[Trad+, RR-] A way to salvage linear modules

OK, this post is written from a perspective of GM-led RPGs. Think D&D. This post assumes that playing trad games, or games with some overlap with trad play, is something that some people do still want to do, at least some part of the time.
Before I started playing, many modules were written in a place-centric way. This is the cave, this is the wandering monster table, this is the porte, the monstre and the trésor. (I would’ve had a much easier time getting started if I had had access to something like B4 The Lost City back then.)
But when I did start playing, most modules I found were written in a timeline-centric way. First this happens, then this happens etc etc. Places are described only when the protags were expected to go there. Characters were described only when the protags were expected to meet them.

And as I’ve recounted many times before: this baffled me to no end. “But… but… what if the players don’t do that?” and I started improvising instead but the way I improvised was in a way that didn’t really give the players agency or really build on what they did. Some players did love my style but I found myself with a dwindling group without any real feedback of what I was doing wrong. When I found story games I discovered how to improvise together, which was better, and then when I found OSR I found how to get back the GM/PC split [which has its upsides] but also maintain some real agency.

Now that I’ve many years of experience running sandbox games, many of those timeline-centric modules that just left me confused and frustrated suddenly seem playable again.

The solution is to both

  1. Place them in a sandbox. Have other stuff going on, have a full setting book and random tables etc etc where you can play without any rails. Like if you have a full, working, set up sandbox, then you can place the starting village of Torch (from the linear Iron Gods adventure path) on your map and the starting event from that village on your rumor table. And,

  2. Treat them as a sandbox. The thing that I didn’t fully grok (and old-timers are shaking their heads as they read this, if they knew how to use these modules. Well, I didn’t) is that even linear modules have places, people, monsters, dilemmas etc set up. If they don’t go to the place the module-writer expected? Well, hopefully the place they do go to is described somewhere in the module. Extracting places, people, items etc from these modules is often a lot of work but it can be done.

And that’s something I wish I had understood 20 years ago. Better yet, I wish that I had started with some small, easy sandbox modules like B4 The Lost City.


This isn’t new, every DM worth their salt knows this!

A lot of people are able to run linear modules successfully. Some using this technique, and others using the opposite technique (and by opposite technique I mean keeping them on the rail — often overtly and with buy-in). And I’m not saying I invented this technique. In hindsight, I guess people must have been doing it this way, since it makes so much sense to me. But I’ve never learned it nor did I see it written down. Hence this post: to write it down so that people can learn it.

Does this work for every linear module?

Not as far as I know, but it works for a non-zero amount and even that is cool to me. It means that some pretty darn interesting setups and people and places that I’ve let gather dust over the years now can find a new life.

Ugh, is it even worth it?

Probably not. There are so many good sandboxes small and big now. And it can sometimes be a lot of work. But if you have a pet module that you always wish you had been able to do something with…

Wait, does this mean that presentation doesn’t matter?

Oh, it does! As an example, I had a very hard time running Princes of the Apocalypse even though it is a legit sandbox, and even though the setting and characters are (to me) very compelling, because the book is so weirdly organized. And then extracting info from linear modules can be even harder.

Wait, does this mean that you think that timeline-centric presentation is sometimes the right choice for a module?

No. My own opinion is that it’s the wrong choice but I’m trying to write down a way to work with modules that made that choice, work with them spite of that choice.

I often see that intro adventures to trad games are linear…

I know, and that’s bull! I know that many people believe it’ll be a lot easier and more scripted and tutorial-like for new GMs and DMs but… there are so many ways things can go wrong.

What do you recommend for trad game makers creating intro adventures?

Make a small sandbox. A town, some small dungeons, rumors that lead to those dungeons, and well-stocked and evocative encounter tables for every region. Emphasis on set up over story. Teach when to look things up and when to make things up. (I made a first draft of a a game mastering tutorial text in Swedish, here.

What do you recommend for adventure writers who’s working on a big adventure?

Organize it as a sandbox. I think Curse of Strahd is a usable and interesting example. As are Stonehell and Barrowmaze.

What do you recommend for GMs who don’t have a big sandbox?

Start with a small sandbox. I’m trying to create a way to grow a sandbox, and one of my first attempts at that is the Quest Queue.


  • I feel like the ideal way of doing really timeline-centric modules is to have everyone read the module (and potentially even have everyone have a copy of the module that they're following along with in play) so that then everyone is on the same page and is working towards telling the story of the module together.
  • edited April 2018
  • edited April 2018
    Related to Emma's suggestion, I recall reading of a very unique Pathfinder group that had a similar approach. Instead of everyone reading the module in advance, the DM would read the module aloud to the players at the table. Then, when they reached a combat encounter, they would set it up and play it out, players vs DM.

    That way, the DM didn't have to do any prep before the session, and nobody had to go through a facade of the players having to guess what they were supposed to do in order to get the correctly planned story and series of fun encounters. Always stuck with me as an interesting way to turn a kind of flawed rpg into a focused storytime/wargame combo.

    Of course, clearly that was for a group that wasn't interested in the roleplaying/theatrical aspect of the game.

    (I wish I could find the original post where I read about that... it was somewhere on reddit but I can't think of what keywords I would use to find it now. I recall they had a slightly satirical approach, where the DM would read aloud, "If the players talk to the old man..." and then pause a moment, and the players would say, "Yep, we talk to the old man," and then the DM would continue reading.)
  • edited April 2018

    Both of your suggestions are on how to railroad. This thread is about how not to railroad. That's why I put the "RR-" in there.

    What I was after here was instead a way to use ideas (like people, places, dangers, dilemmas, items) from linear modules in an non-linear, open-ended way. Placing the railroad in a bigger living landscape, making it so that people can fall off from the train and land into something else that’s also interesting, do other stuff, and then maybe they later encounter things that also came from the linear module, and maybe they don’t, and maybe they do things out of order, and maybe they don’t do them.

    In my imagined history of how trad RPGs developed, there has had to have been some DMs who used the modules this way, or some writers of linear modules who intended the modules to be used this way.

    They had the world source book, they had the encounter tables for the region, they perhaps had some extra dungeons they had dropped in there.

    “OK, we don’t talk to the old man! We go out into the forest!”

    DM rolls for random forest encounters, and gets ‘A tribe of bulliwug, tracks.’

    “OK, you meet… roll, roll… you see some toad-like tracks in the wet mud here, and smell of charcoal, what do you do?”

    etc etc.

  • 2097 said:

    Both of your suggestions are on how to railroad. This thread is about how not to railroad. That's why I put the "RR-" in there.

    I'm sorry. I didn't understand what the RR- meant, and I got confused, and I apologize.

  • Oh, I guess that's on me. Don't worry about it. And now we have two good topics in separate threads
  • edited May 11
    I think this is a really challenging thing to do, and a major project.

    Things I would do or consider doing:

    * Consider each NPC, location, treasure, encounter, and rumour as a separate entity and list them.

    Use them separately, as you wish. That's already some utility - some material.

    Or, you could go more in-depth:

    Make a relationship map between them, showing connections from each to the next according to what's in the module (and add some where necessary). Allow the players to navigate these elements as they wish.

    For each element, write down its goals and likely actions they will take. (And perhaps how long they might take to accomplish those goals, if no one interferes.)


    * Find a good early starting place in the adventure,and roll from there. Have the "elements" you derived pursue their own agendas while the PCs do the same. Anything leading up to that early starting place has already happened, before play began.

    Some alternatives to that:

    * Start later in time: assume some percentage of the adventure has ALREADY HAPPENED, with a group of adventurers who ultimately failed.

    (It could even be that the entirety of the events described in the module have already happened!)

    Ask the PCs how they wish to deal with the aftermath.

    * Invert some aspect of the structure, like putting the Maguffin in the hands of The PCs and having the steps outlined in the module being taken by an enemy group to find and claim it.

    This will force you to reconsider the elements at play in the module and create an emergent story instead.
  • edited April 2018
    Paul_T said:

    Consider each NPC, location, treasure, encounter, and rumour as a separate entity and list them.

    Yes, extract them. When you’re experienced doing this, you can do it on the fly, but it’s a useful exercise to Actually Do It sometimes.

    Paul_T said:

    Invert some aspect of the structure, like putting the Maguffin in the hands of The PCs and having the steps outlined in the module being taken by an enemy group to find and claim it.

    That is exactly what we did for Corsair Council. Was awesome.

    Paul_T said:

    Find a good early starting place in the adventure,and roll from there […] Assume some percentage of the adventure has ALREADY HAPPENED, with a group of adventurers who ultimately failed.

    This is a good technique. I also sometimes do the opposite: I put the “starting event” later on the calendar. Maybe every intradiegetic day there’s a random chance that the big starting event will happen. Makes it so that the players might be able to discover some of the elements from the campaign on their own before. Especially good when you plop down stuff from a linear module in a bigger sandbox. Like the example with Iron Gods; you can place the city of Torch in your sandbox and have it working juuuust fine until one day: boom, the starting event from Iron Gods randomly happens. Or Carrowmore from Deep Carbon Observatory. One day the levee breaks.

  • edited May 11
    Randomizing the timing of the "inciting incident" is clever! I like that. Basically, anything to throw a wrench into the careful plan of the module should work, but that's a good one for the "campaign at large". Interesting!

    I was thinking about this further, because I won a "Call of Cthulhu" starter book/package, which includes basic rules and a "starting adventure".

    The adventure isn't a complete railroad, but it does have a basic sequence of events which will occur in some order or another. There is a "haunted house", the investigators are expected to "investigate" it, and there is a monster inside who wants them dead or gone, so sooner or later it will come down to a fight.

    The setup for the adventure includes four or so places the PCs could look for info first, and hints at a backstory at each one (e.g. the Federal Archives, places like that). (Oddly, it doesn't seem to me that any of these sources of info interface with the final encounter in any meaningful way, but at least they're atmospheric and each present some colourful detail.)

    How would I use this in a campaign? I'm going to brainstorm right now, so bear with me, and we'll see how fruitful this all is.

    Here we go:

    First of all, I would create those five locations - the haunted house, and the four information-gathering sites - and place them "on the map" (whatever that means for my particular hypothetical campaign).

    I would also introduce the current owner of the property (who, in the adventure, hires the PCs), with the given motive of "looking to hire someone to help him deal with his haunted house". Finally, I would find a way to hook in the original owner of the haunted house, since he is referred to in each bit of "info" you can find in the adventure (he's long dead, but hinted to be a part of a historical conspiracy).

    Next, I would want to integrate all this with other stuff in my campaign. These connections will form the framework for long-term play.

    For instance, let's take the four locations. For each, I'd make sure that an important NPC, item, or piece of information related to something else I've prepared is placed there.


    The module has the Federal Archive as a location the PCs could visit. In my (hypothetical) campaign, a PC has a love interest, or another adventure requires a mysterious clerk - that person is now placed at the Federal Archives, as the one in charge of the information related to this "haunted house" (and perhaps searching for that info themselves!). So, now, interacting with them could potentially lead to drama and/or at least two other "plot hooks" or adventures.
    Notably, if left alone, this person might stumble on this info themselves, go to investigate the haunted house, and get killed or cursed. (This is almost guaranteed to be interesting, since we've already established that this NPC is quite important to some other aspect of the setting or the PCs - a love interest, a linchpin in some other plot/adventure/intrigue, or something similar.)

    I'd do the same for each other "information" location, and perhaps the owner and the haunted house, too. Much more interesting if we find out the haunted house used to belong to the kindly woman the PCs rescued last session, for example. And what if one of the PCs has a mysterious history? They could find out that the original (long dead) owner of the house is a missing relative they've never been able to track down before. Now that PC has a personal stake in dealing with this situation somehow.

    By the time we're done, each "element" in the adventure has connections to other stuff in my campaign. All these connections mean that all kinds of interesting emergent stories can be told. For example, it might feel quite organic if, over the course of play, the PCs visit those four "information locations" and slowly collect clues about the history of this house. They're not looking for the house - they're searching for clues or people related to some other adventures (since I've seeded them at those locations). But, in the process, they'll slowly learn about this house with the weird history.

    When they look into the house, they find that the owner is, strangely, trying to hire someone to help solve the mystery of the haunting. Now it feels like something they've found themselves, rather than something I pushed on them, emerging organically from small, seemingly irrelevant, details several sessions back.

    If there's something deeper I can connect this to, that's even better. Perhaps we have clues about the deeper historical conspiracy, for instance (it's only very barely hinted at in the module itself, but could add depth to my campaign if I have other unexplained incidents and historical tidbits I can tie into it).

    Finally, I'd give a thought to what happens if the situation is left alone. Who else might the owner of the property hire, instead of the PCs? What might happen to them when they investigate the property? (Either or both could be randomized, for extra surprise - "Ok, it's [rolls a 3] the old woman from the FBI, and her going there means... [rolls a 9] she survives, but becomes haunted/possessed herself."

    Are there NPCs in my setting/campaign who would jump on this opportunity? Then they do!

    This rather lacklustre adventure is now giving me all kinds of interesting material for a campaign.

    I'd also consider assuming the adventure is already played out (presumably by a group of NPCs), and see if that gives me anything interesting. Did these NPCs "win", defeating the creature, or "lose", by dying in the house?

    This would be fairly weak as a "bang" introduction, I feel: "You wake up on Monday and the newspapers are reporting a strange murder at 1506 Penny Lane..." That's not too my taste; it feels too much like I'm leading the players by the nose. But it could work.

    However, if my campaign has some unanswered mysteries I'd like to see resolved, then this could be a good opportunity. Instead of the boring "you learn about a murder" opener, I get to answer existing questions in my campaign. The material in this adventure module plugs into those things I haven't determined yet.

    Let me illustrate with an example:

    Has an NPC disappeared recently, and under mysterious circumstances? Ok, I now know what happened to them: they're lying, dead, in the basement of this haunted house. Or, maybe, has another NPC suddenly starting acting very strange or suspicious? Ok, maybe they've discovered the "haunted" house and are now acting under some sort of occult influence from its monstrous denizen.

    Either way, I get to tie a fairly uninteresting "explore a haunted house and get into a fight" module into other character drama, relationship maps, and events in my larger campaign. I'd imagine that, with a better adventure, it would be even more fruitful.
  • Paul, I'm now super happy with where you're taking this thread. My hope is for this thread to be a resource people link to for years and years and you really contributed well. Thank you from the heart.
  • Some of these techniques I've heard on Behind the DM Screen podcast and some on a now defunct youtube channel, something with blu or blue? But I wanted there to be a text resource too. For the canon ♥
    Just like how I always go back to old rgfa threads and forge threads and "anyway" posts.
  • edited April 2018
    Thanks, Sandra!

    It would be interesting to compile this kind of advice into some more easily digested form at some point.

    I think the key in what I wrote is to find variety, spice, and liberty (in terms of player choices) by combining multiple sources. I haven't ever tried this, but maybe weaving together three or four rather uninteresting adventures like this one into a coherent whole makes for a really cool sandbox. It would be an interesting experiment, in any case.
  • What I've found is that even publishers of linear adventures do publish "world books" that provide at least rudimentary sandboxes that their linear adventures are put in. For example Iron Gods has the Numeria book. So that's the "second source" right there, which is another sign that this mode might have a long history.

    On my first browse-through of al-Qadim, I was disappointed. So many of the modules had linear elements. But once I had gotten to the more sandboxy book, including the random encounter tables, many of those modules suddenly felt usable again.

    What does seem to be an even more central key in your posts is a set of techniques that are designed to sort of "shock the DM out of linear mode"; changing who has the McGuffin, changing the start date, and, yes, adding in multiple compelling threads to pull at by combining modules.

    Which is good, I do that. I'm learning to be careful about it; being able to pick between funnel A and funnel B is sometimes more frustrating and less rewarding than just having funnel A. The idea is instead to use the modules to create a living world that the player characters can truly live in. For that reason, it's sometimes good to only have one megadungeon in the vicinity. I dunno. I do like putting a couple in there… I just love dungeons…
  • How would you approach it? ( Placing timeline modules within a sandbox)

    Would you have some kind of random when ? chart that hints at preliminaries if it doesn't indicate the event itself begins?
    A bit like rumor tables crossed with the Threats/Fronts from Dungeon World?

    A lot like rumor tables, yeah.

    Here's how I'd math it out:
    let's say I want the event to have an EV of happening 36 days from now.

    I'd then roll a 2d6 every day and look for snake eyes and that's when it happens and the rumors go live.

    As an example the island rising in Corsairs of the Great Sea. I had that put to happen at average of two years (because of "long tail" maybe it'd almost never happen, but in two years we've got a 50% chance). So I made a dice expression that had a one in 730 (because that's the amount of days in two years) chance of succeeding and rolled that every diegetical day. Like I think a 3 total on 1d8+2d10 or something.

    If it would've come up, that's when the players would've been notified that the island had rose.

    You can have the list be more of a tree than a list; if the events have prerequisites, you'd make it so that the prerequisite is what you're rolling for and only after the prerequisite has happened, you start rolling for the real event. On a different die expression possibly, maybe you want an EV of 6 days so you roll a 1d6 every diegetical day.

  • Yeah, that's similar to what I'm thinking, but with possibly even more cross-referencing.

    Heh, I guess for Gangbusters or Boot Hill, the set-up rumors come from the newspapers!
  • That does sound awesome. I'm not too familiar with these two games
  • edited June 4
    Boot Hill is generalized Old West.

    Gangbusters is 1920s/early 30s America at the height of Prohibition ( meaning the birth of the American Mafia as large criminal enterprises) crossed with similar period detective fiction.

    Both are late 70s/early 80s designs. The modules clearly come from a time when designers are trying to figure out how to do something that doesn't fit a site exploration model of adventuring.

    The only really strong concept for core activities ( Yes, but what do our characters do in this game?) is the criminal class in GB ( Get rich and level up by making your own breaks in this world) followed, to a lesser, more reactive degree by certain law enforcement and investigative types ( catch and convict Big Fish, level up, forget getting rich, unless you're crooked).

    Boot Hill is waaaay more open-ended than that.
  • The Gangbusters modules are mysteries* with timelines. That's where things get a bit trickier. Without those modules or something similar, PIs, Plain clothes Division Detectves, and Reporters have little to nothing to do.

    Check out Silent Legions
Sign In or Register to comment.