illusionist with thrown daggers and good dexterity is a deadly opponent to a first level fighter, but the later the illusionist comes to find out that being alone in the woods makes an urban snob easy prey for determined hunters.

Most of the smugglers are caught. There is lots of loot in the dungeon, so after some discussion word is sent to Saltmarsh that people will get paid if they haul out the loot; the assassin is one of the volunteers. While waiting for them, the player characters explore the underground. Treasure that belongs to the baronet, some green slime that surprisingly does not corrode one armour (it happened to fall exactly on the character wearing a full plate the discovered earlier which later turns out to be magical; very good luck there, it could easily have been deadly otherwise) and one dead-end room with skeletons. After killing the skeletons the players are really puzzled about it all; no necromancy in the adventure and then a dead-end room nailed shut with armed skeletons? They don't think of searching for secret doors, so the most interesting treasures of the adventure are missed. Such is life.

The assassin makes a final brave attempt to scare away the folk by capturing a random hireling, killing them, cutting their throat and throwing them down on the manor lobby. The player characters are too clever for this and start chasing for the assassin, who does manage to avoid them for a while with the home advantage. There are dry bushes on the outside of the manor and a missed fire spell hits them (missed attacks with any effect whatsoever is not part of D&D 5 rules as written, but we certainly take them into account when appropriate) and the characters choose to burn down the garden, inspired by this. Windy weather and uncontrolled flames, in spite of some noble attempts, finally devastate the manor. The assassin escapes.

Later: Some goblins move in in the abandoned caves (restocking the dungeon). The player characters have a choise of waiting substantial times for the baronet to summon their reward in gold, the now-burned plot of land where the manor stood, or the rights to an abandoned fortress with a productive mine in the nearby mountains; they choose the latter, which turns out to be full of orcs. The assassin kills the smuggler-merchants but does not get their hands on the smuggler leader, who is imprisoned. The smugglers are building a wall towards a swamp area infested with lizardfolk. What the pirates doing the smuggling, the lizardfolk or the sahuagin, part of the follow-up adventures to this, are about has not been seen in play or mostly determined yet, though some signs are in the air. The player characters are currently active elsewhere.

The bird-man, as the aarakocra was known, was an estalished and well-like character. Their death was earned. The eladrin was a new character and well-played; they simply did not know what they were up against and, even then, it was a close call, but such things happen.

The scenario that we played here came out from an interaction of two adventures, one a typical starting town plus some caves, the other a more intricate adventure with lots of potential. The player choises in approaching the entire situation were crucial to creating the scenario.

The ruleset, D&D 5, is mediocre. It is somewhat cumbersome but not awfully so and does give the tools to resolve most adventurous situations. The characters are brimming with boring magic, which I find aesthetically unpleasing, but it does not really affect how fun playing is. The most bothersome aspect is that it neuters the consequences of many actions. Magic coming from demons, Cthulhu, or dark gods has no price. Shooting into melee is trivial. Resting we have house ruled so that not everyone heals completely with a night of sleep. We pay more attention to encumbrance and carrying then the (ridiculous) default assumptions of a random man on the street carrying 70 kg of stuff around while marching 50 kilometers per day and fighting some skirmish combats, all without getting particularly tired. These house rules were not particularly relevant during these sessions.

Actual Play


badspeler's picture

Hey Tommi,

Interested in this style of play with DnD 5, and have a whole host of questions. Don't feel pressured to answer all of them; any help would be great.

(1) The baronet seems like an important character. Were they introduced from a module or were they your design? How did you instantiate their motivations "in the middle of play"? In particular how did you instantiate their resources of wealth and soldiers? Or was all of this part of one of the modules?

(2) How do you manage "fleeing"? It seems like you improvised something for the smugglers to decide to flee, but once they started fleeing, how did you handle it? When situations like this come up, I feel like I have to be a designer in the middle of a tense combat, force to choose between something "unfair" or "pulling my punches." Especially frustrating because I just can't imagine fleeing not coming up in playtesting.

(3) Did missing out on the secret doors impair play at all? It seems like it didn't come up in play because it wasn't a natural extension of what you guys were doing, so I wouldn't have missed it at all. Like you said, skeletons have little/nothing to do with anything established.

(4) Can you say more about "missed attacks"? In my experience, what should be the interesting emergent play stuff comes from ad-hoc calls/rules like this. I would like to instantiate many of these "calls" or "house-rules" as formal rules in my game.

(5) If you would indulge me more, could you talk a little about the process of "combining modules" like this? It reminds me of Raymond Chandler's strategy writing his novels. If I'm doing a west marches style hexcrawl, part of the idea is to "stick modules" into various hexes but I'm sort of lost for how they actually interact together.

(6) You mention several times about the destruction of locations (burning) or falling through floors. Was this instantiated in rules somehow or were they ad-hoc calls each time? I really enjoy the ability to navigate environments in clever ways like in video games like Thief and Nethack.

(7) You mention using "different rules" for the NPCs that came from non-5th edition modules. Can you explain this? I'm at a loss for converting an entire module to a completely new system.

Thanks for your post. Hoping to learn as much as I can.



Sean_RDP's picture

Eric, I have some comments on #7 but want to wait until Tommi gives his answer. 

Hey Eric,

(1) The Morgansfort adventure module happens to have stats and lots of detail on the town; it is free (when it comes to price, and I guess OGL when it comes to licensing), so you can easily find it on the linked website. When such sources are not available, or they do not provide what I want, I might use a random personality generator and just guesstimate how much resources they would have available based on what I know, maybe rolling a d6 with low results meaning I adjust downwards and high upwards or some similar ad hoc methodology. The key principle being to avoid making decisions that have a clear or immidiate effect on player character success, but rather always disclaiming them by dice or preparation.

(2) To determine that they flee: Since their situation was bad, with knights defeating their elite (and feared) gnollish front ranks, I figured there is good chance that they would be afraid; hence a morale roll. I typically use a wisdom save, but it might also be a charisma roll of a leader or whatever seems to be the key issue at hand. Since they had prepared for escape, I would have made them panic and run all over the place only on a very bad result; maybe 3 or less.

Since these guys are smugglers, or at worst pirate-smugglers, they don't want to die and their threshold for running or surrendering is quite low. There is no fanatism here. I don't remember what I did then, but now I might set the DC to stand and fight to be 15 or maybe even a bit more.

As for the mechanics of escape, as soon as we are no longer in a situation where turn-based actions make sense and decipt the situation in a useful way, I stop using them and instead ask: the smugglers are running up this stairway, arms full of stuff. the first one opens a hatch in the ceiling, while the last one is just getting onto the stairs. What do you do? And then play as usual: figure out difficulties and consequences, roll dice to select among them, etc.

(3) The players did not know they missed something. I felt a small bit of disappointment, since it was something there, but such is life. Maybe someone else wil come there later. So no impairment in play.

(4) I go by ad-hoc rulings. I would guess that my process is to imagine what is going on in the fiction and what are the likely effects and side-effects of it. Are you in tight quarters? Disadvantage with long weapons, aside from thrusting ones, because good luck using them there. Fire and flammable environment? Think of the consequences. On a rickety rope bridge? Being hit with a heavy weapon demands a saving throw to not fall.

I try to always be explicit about these beforehand so players can suggest stuff, cry foul, and so taht we secure we are on the same page about the circumstances. Codification of these leads to publishing these huge tomes full of tables of modifiers and special cases. I much prefer the organic method of ad hoc rulings, accepted on a group level, which also leads me to learn new things about for example climbing, how to make oil from animal fat, and whatever the players happen to know more about than I do.

(5) Can you ask a more specific question? I don't really combine modules as much as slot them into the game world. Then I assume nearby settlements typically have some knowledge of and relationships with each other (add to rumour table: rumour from the next town over) and things move around (add to random encounter tables: npcs from adventures, rolls from their encounter tables). The world is a living place and the adventures provide the starting situation.

(6) The TSR adventure specified certain weak parts of floors. It is a pretty decent adventure when one ignores the railroading. Burning stuff is a common problem solving method among players, but here it was the consequence of risk-taking and fire magic, so ad-hoc rulings and figuring out what the situation is like.

(7) The sensible way to do this is that if there is a gnoll, you use D&D 5 stats for a gnoll instead. What I do is adjust those based on the source material; hit dice and armour class I adjust, for example, based on what is says on the module. I can do this ahead of time or on the fly. But it is much easier to just replace everything with the equivalent in whatever game you are using. If there is no equivalent or it is very different, make your own.

I just happen to enjoy the chaos of having material that breaks the assumption of the game I am using; magic and monsters are not stuff that is codified and in the rulebook; these walking corpses might resist cold while those others might have more hit dice, and these others might resist weapon attacks while those are immune to non-magical weapons. But this is a messy creative choice on my part.

Ron Edwards's picture

So much to discuss! Some of what I have in mind parallels Eric’s comment above, but I’m starting this as a new comment in order not to gum up his priorities there.

I’ll start with the concept of “going off-script” in relation to published text, which I think of as actually playing in the first place. Strong words, right?

I hope it makes sense that I do not actually think the publication is the key feature; the key feature that I’m contemptuously dismissing (“not real play!”) is the notion of a planned plot to be delivered from one person to the players and, if necessary, protected from them too. By planned, I’m referring not to the entry of troublesome or challenging components, but to how they turn out, especially as conceived in a sequence.

This is why I am so intense lately about the difference between situation-authority and outcome-authority. Situational authority may be centralized extremely tightly and include considerable preparation before play without violating outcome authority, e.g., when the latter includes extensive procedural constraints (mechanics). A great deal of current rhetoric falls flat on its face by misunderstanding this point, and by confounding group-distributed, improvised situational authority with some kind of benefit to outcome authority. This what Workshop: The plot thickens is all about, as well as my discussion of what I’m calling Bounce.

Anyway, to apply this gibberish to your account of play, I’m talking about how the players’ perceptions, priorities, and approach interacted with your own commitment to play the NPCs as interested, responsive characters in their own right, as well as adding more NPCs to arbitrary components. Therefore the entire play-group treats whatever they “see” in play as an ongoing starting point for “how to handle this” and “what’s going on next” rather than points “to get to.”

In terms of content, this matters too: it’s not as if there are several nodes of carefully-planned combat set-pieces, to which you must nudge and guide everyone to get to. Instead, you’ve treated the location as a politically and economically active place, caught in its present moment partly due to the dynamic situation among the members of a particular family. All this content is not merely a “skin” for the encounters, like the backdrop in a combat video game, but instead a completely solid aesthetic and human basis for playing the NPCs.

Therefore even if we do crudely define “the scenario” as who fights whom and where (and why not? The purpose of play includes risk-and-reward, which is perfectly understandable), and, its sequentially resulting and resolving events (a.k.a. “plot”), are literally unwritten, unplanned. As long as the people enjoying it play it, then such a resulting/resolving sequence is certain to arise somehow.

I’d like to repeat and stress here that nothing about this is a “style.” It is the minimum necessary actually to use the medium we are talking about, regardless of what purpose is involved or any specific procedural detail. Are the outcomes of play subject to the procedural determinants of play, or aren’t they? Period.

Going all the way back to Monday Lab: Kittybox, that Lab resulted in some frustrating fallout. I perceive a very tightly-held mythology in role-playing culture, especially amon those who have learned to do what I’m talking about, yet cannot free themselves from romantic ideals about the published sources they’re using. It’s the error of believing that one did this (plot-results-from-play) as if the text “allowed it,” when the fact is that one ignored the text in order to do it.

It’s a notable, even frightening gymnastic mental trick: to say, “well, the authors played it emergently, but they had to write it as a script or no one would understand it, but since we understand that, we dial it back to its potential rather than the script, so we really play via this tacit understanding between the author and ourselves, rather than the sort of false text that happens to be on the pages, so if you criticize what’s actually on the pages, then clearly you just don’t understand.”

Whew ... and more to go ... as a subset topic, I’d love to address the technique of combining materials, or more generally, to have “situation A” embedded in or connected to “situation B.” As far as using published materials for this (not necessary but a useful focusing point for discussion) is concerned, my experience in this regard is deep and detailed, mainly in playing Champions. I think I’ll follow up about it in Eric’s comment.

Hej Ron,

Thanks for the comment.

I agree with respect to going off-script with respect this way of playing or philosophy of play or whatever you want to call it (you seem to reserve purpose of play to a larger category; I am including a bunch of specific techniques here and narrowing the purpose from general competition to this more specific combination strategy, tactics, fiction-based problem solving etc.. but ignoring for example sophisticated mastery of a formal written system or effective cajoling of other participants). I am not certain it extends to all possible functional ways of playing roleplaying games, depending on what precisely you mean.

Agree concerning outcome authority etc, though I did not manage to listen to the entire workshop and have no idea what you mean by bounce. But I think we are strongly on the same page here, terminology aside.

It’s the error of believing that one did this (plot-results-from-play) as if the text “allowed it,” when the fact is that one ignored the text in order to do it.

Here I am not sure I quite agree. With the TSR adventure I did explicitly and with will ignore the railroads. The rest I use as written as a base situation. The text both allowed this particular adventure to happen and parts of it were utterly ignored.

But many adventures are also like Morgansfort: the adventure is written as a snapshot of a location, often (but not always) a static one. Is it ignoring the adventure to treat that snapshot as a basis for a dynamic situation, where the material is embedded into a living world and different entities start to act?

Anyways, I do say that not taking these adventures as a basis for a dynamic situation is (generally speaking) using them in a sub-optimal way, regardless of how they are written or what the authors intended. I should say that my interest is more in how a particular game can be used for a particular purpose, rather than what a particular game claims it is trying to do. Substitute "game" with "adventure" or whatever.


Ron Edwards's picture

My mistake in communication: I did not make it clear that my criticism, the error I described, explicitly did not occur with you, and that your session (and general sense of play across these games) is living proof against the error. I was taking the opportunity to complain bitterly about other people who are not you.

LorenzoC's picture

Hey Tommi, there's a couple things that caught my attention reading your post, and I'd love to hear more about it.

At some point you mention two prisoners being held with daggers at their throat. I'm wild guessing you handled this with an ad-hoc decision, because if I recall correctly, there's no "coup de grace" rules in 5E. The game assumes you'll struggle and fight and get lucky and whatnot, but people simply get critical hits on you for free - they have to go through your entire HP pool.
Was it like this? Did you feel like it was necessary to be able to say "do something or die" to make the experience work, in terms of mood or drama or simple delivery, even if that meant blowing past rules?

As for players missing the secret door where the skeletons were, as sad as it for the players it's definitely something "working as intended" for me. Anything that may even slightly lead on the players ("the skeletons immediately turn to you, and form a shield wall with their backs on the southern wall") is going to be too much, and remove the value of exploration itself. But it would have been interesting to see a scenario develop where the players suspect there may be a secret door, roll to find it and obtain a mediocre result. In my experience that is the tabletop equivalent of a "blue screen of doom".

Lastly, concerning the use of published modules, I've learned to be extremely unfraid of using them as the equivalent of DM preparation. In the Pathfinder game I'm discussing elsewhere, we used the first officially published adventure (the Fall of Plaguestone) as the stepping stone for our adventures. That adventure begins with the players arriving in town at the hells of a merchant caravan; their benefactor (a dwarf) gets killed and the adventure focuses on them chasing the killer.

In our game, the assassination attempt took place, but due to their choices some of the players happened to be close to him. One of the players happened to have specialized in Medicine, with both feats dedicated to it, and he happened to roll 20 on his medicine check. Despite what was mandated by the book, the poison was countered, and the dwarf didn't die, and we had plenty of material to go on to do completely different things than what we were "expected" to. 

Hey Lorenzo,

On daggers at throat: The prisoners had been beaten and the Aarakocra has suffered minor torture (plucked feathers), so I could have gone through the entire rigamarole of figuring out how many hit points they have. Instead, I made a ruling about how difficult it would be to escape: on success, blah, on failure, you are dying. D&D hit points are an abstraction of combat endurance in skirmish combat situations, and also a measure of heroic plot armour. I am generally careful about applying them in other contexts and by no means feel bound to them.

Had the character been of very high level, I might have suggested that they have a number of luck / divine favour hit points equal to, say, their level and those are what protect them against the knife on their throat, the rest being endurance to dodge and sustaining minor injuries and whatever. But at low levels this does not make a difference, so I have not figured out the details.

An amusing alternative would be to treat the rules exactly as written and see what kind of crazy fantasy world would emerge. But that is not what I am doing at the moment.

This was not a matter of drama; rather, a matter of what is a playable way to abstract a particular event within the present rules framework so that the ruling is not in favour or against player characters, but rather respects the fictional circumstances.


As for the secret door: Definitely working as intended, yes. If the players had decided to search for it, I would have asked how much time they want to spend on that. Just a quick look gives bad chances of success (maybe DC 25), 10 minutes per ten feet of wall gives some chances (maybe DC 20), while an hour per 10' makes it pretty much automatic that they find it, or a trivial DC in any case. I typically let them roll and they can always choose to spend more time, but I do not tell whether there is something to be found. (With an OSR system with 1/6 chance of finding stuff I would roll for them, but here the uncertain DC compensates for that, I feel.) After the place had been cleared I would have rolled a couple of random encounter checks per day, maybe, so it would have been entirely safe to check that one room throughoutly. Checking the entire dungeon with full care would still have taken days.

I have thus far not faced any dysfunction with this methodology.


With respect to using published material: agreed, though I also use them to add variety to the adventure locations, too.

LorenzoC's picture

Yeah, the explanation makes sense. I was asking this because I've found that to be a frequent element of friction between players and GMs in OSR play.

It's clear the hostages could be dead well before the players got to a situation where they could get to free them, so the "dagger at the throat" situation could even be seen as a form of mercy from the GM; on the other hand, looking at a few similar situations I've run for my players in the past, I had a certain number of complaints from players who said I was bypassing the rules to force a "moral dilemma" on them, which they felt like a form of railroad ("do something that fits my own perception of how the narrative could unfold, or die").

It's an interesting gray area to explore - on one hand you have the need to be able to portray a certain situation in a way that makes sense (such as the hostage with a dagger on their throat, or the noble making demands with 20 crossbows trained on the player); on the other I've found some players consider the rules to be a form of insurance against the power of the GM, or at least equidistant from everyone at the table. HPs are a powerful manifestation of that "protection" (precisely because they're narrative currency and not "meat points") and if you take them away or alter how they work, you may rustle some feathers - at least in my experience.

LorenzoC's picture

Clarification: when I say "OSR play" I mean the general phenomenon people refer to, rather than any specific ruleset or group of rulesets.

Hey Lorenzo,

I would say that the double approach of discussing these matters in public and relying hard on the fiction (and secondarily on the rules) is the key for constructive handling of these issues. Some other techniques:

1. Have a coherent interpretation of what abstract matters, such as hit points, armour class and spell preparation, mean in the fiction. Then you can deduce how the rules should be interpreted in a particular case from first principles, so to speak. This is a work in progress for me.

2. In campaign play, and with a fairly stable group, don't be afraid to have a discussion of rules philosophy under play. Though it slows down play at that particular moment, it pays dividents in the long run as the group develops a shared understanding of what the rules actually mean. Though try to have such a discussion before you get to the most intense moments, if possible, if you see something like an assassination attempt (a typical place where D&D combat rule abstractions break down) coming up.

3. You can still use the rules as written as a fallback and an insurance; if the group does not come to a shared understanding about how to resolve something, just default to the rules as written and talk about it later if there is interest or need.

As far 20 crossbows, that is still on the order of 40 points of expected damage unless the character has strong defensibe abilities. To survive that the character needs to be of fairly high level, and it still hurts, so it feels completely appropriate - high level characters are powerful in exactly that sense.

Add new comment