Month: June 2025

This is part of a “blog challenge” I decided to kickstart after discussion on Discord. It would be interesting to hear about how referees prep their games.

Prep is the material I need to comfortably improvise during play as a reaction to player action and to conduct what is the shared goal for the session. I start from the assumption that in-game needs trump what I expected, and that my job as a referee is to be sure challenges are fair and engaging. So if play reveals there’s space for a challenge, I look at my understanding of the world and prep to see what addition would be plausible. If a challenge in practice is unfair or would kill the game’s pacing, it’s liable to quick adjustment. The goal isn’t to help players or “fun” that assumes enjoyment only comes from victory or the illusion of difficulty, but to make the game matter.

I’m a fairly high prep referee because I like to understand what is going on. If pushed into a situation to which I have absolutely no prep reference, I’m capable of making something up like everyone, but I’m also likely to tell my players I have no idea and either call the session a bit earlier or if feasible ask them to, for now, pursue a different course.

So prep is highly tied to my own understanding of improvisation. I improvise, whether I’m running an OSR exploration, an investigation, or a “trad” game, from the key structure of challenge. Players arrive at a situation tied to something agreed as desirable and must creatively solve a challenge. Even in a drama-focused game, where my refereeing follows cues of genre logic as much as world logic, the structure is challenges.

The First Steps

If I want to run a game, it’s either because I had a genre-related idea (either “I want to run something inspired by X” or “players are X doing Y”) or because I thought about a game setting. In both cases, I start by doing free association of words and imagery on paper to create a “moodboard” for that project which I can always return to (I believe this is inspired by very early contact with Alan Moore’s essay on how V for Vendetta started being written). An actual imagery board also works fine, but I put words first. An Appendix N of inspiring media and art helps.

I follow by thinking about possible adventures within those bounds, to make sure the premise is actually gameable. This is where thinking games as challenges help: how many challenges does the premise sustains? If I can think of around 5, there’s enough for a short campaign at least. Challenges can also be structured as questions we are trying to answer during play (“Can the players rescue Irina before the rebellious angel of death takes her as her bride?”).

This process happens simultaneously with the more practical concerns: how many players do I want for this project? Is it better for a duet or group play? It will happen in-person, by online voice, or text? The speed of play in each case is a major factor for prep.

Since games take place in a fictional world, I need to have a solid grasp of that world. This is where a structure like Kevin Crawford’s World-Region-Kingdom is useful, since I can have broad understanding of the wider world beyond the actual game focus that further informs the focus.

If a particular point of the world or potential adventures seems unclear, I deploy the method of writing down questions I need to understand about the issue and try to answer those questions, snowballing further questions.

Campaign Prep vs Session Prep

There’s a different between the prep before the first session and between sessions. Campaign prep should be enough to get me to the first session, but it will inform all prep after. Hence it’s better to have enough for the second session: I think about where’s likely for the first session to end and, without marrying such hypothetical result, sketch what I would need to know for the second session. That sketch in general is helpful to further inform relevant details for the first session.

Some questions are paramount when doing campaign prep:

• Are the PCs complete newcomers to the situation, or do they have previous knowledge the players must familiarize themselves? If they are newcomers, which sort of adventure or procedure will best inform them?
• Do players hold responsibilities and social power within the world, with major ties, or are they more adrift?
• Are players expected to take initiative on their own objectives upon certain agreements, or are objectives handled to them and their agency comes from how to achieve those?
• Is play heavily proceduralist and is prep based on gathering material for those codified procedures, or is it “mission-based” and traditional, with prep based on contextualizing all that?

I like to have a list of the major locations and major NPCs in any case, and whatever prep is needed according to the questions above. Such information is best organized in spreadsheets for ease of reference.

Session prep is highly guided by what the players announced to be planning next. I generally take their feedback and combine it with the notes I took about their actions during the session, notes taken about what such action could snowball into. Sometimes in the heat of the session I don’t know what that is, but with experience we get the sense of when a fairly impactful action was done. I only prep enough for the next session.

Procedures and Rules

Procedures guide further prep. When we think procedures in a traditional context, we think OSR exploration and random tables. I think mission-based games (this refers to a style of trad-adjacent gaming where the GM is curator of highly specific premises and challenges that lead to a problematic climax, and which I prefer to the “event-based” “narrative” mentality that historically dominates trad gaming) are also proceduralist in a sense, although less based on ruleset mechanics and more in mentality. Among options for mission-based prep:

• By picturing the crisis, I can trace backwards the node map of challenges that will lead into it, with the challenges results influencing the circumstances of the crisis (also a method for investigation, which is mission-based anyway). Node maps are generally good because they organize information efficiently while tying prep back to making sensible connections players can follow to make decisions.
• A timeline of the mission and crisis unless players interfere.
• Dividing the scenario or campaign between Assets to be seized and Hazards to be destroyed, with the cycle of play being based on players doing one of those things at a time, and attaching small mechanisms for how the untackled elements develop without their action.
• Developing random tables tied to a situation for developments during the progression of challenges.
• Something like PbtA GM Moves, written and attached to different setting elements and challenges, to be considered during rulings and prep.

This is a good moment to discuss ruleset. By this point I will likely have chosen a base, considering what kind of activity is involved in the central challenges/play questions, which informs me what will be recurrent enough to ask for codified resolution or what I want to elide. I trust that play itself will add more rules as needed.

I avoid rules that demand more prep. Central resolution mechanics that bake in degrees of success, for example, demand more preparation to account for reliable rulings than binary. Combat rules are the one thing I consider as soon as possible since I don’t enjoy ruling violence spontaneously and prefer a robust procedure to ignore that aspect of play. Skill systems work great to elide detail for resolution of things that are relevant but better fast-forwarded or quickly answered.

Specific Prep

Let’s get on some specific tools that I like to have.

• A list of names for spontaneous NPCs, since I dislike coming up with them.
• I think about which locations or features are likely to show up in the campaign, either as major spots or as places players might spontaneously like to visit, such as a bar, and keep very small tables in my folder for that.
• I like very small encounter tables to guarantee the ideas there see use.
• For NPC prep, I need a name, one visual detail, what are the assets they have that might interest PCs, what do they want, and a general approach to problems. Regarding personality traits and sometimes appearance, I select a movie actor or cartoon character to associate with them, which allows for very effective references while portraying them. FATE Aspects served a similar niche well before.
• Dusk Witch’s Context Engine is worth considering for campaigns, especially in the first steps of prep.
• I think having one spark table ready as a last improv resource doesn’t hurt.
  For dungeon prep keys, besides adding a description of room dimensions that can be read effectively to players, I add sounds and smells coming from exits as much as possible. I enjoy using Priestess of Spiders’ Ten-Room Dungeon in an expanded fashion, starting with the random encounter table to stock fixed encounters based on it. I prefer to automatically generate or outright recycle maps, as I don’t enjoy creating them.
• For hexcrawl preparation, Josie’s method works well. I don’t care to do the whole Landmark-Hidden-Secret system in most cases, and would prefer to do so to stock and organize a city.
• Speaking of city prep: having determined the major players, I assign each a die and drop on a table, assigning connections between them that can either be relations or sparks for challenges and scenarios tying both (the same method works well for building a pointcrawl geographically, and even for organizing the politics of a hexcrawl). I prepare a list of the main services and locations players would know and handle it to them. A rather large random encounter table with small attached tables for certain districts works well; I would be tempted to use Tetris-like blocks to organize a city into districts, indicating the general paths players could take, from now on. Locations most certainly can be created with Josie’s hex method for support.
• For the world map, I take one from the real world with the geographic features I need. Hate mapping.

A d666 table is an useful structure to contain some of the above. It comports 216 entries, and the horizontal rows divided into 6 parts can refer to motifs or central ideas of the campaign, meaning I just need to think of six items referring to each vertical row, which can be divided into more practical categories such as Locations, Assets etc. As a spark table and context engine, it works magnificently.

The Necropolis of Nuromen is a Blueholme module of vernacular TTRPG fantasy (aesthetically medievalist fantasy) that I selected to run as a campaign – or better, several – from March to late May. It was divided between an open table dungeon campaign over voice and three duets over live text. Despite an incredibly chaotic life and schedule, I’ve done my best to run it every week regardless of my mood and energy, which amounted to four games per week ideally, although late April and early May made it more slippery due to personal disasters. Regardless, no week went without at least two sessions (from my recollection) until the point I had to cancel it.

I have some thoughts now that it’s over. This isn’t a play report where I describe what happened during it; I might do that some other time. That’s about my thoughts as a referee on the process of running those games.

I wanted to get back at running games after a few months without doing anything of the sort and was fairly worried as to how, and afraid my skills were rusty (they were, alas). I decided to grab a module and do something out of it, and proposed Nuromen as a duet campaign centered around a wizard’s apprentice PC. A lot of people were interested, more than I could conceivably run for as duets, and I found myself having expanded the dungeon significantly. Hence, three duets and the open table.

The Open Table

This was my first time running an open table, running a campaign without any activity outside the dungeon for maximum casualness, and doing both in my second language (English). We had a varied number of players for each session, generally in the 1-4 scale.

It was played with a small hack of Tunnels & Trolls with corruptible freeform magic I took and adapted from Streets of Marienburg. I believed it was best to keep the ruleset open to change and development according to need and player feedback, which we did.

What Did I Learn?

Although I was fairly satisfied with the rules and exploration procedures for the loop of play, magic was an issue due to constant bad rolls and too harsh consequences. It confirmed a long-standing suspicion I had but couldn’t prove in my own play experience until now that corruptible magic of the sort doesn’t integrate very well with my vision of resource attrition and strategic concerns during the classic dungeon crawl, and that I prefer to tease out its consequences during the course of drama-focused play. The classic Vancian magic or the point spending of BRP and T&T are a better choice for that structure, with magic as a reliable tool to tackle the problems of the dungeon.

I had plans to complexify the campaign with a network of rival adventuring parties, the actual rhythm of the open table asked for simplification. I didn’t care too much about restocking as things went. I started the game wanting to guarantee that I remembered the internalized procedures and philosophy of adventure play; I ended it deliberately forgetting the pieces of the intricate machinery that turns the dungeon from artificial play space into a microcosm of conflict. Revisiting my thoughts, that’s partially due to my inexperience with open tables, for which “best practices” amount to “have things ready for whoever will show up and be consistent”.

Something wonderful and terrible about TTRPG play is that you can identify a problem, and sometimes identify the solution to the problem, but the more you think about it you can’t point out all the issues that originate that problem. Example: I realized that for campaigns without a consistent group (which establishes an invisible code of communication) I need more prep to run in English, and to further codify my talk into a reliable, repeatable structure. The open table demands a mechanical efficiency of description and establishing the scene, even more than the closed table, and for my next attempt at the sort I will certainly spend some time considering it and writing keys that can be read in a perfect manner to deal with the problem. But does the problem originates from running in a language I don’t master, from the open table structure, from working with material that wasn’t originally written by myself, or all three, and if so, which factor is more relevant? It doesn’t matter… to a point. I will return to this.

Experience distribution, as I noted soon, would become a problem. I don’t even mean in the sense that the players weren’t finding enough treasure, or that there wasn’t enough treasure, but that the distribution of experience itself, considering the rules for combat and saves, wouldn’t work for the open table structure as soon as they got to different floors and difficulty went up. I considered to not change the difficulty of saves and such when they got to other floors to account for it: the continuity of play is more important than some abstract ideal of escalating play.

I’ve also learned that a long time without play makes me fucking suck at describing room dimensions in an actionable way, and it took some sessions until I was fully back into a reliable system to describe such. Doing pre-session notes focused on room dimension descriptions is now a very obvious part of prep I need to consider.

What to Do Different?

• Different, more reliable rules for magic.
• Restructured my dungeon keys to account for the linguistic barriers.
• Used a mixed of paper and digital prep for quick reference and to better track the player characters besides my own GM-facing information. I believe I find myself more comfortable having all the information always in front of me, but digitally it can become a hassle.
• I prepared party sheets in Google Sheets for ease of tracking from the players’ side, assuming that since groups would be inconsistent and people could just land on a role, that would facilitate play. While I don’t think the theory of it was wrong, and I did encourage players to alter the sheets to what seemed easier to them, ultimately I believe I must search for a new collective sheet structure for open tables than the ones I’m used to from closed groups, and perhaps let the players themselves sort it out. I believe my sheets were carelessly constructed for the task and created more problems.
• My philosophy of prep is that it serves the purpose of giving you the structure to improvise content. While that’s true, I believe I was underprepped for the organizational challenge of the open table and unable to find a workflow and “GM Screen” that worked for those demands.
• Not playing Nuromen (more on that later). It’s a great dungeon, but I shouldn’t have used it.

The Duets

This was certainly not my first time running duets. We had three duets, with three different wizard apprentices interacting with the same characters, in some cases the same scenes (I could recycle a fair bit of description and dialogue from how situations were going, especially in the introductory sessions, but things quickly became vastly different). I was partially curious to see how different play could get. As is my preference, it was blackboxed, with solo PCs having a fairly heroic capacity to do stuff, and players had access only to the fictional understanding of the situation.

What Did I Learn?

In general, nothing.

It’s strange to say, but the more I try to think about unusual situations or difficulties from the format itself, none really showed up. There was no difficulty happening during the actual refereeing of play, or organizing my notes for what was happening. However I did learn a couple things about myself which I will get to later.

I’m lying, I did learn something: I have a fairly reliable sense of what players are probably thinking, and I should exert much more confidence in that sense.

What to Do Different?

All the issues with the duets can be easily summarized, for different reasons, to the final point of the open table: I shouldn’t have used Nuromen.

What’s So Wrong About Nuromen?

There’s absolutely nothing wrong with The Necropolis of Nuromen. It’s a very well done dungeon. If I have used absolutely any other dungeon, this section would be named What’s So Wrong About X. The problem of the Necropolis is that it is pre-written, which lead to different issues for different play formats, but also to the same issues.

The point of modules is threefold for me as a referee, and let’s check how Nuromen served in this context:

• Modules are pedagogical material: I don’t feel I require module pedagogy anymore as a referee, and I didn’t want to learn anything about structuring play from Nuromen anyway.
• Modules as a time-saver: I wanted to quickly return to play, and Nuromen served that perfectly, but it turns out that the time it saved me to just going back to play was just that. It got to me to the chair and allowed me to quickly run sessions, but it didn’t save me prep and writing time I should have put on to truly making the campaigns meaningful and interesting for the players, and more than that, it didn’t save me the prep time I needed to understand the campaign itself.
• Modules as a structure to improvise and riff on: I think many are against modules because they think it is false gaming, as laziness or as a straight-jacked. I think a campaign played with modules can be just as original and challenging as coming up with everything yourself, especially because when you come up with anything you are still riffing on hundreds of references you have anyway. My problem was that, in the rush to return to the table and to the actual experience of playing and conducting multiple sessions, I wasn’t willing to change the module enough.

Looking back, what I should have done is writing everything from nearly zero.

The Duets: After I announced the end of the duets, one of the players wanted to give me a piece of feedback. I knew exactly what she was typing even before she finished it: that moving from the social intrigue and interaction towards the dungeon was a step down, not because anything was refereed incorrectly in the dungeon or because it wasn’t enjoyable, but because it was very clear that a) I felt more comfortable with social interaction and intrigue over a duet in general and b) it was very clear that all the social stuff was much more my material than the dungeon, and there was a very clear break from voice and logic between the two stages of play.

To be clear, all the social stuff was already suggested in the module over a page: the castle, its inhabitants, the economical issues looming. However, everything there was developed by me in my prep and tied to other original ideas I had, and creatures from the module (vernacular sorts such as hobbits and elves) were replaced by my Dodgers and my rendition of ghouls, as well as wider distant things that maybe players could get to eventually. The dungeon stood out not because things weren’t modified (they were), but because there’s a clear break away from what purpose those changes served.

Looking back, I should have replaced the actual dungeon with a much smaller lair that could be quickly explored and the ramifications of intruding on it and finding out its secrets should have been developed socially back at the castle. And I knew it, even though I had fun and the players also enjoyed the dungeon exploring itself. But I knew the disconnect would kill the duets, and I should have leveled with them. But if I did so, I’d be back at the square one of writing something new when I was worried about getting un-rusty for actual refereeing.

The Open Table: honestly, when it got to the point that I wasn’t restocking anymore, that I had given up on complicating the dungeon exploration with rivals and other factors, that I was fully committed to handwaving a lot for the sake of pacing, my only thought was “you should, then, have written the dungeon from zero and construct it further as stuff goes, but make it yours”. I didn’t feel comfortable describing something that I couldn’t picture very well even though my notes were immaculate and Nuromen’s writing is very strong. But it wasn’t mine enough, and all the concerns one usually has with writing dungeons (accounting for the reality outside it and how it will mesh with play) weren’t present in the open table anyway, so I should have modified it further (all my modifications were additions the players never got to anyway) or, better yet, wrote it from zero.

Conclusion: I don’t consider any of the campaigns a failure, firstly because the act of play is not a success/failure binary. Everyone had fun, I got back on the saddle, and I learned stuff about a mode of play I had never GMed. There were successes. But I do think I failed myself on some level by not prepping different and not using the rush of having the games scheduled to organize myself and write more. I’m someone who enjoys deadlines even though I can organize myself without them. Having a deadline to start the Nuromen sessions when I had a vague idea of what I was going to do was helpful. I should have used that time more.

What Will I Do Next?

“Stay the hell out of dungeons” would be a pity statement that gives the impression I’m tired of running dungeons. That’s not the case, I’d love to run more dungeons, especially written by me, even an open table dungeon with all that I learned. But as I was running everything, as I was getting the feel for the duets and thinking about what I wanted to do, my conclusion was, to my own surprise, “I miss running things that aren’t dungeons.”

I mean social sandboxes of the sort I grew up with, I mean trad gaming focused on a certain sort of emotional escalation from the adventure, I mean investigative horror, I mean having the dungeon itself as a short but memorable set piece structure of play as a part of a larger thing (see my comments on what I should have done for the duets). I was also thinking so hard about the urban and contemporary fantasy that filled my recent content posts, or looking back at the metaphysics of Ravenloft, that I was probably out of step with what I was doing, aesthetically, which made my enthusiasm and focus more difficult.

I don’t know what I will run next. I know I need to run something until the end of this month, even if it’s a one-two shot of investigative horror with a single player, using a module if necessary, just to keep playing. Playing isn’t an obligation, but I do think the point of a game hobby is to play, and in life it’s extremely easy to never play. I don’t know what comes next, but I know something must come.