Welcome to the GM guide, make yourself at home! Grab some cocoa, pick your favorite chair, and get comfy.

This may seem like a lot, but don't be intimidated. I'm mostly just taking time to break down and draw attention to things that you're either already doing automatically or that will become automatic very quickly once you get some practice.

I'm not going to pretend GMing is trivially easy, or that it's right for everyone, but it doesn't have to be some scary arcane incomprehensible art passed down through secret rites and oral tradition either. I ran my first game after only having played a single session in my life, that used a completely different system than I ended up running, and basically fumbled my way through it and figured out what I was doing as I went along. And it was still a great time! Your players are right there in the trenches with you, and as long as there's mutual respect and compassion between everyone at your table, then even if you get every single rule wrong and have to pause every ten minutes to figure out what happens next, you'll be fine. Don't sweat it - your next session will be better. And so will the one after that, and the one after that. If this chapter helps one person to hit the ground running with more confidence and understanding of what the moment-to-moment craft of GMing actually involves, then it's done its job.

If all else fails, there's only really one thing about running RPGs that you must always remember:

You Are Not A Storyteller

You are the FACILITATOR of a conversation between yourself and the other players, guided by game mechanics, that GENERATES a story.

According to legendary designer Sid Meier, "a game is a series of interesting decisions." Your first duty as a GM is to continuously tee up those decisions for the players to make, usually in the form of a PROBLEM that demands some kind of response, an OPPORTUNITY that needs proactive action to seize, or a STEADY STATE that the players can freely interact with - and that will interact right back if pushed. The process of prompting player decisions, adjudicating their outcomes, integrating their results into the fiction, and using that changed fiction to prompt new decisions is THE LOOP. The story of your game EMERGES naturally as a function of you and your players repeatedly iterating through the loop.

Your second duty as a GM is to MANAGE ENERGY AND TONE of the world and situation, and it's a critical process that occurs throughout execution of the loop. This is where you shape the kind of story that emerges from play. It's an art, not a science, and there's no one objective best way to do it. The more experience you get, the more you'll learn about what works well for you and your group and what doesn't. I call the process of making these continuous judgement calls and adjustments RIDING THE FADERS.

If you've ever run a game, in any system, you have already executed the loop and ridden faders. Nothing about either concept is revolutionary; this is fundamentally descriptive, not prescriptive. I'm breaking them out this way because I believe it's the clearest way to organize my advice on how to execute every step of the process, and because it fits the process that this game and the loop itself rely on. The loop is what you're doing; riding the faders is how you're doing it.

The Loop

Running a role-playing game, in nearly any system that uses a GM at all, boils down to iterating through this loop of questions:

1. What's going on in the fiction right now?
2. What are you trying to accomplish?
3. How do you try to accomplish it?
4. Resolution:
   1. Do mechanics take over to resolve uncertainty?
   2. What could explain this result?
5. What are the natural and/or dramatically appropriate consequences of what just happened?
6. Integrate 5 into the fiction, then return to 1.

Let's break it down.

What's Going On

This is the broadest step in the loop, covering nearly everything about the current state of the shared fiction you and the players are all operating within.

Your narration should focus on the immediate situation at hand and the currently available information, since that's what will matter to your players' decision making. But "What's happening right now" is not limited to "what's happening immediately in front of the players right now." It can include:

• Things that have already happened and things that could happen
• Things that are definitely true and things that are only provisionally true
• Characters in the scene, characters not in the scene but who are relevant to it, and characters waiting in the wings
• Ticking clocks counting down to something that will interrupt a scene and turn everything in it on its head

and more. Think of this as a complete snapshot of everything RELEVANT TO THE DECISION you're about to present to the players, whether it's a known factor they're weighing or an unknown one that still affects the stakes and consequences of what they do (e.g. a bomb under their table silently ticking down that they haven't discovered yet).

This might sound overwhelming, but in practice, you'll almost always be working with just a small sliver of the full world state. Most of the time, this is as simple as: "The PCs come to a fork in a tunnel. I know that the left path leads down to the den of a cave bear, and the right path leads up and out to the surface but is home to a colony of roosting vampire bats. I need to present the information the players can glean from their current position about their choices. Therefore, I'll describe the smell of wet animal and rotting gristle to the left, the chittering of the bats and the faint air current to the right, then ask the players what they do next." When they answer you, they trigger the next step of the loop. There may be much more happening elsewhere in the caves right now, or all kinds of drama happening on the surface, but none of that is relevant to the players' next decision, so unless and until it becomes relevant, you don't have to worry about it.

The world shouldn't just react to the player characters. The PCs are the stars of the show, but they're not the star the world orbits around. To make your game feel like it takes place in a living, breathing world, and to make it meaningful when the players push against that world, THE WORLD ITSELF MUST ACT and prompt new decisions: a man with a gun bursts through a door; a character receives a desperate panicked phone call from a friend that cuts off right as they're about to explain what's wrong; a messenger arrives bearing a formal invitation to a noble's gala to reward the party for their heroism; an infamously petty teacher singles a character out for humiliation in front of their class.

Whatever's happening, always remember the ultimate purpose of this step is to set up the players' next decision point: here's the situation, now what are you going to do about it?

Prepare To Improvise

You may already be familiar with Ernest Hemingway's "Iceberg Theory" but to restate it in brief: The words on the page are like the top eighth of an iceberg, floating above the water. The unspoken truth informing those words is the vast unseen mass of the iceberg, hidden beneath the sea, that the reader feels the weight of without ever needing to be directly shown.

A similar concept applies to game prep: don't prepare plots or sequences, PREPARE FACTS. Those facts will, together, add up to a SITUATION. Place your players inside the situation, and they will come to learn its shape one interaction at a time as they poke and prod at it - or, as the case may be, as it pokes and prods at them.

A prepared situation is flexible and resilient in ways that a planned sequence can't be. A situation doesn't rely on the dice landing the right way and doesn't require any particular order of events. Most importantly, is immune to the fatal knife through the heart of your idea that occurs any time the phrase "and then the players will ____" appears in your notes, which all but guarantees your players will do anything but what you expected. By establishing the facts of the situation and then working outwards from those facts, you gain the ability to respond to the players, build in the direction they're already moving, and have an answer ready to any action they take, no matter how insane or out of left field.

Don't just dump a complex setup like "Lord Paxton is a vicious warlord who subjugated this region five years ago and brooks no dissent" on the players all at once - they probably won't care, since that information has no emotional hooks, context, or stakes associated with it yet. Instead, let them discover it for themselves, one puzzle piece at a time, through lived experience. Have the greasy, sinister guard at the city gate demand a hefty toll to enter, in Lord Paxton's name. Have tax collectors bearing Lord Paxton's livery extort a merchant in front of the PCs. Have a parent cut down in the street without mercy by Lord Paxton's troops as they forcibly conscript their teenage child. Have an old woman a little too deep in her cups complain a little too loudly about how "none of this would've happened under old Duke Thorne" before the other frightened townsfolk desperately shut her up, and the innkeeper politely but firmly asks her to leave, their fear plainly visible to anyone looking. Have someone respond too quickly, too enthusiastically, through a sweaty brow and a fake, too-wide smile when the PCs ask about this Lord Paxton fellow, eyes darting all over as though he's sure someone, somewhere, is listening.

Critically: this situation does not care how the players engage with it. It doesn't assume they'll ask someone about Lord Paxton, expect them to take a stance on Lord Paxton, or demand they interfere to save the conscripted teen. But if they do, or if they do something else, you're already prepared for it, just by knowing "Lord Paxton is a vicious warlord who subjugated this region five years ago and brooks no dissent." By setting up facts this way, you have a baseline that you can use to respond to your players, no matter what they do or how they choose to handle the emerging situation.

Further Reading: The Lazy GM's Resource Document by Mike Shea (especially the section on Secrets & Clues) - if you like it, buy his book, Return of the Lazy Dungeon Master.

Flexible Truth

THE FICTION ESTABLISHED THROUGH PLAY IS TRUE.

Whatever you improvise, you have one absolute constraint: you must maintain the integrity of the story that's already emerged at the table. Otherwise, the entire framework that enables your players to make meaningful choices collapses, since the baseline against which they make those choices loses its stability.

As a direct corollary:

UNLESS AND UNTIL IT IS ESTABLISHED THROUGH PLAY, NOTHING IS TRUE.

The map is not the territory, the prep is not the game, and your notes are not the fiction. Your prepared notes are provisionally true, but they're best thought of as the strongest and most interesting explanation you've come up with for the current state of the fiction.

You should respect your prep, but you don't need to be wedded to it: if a better idea comes as a bolt from the blue twenty minutes into your session, and that idea is entirely consistent with the established fiction so far, then, if it still passes the consistency test and holds up after careful consideration, go with the better idea.

Knowing when to commit to the bit and when to pivot is one of the harder parts of improvisational play. The more experience you get, the easier it will become to make the call. When in doubt, trust your gut.

Problems, Opportunities, Steady States

Now that we've established how to prepare, we can start to talk about what to prepare. Most situations the players will find their characters in can be mapped across three broad categories: problems, opportunities, and steady states.

A problem can be an active threat or a passive obstacle. This is where the majority of moment-to-moment gameplay lives, especially with games that dedicate a lot of attention to combat (e.g "The marauder bellows as she readies her axe for another blow, but leaves herself open as she does - it's your turn, what do you do?") But it's also a fantastic source of scenario hooks - someone, somewhere, has a complex problem that needs solving, and the players might choose to make it their problem, or it might make itself their problem.

#TODO Opportunity - needs active player pursuit. "There's gold in them hills, go get it." This often exists at the adventure level - "what are we going after this week." Players deciding to pursue an opportunity and dealing with problems that emerge between them and their goal is one of the primary top-level loops of proactive play. In a sandbox, you provide a lot of adventure opportunities and give the players free choice; in a more linear campaign, you might provide only a few. #TODO moment-to-moment opportunity discussion

#TODO Steady State - throw a rock into still water and watch the ripples; throw more rocks in different places and the ripples collide; throw really big rocks and the colliding ripples become waves. Unleashing your players on a steady state is similar. This is most immediately visible in social scenarios or faction conflicts, where the players enter an unstable equilibrium and become, essentially, self-propelled cue balls.

#TODO These options are deeply intertwined - within any steady state there are problems to be dealt with and opportunities to pursue; within any opportunity, there are problems that must be overcome and steady states that can be engaged and disrupted, within any problem, there are opportunities to apply leverage or find workarounds and steady states #FIX that are either causing, threatened by, or already being disrupted by the problem. Each one can exist at all levels of play, from moment-to-moment decision making through a single iteration of the loop to campaign-level situations and goals.

I'll talk more about how and why to chain problems, opportunities, and steady states together on the fly in ways that create strong narrative momentum when we get to #What Follows the Result and #The Golden Path later in this chapter.

What Do You Want To Accomplish

This question is really about defining WHAT SUCCESS MEANS in context. When Allenby hits someone in the face, is she trying to provoke them? To knock them out? Specifically to break their nose and disfigure them? To snap them out of mind control? Something else? Every single one of those answers can have drastically different effects on the fiction, even if Allenby's actual behavior and action is identical across them all. The GOAL is what matters in this step, not the method.

Everybody Wants Something

This applies to your NPCs just as much as it does to the player characters, but it's worth remembering as more than just "the loop also works for NPC action." Everybody wants something - from the archvillain who wants to conquer the world to the stormtrooper who wants an excuse to crack some skulls to the wilderness explorer who wants to find a safe place to shelter for the night to the barista who just wants their shift to be over already and to never have to make another venti soy pumpkin spice latte with extra splenda and foam art in the shape of a trout again in their life. NPC goals are the pull to your players' push, and neither work well without the other to press against. Drama emerges largely in the places where multiple characters' goals push against each other, against the world, or against themselves.

The fiction is in constant motion, and so too are goals. Most of the time, what matters most is what a character wants right now - as far as the loop is concerned, that's the only thing that matters. We'll talk more about letting NPC goals drive the fiction forward when we get to step 5 and talk about integrating results and changes into the fiction. For now, remember: If a scene is starting to stall, look to the characters' goals as the spark plugs to restart the engine.

The Seven Dwarfs Method

If you need to sketch out an NPC in a hurry in the middle of the session, and can't figure out what this person wants in this moment, Taliesin Jaffe's "Seven Dwarfs Method" is a great jumpstart. Choose one of the seven dwarfs from Snow White - Bashful, Doc, Dopey, Grumpy, Happy, Sleepy, and Sneezy - as the defining adjective for the NPC. Each of these names stands alone and instantly evokes a single strong, defining trait that offers a clear cue for what the NPC might be pursuing right now.

If you want to further distinguish this NPC through performance, go to #Optional 3-Axis Acting.

How Do You Try To Accomplish It

Drill down. SWEAT THE DETAILS. What is the character actually, physically doing in this moment? We don't need Allenby to improvise a full paragraph every time she throws a punch, but we do need to know that she is punching, as opposed to kicking, or swinging a baseball bat, or a sword, or shooting a gun. And, IF IT COULD MATTER, we need to know how she's punching. Is she keeping her guard up and deploying tight, well-practiced jabs while leaving few openings for a counterattack? Is she throwing wild, untrained haymakers? Is she darting about and sneaking in pinpoint-accurate blows to joints, nerves, eyes, throats, and other weak points? Is she throwing caution to the wind and unleashing pure berserk aggression? Something else?

The extent to which detail matters depends on the rules you're using and the nature of the action. If you're playing in a system that abstracts combat more than this game does, then you don't really need to know whether a player is slashing or stabbing; it's the same roll either way. In contrast, if you're exploring an ancient lich's tomb, with intricate deathtraps around every corner, the minutiae can be a matter of life or death - are you lightly tapping the ground ahead of you with that pole, putting your weight on it, or something else? Are you pushing or pulling on this wall section you believe contains a secret door? Are you touching that altar with your bare hands? Which eye, exactly, do you use to peek into that hole in the wall?

"Or Something Else"

When prompting players for a decision, if you lay out multiple courses of action they could take, then wherever possible, add "or something else?" at the end, as seen above. This sounds small, but it's important. Offering the players a few options to choose from comes across as presenting a menu from which they may only choose something you've presented; adding "or something else?" leaves the door open for them to come up with something unexpected and preserves the intended effect - you're naming the most obvious (to you) actions available in this moment, not the only actions available.

Resolution

This step is broken out into two questions, since if the mechanics don't intervene to impose a result, then there's rarely any need to explain that result - the player's action either succeeds or fails precisely as they described it.

Do Mechanics Take Over

MOST OF THE TIME, the mechanics don't need to kick in for you to declare the result of an action. If there's no meaningful uncertainty or resistance in the fiction, then you shouldn't apply mechanics that inject that uncertainty or resistance artificially.

This applies in both directions - you shouldn't make the player roll for anything that has no chance of failure, or no chance of success. For more on impossible rolls, see The Limits of Rolling.

If there's no uncertainty for the mechanics to resolve, proceed to #What Follows the Result. Otherwise, follow the appropriate rules for whatever's happening, then move on to the next question.

What Explains The Result

If the mechanics took over after the previous question, then you need to explain the result that those mechanics imposed. Since success is often self-explanatory once the player has established their goal and their method, this typically takes the form of "what went wrong or changed about the scenario that kept the PC from achieving their goal with the stated method?"

As the GM, you should try to avoid narrating failure as a result of player character incompetence (though if a player chooses to lean into that for themselves, as with the Gimli example, let them). "You just screwed it up" often undermines the identity of the character and, therefore, part of the fun of playing them. Instead, it's best to figure out why the window for whatever was being attempted closed. Did something go wrong around the PC? Was it bad timing? Bad intel? Dumb luck? Something else? Or, if you're struggling to figure out something that fits, you can pass the narration back to the player and ask them to explain to the table what went wrong.

Besides preserving the fantasy of the player characters, finding other reasons why they failed often results in new sources of tension or drama in the scene - this way, failure becomes a new problem to deal with on the next run through the loop, fuelling the fiction instead of closing it off.

What Follows the Result

Your player has acted, the dice have spoken, and you've figured out everything that's happened leading up to and during the action. Now what?

In this step, you figure out what effects emerge from the results of the player's action. These effects take two forms: THEREFORE or BUT.

Therefore, Butts

The quote at the top of this section is fantastic advice for writers. The point is to ensure that there's always a clear flow of cause and effect through your story, giving it a propulsive momentum and sense of inevitability: This happens, THEREFORE our hero does something about it. BUT, something else gets in their way, THEREFORE the hero changes their plan to adjust for the new obstacle, THEREFORE they get what they wanted, BUT that angers someone else, THEREFORE the hero now has a new enemy. And so on.

This idea separates a story that makes sense and has a clear throughline pushing it forwards from a story that's more like a series of random events that happen to occur in sequence. There is, however, an obvious disconnect: this advice is meant for storytellers, and we're not storytellers, we're game masters. How are we supposed to use this advice without falling back on prepared plots?

Simple: Figure out how every action leads to a THEREFORE or a BUT, and bring that THEREFORE or BUT in before you restart the loop. This way, the world always reacts to the players' actions. Sometimes it pushes back against them; sometimes, the players do unexpectedly well and the world's opposition to them melts away like ice cream in front of a heat gun. This logic flow is how you keep things interesting.

Here are a few broad ways this can play out. This is not a complete list, but hopefully it's enough to help you start thinking about how this can come into play in your own games:

• The player succeeded and accomplished what they set out to do, THEREFORE they've changed the situation, BUT that changed situation comes with its own built-in set of obstacles, THEREFORE there are new decisions to make.
• The player succeeded, BUT something else went wrong in the process of accomplishing their goal, THEREFORE they now have a new problem.
• The player would have succeeded, BUT something went wrong and is standing between them and their objective, THEREFORE they now need to deal with whatever went wrong before they can achieve their original objective.
• The player failed, BUT a new opportunity opens up, THEREFORE there's still a path forward, BUT it'll be more difficult in some way than their original idea.
• The player failed, THEREFORE their original goal is completely cut off, THEREFORE they need to figure out something completely different.

"Dramatically Appropriate" Is Valid

Not all complications or other effects of a result have to stem from something already in the scene. Often, the most interesting option available is inserting something new into the scene. This can be something from your prep that makes sense to bring in here, or an entry from a random table that turns out to be just now rounding a corner, or something you improvised on the spot because the dice said something had to go wrong.

Sometimes, the effect of a result might not be felt immediately within the scene. It might make sense to progress something happening in the background, like ticking a clock to decrease the remaining time until reinforcements show up or #TODO

Fuzzy Logic

The boundary between "What Explains The Result" and "What Follows The Result" is occasionally fuzzy. Sometimes - usually when mechanics taking over results in a failure - the "BUT" is contained entirely within step 4B, and leads so directly to the "OK, now what?" question and integration that there's nothing left to add. This is fine; nobody's going to come take your GM privileges away from you for not dogmatically adhering to the precise structure one indie writer came up with to explain what most GMs are already doing.

Integration

Now that you know what's changed at the end of the loop, it's time to put it all together and start the next cycle. It's important to re-center the narration on the key facts influencing the next decision, especially in complex or chaotic situations. If the change is subtle or the situation is relatively stable, it should be fairly easy for everyone to keep track of the scene state, so you don't need to belabor things that aren't obvious factors in the next decision. However, if there's a lot of moving parts in the scene, then it's best to quickly re-establish the current state of all those parts for any decision where they're relevant.

The Fader Board

"Fader" is the proper name for the large slider knobs you see on a sound mixing board. Throughout a concert, a sound engineer constantly adjusts - "rides" - the faders to make sure the music as a whole sounds the way it should. Sometimes these adjustments are drastic, sometimes they're subtle. But together, they're the central control board for the entire experience of the show.

Every decision you make while running a game is, in some way, riding the faders - if you've ever run even a single session, then by definition, you already have experience with this. You've made judgement calls, decided how your NPCs act, described your world in your words. Different fader-riding choices are why first-time GM A's Lost Mine of Flapdoodle can feel completely different from first time GM B's Lost Mine of Flapdoodle, even though they're both using the same source material. This is where all the micro-decisions that collectively form "your style" live - it's the bucket containing all the things that make your game uniquely yours.

So I'm not going to pretend I can tell you "this is the correct way to ride the faders at your table." I'm just going to do my best to explain how I do it.

Everything Is A Fader

Faders I commonly reach for include, but are not limited to: tone, mood, atmosphere, consequence type and severity, difficulty, environment design, nature of challenge, npc design and behavior, player spotlight, pressure, speed of play, stakes, transparency, vocal performance. I can't speak to every fader, since again, every single decision you make at the table counts as fader-riding in some way, but I can elaborate on some of the more frequently-needed and impactful ones on this list.

All of the prior advice in this chapter has already been about how to do this within each step of the loop. What follows is valid no matter where in the loop you are, and can also apply across multiple steps and iterations. It's important to understand that none of these are completely independent variables - adjusting one of them inevitably has some kind of effect on most, if not all, of the others. This also is far from a complete list - again, these faders are just how I conceptualize what I'm doing when at the table. Yours might vary wildly.

Tone, Atmosphere, & Mood

TONE is the thing you'll be adjusting the least often, but also the most important fader to keep in mind - it's a filter that will affect every subsequent decision you make. Tone exists at the highest level and defines the feel of your entire campaign. It can and should change over the course of your game, sometimes quite drastically, but it tends to change slowly, through a series of small adjustments over time.

Instances where something radically changes the tone of the entire narrative in a single moment are enormously impactful, and should therefore be executed very rarely and with caution (I recommend only doing this once in a campaign, if you do it at all - most campaigns don't call for or benefit from this kind of seismic shift).

MOOD is how you want your players to feel right now. You'll modulate this more actively on a scene-by-scene basis. Most scenes will maintain a single dominant mood throughout (e.g. funny, relaxed, gloomy, intense, terrifying) with only minor adjustments, but mood can also shift drastically within a scene (e.g. surprise reinforcements instantly turn a battle from a grim last stand into a triumphant victory; a surprise reveal recontextualizes a lighthearted scene as something horrific).

ATMOSPHERE is related but distinct. It's the external expression of mood. This is where a lot of your moment-to-moment execution lives. It primarily derives from the wording and texture of your narration, but also includes things like character voices, music choice, handouts, and ambient lighting, if you're into that (personally, I enjoy doing voices and need music at the table to put myself in the right headspace and dedicate a lot of time and effort to picking out the perfect tracks, but for most people that's complete overkill - and I play primarily online anyway so lighting and handouts are a non-factor).

#TODO Modulation, Pacing, Story structure

I recommend Hamlet's Hit Points by Robin Laws as further reading. In it, Laws breaks down a selection of popular films in depth to learn how they modulate mood and atmosphere using sequences of "up beats" and "down beats," and explains how these techniques can be applied at the game table. The advice he gives works well, even when those beats are often up to the whims of the dice.

Optional: 3-Axis Acting

I am not an acting coach or a director, so I'll parrot someone who is - once again, Taliesin Jaffe. Once you have an NPC's personality and immediate goals worked out (see #The Seven Dwarfs Method in a pinch), if you'd like to go the extra mile and inhabit the character as performance, the easiest way to distinguish them is by making two or three snap judgement calls, each of which can be High, Medium, or Low.

First, choose a posture - the actual, physical posture that you hold your body in while performing the character. High posture - straighten your back; raise your chin. Medium posture - however you normally hold yourself. Low posture -hunched over or curled inwards. Again, you should actually change your body language and adopt the posture once you've selected it. If you're not used to it, it might sound stupid, but I promise it works.

Second, choose a pitch. Again, High, Medium, or Low. Combine this with posture and you have nine starting points for your performance; combine those nine points with the Seven Dwarfs method and you have 63 base performance templates ready at a moment's notice.

If you'd like, you can add an even-more-optional third axis: sophistication. Once again, High, Medium, Low. This primarily affects word choice, but that can have an enormous effect on how the character comes across. A rich, successful, and somewhat snobby lawyer might be High posture (arrogant and snooty), Medium pitch, High sophistication (well-educated); that lawyer's mob enforcer sibling might be High posture (Also arrogant, less snooty, more I-can-kick-your-ass), Medium voice, Low sophistication (speaks in street slang and didn't get the scholarship their sibling did). Combine these with all of the above, and your template choices jump from 63 to a whopping 189.

Pacing, Speed, & Dynamic Range

#TODO Pacing = campaign level, adventure session level, scene level. Information / detail density per loop iteration. Rapid pacing = low density, slow pacing = high density. Speed = real-time duration per loop iteration. Longer descriptions, "uuuuh sorry still figuring my turn out" = low speed. Shorter descriptions, talking fast, attentive and on-the-ball players = higher speed.

Story Structures

#TODO This is also where the story comes from. Put the pitchforks down, I know how I opened this chapter, and I'm not saying to script your campaign - the story is still something that emerges through play, not something you bring. But the central driver of any good story is conflict, and as the person at the table most responsible for generating that conflict, it's impossible not to have a finger on the scale in some way. Think of your campaign as a home cooked stew - you don't necessarily pick the recipe, you don't necessarily even have a recipe beyond "pick ingredients, add and remove from the pot as needed, and manage stove heat," but you do get to decide what's in the pantry to work with. Even if you run a purely reactive exploration-based campaign with no big active threat in the world, like many "old-school" sandbox games encourage, that's a decision you're making to center the story entirely on how the players go out into the world, see exotic places, meet interesting people, and probably rob them blind and/or kill them.

#TODO As with Therefore/But, we can adapt storytelling techniques for use in our games.

#TODO To the OSR community - people, please, I beg you, chill the fuck out you are calcifying good practice into dogma. Ravenloft came out in 1983. You can have a completely player-driven sandbox that still has a clear end goal, escalating conflict, prominent / dominant NPCs, high drama, and a satisfying climax.

#TODO I could write another several thousand words on the western Three-Act Structure framework, the eastern Ki-Sho-Ten-Ketsu cycle, the areas of overlap between them, and how you can use them in your prep, but that's a bit out of scope. Will write an appendix containing a breakdown of both concepts, especially the fractal nature of ki-sho-ten-ketsu and why it works so well with the loop and improvisational structures, through a deep dive into a master class in the concept, Kiki's Delivery Service.

Dynamic Range: The Quiet Around The Thunderclap

Let's return to the fader board analogy for a moment, since audio engineering terms are once again relevant to how you run your game. DYNAMIC RANGE is the difference between the largest and smallest amounts of a thing. In music, that means the difference between the loudest and quietest parts of a song. Ever since the mid-'90s, music nerds have been complaining about a phenomenon called the "Loudness War," referring to the way most popular music has gotten louder and louder over the years at the expense of dynamic range. The idea is that, since louder song feels more exciting and energetic, you can make the song feel better to listen to by making a greater percentage of it loud.

Why not just make everything louder or turn the volume up on the listener's speakers? Simple: within a production environment, there is an upper limit to the volume of a sound. Past that limit, bad things happen to the audio - it breaks up and distorts in a way that's almost universally considered unpleasant. And at the listener's end, even powerful speakers have their own maximums. So instead, the solution is to reduce the dynamic range of the music. Lower the volume of the loudest spikes so you're no longer hitting the maximums, then increase the volume of everything else until you get back to the limit.

In other words, you don't just make something louder. You make it flatter. Your spikes are less spiky, and they stand out less. And when you take this too far, the music loses any sense of variety and contrast and just dissolves into a wall of noise. Because as it turns out, the quiet parts are really important. They're what let the loud parts feel loud.

The same concept applies at your table. If everything is at full intensity all the time, then it stops feeling intense - your players adjust to that as the baseline,, and the lack of variety generates fatigue, not excitement. When everything is intense, nothing feels intense. You need those quiet parts in your game, to contrast with the moments of high tension, adrenaline, chaos, hilarity, or whatever other feeling you want to create when it's at full blast.

I go into this in much more depth in Appendix - Shiny & Chrome, but it's worth studying Mad Max: Fury Road as an absolute masterclass in the manipulation of dynamic range across every creative discipline. When you're watching it, it feels like nonstop intensity. In reality, when you look closely, it's constantly spiking the intensity up and then pulling it back down again, allowing it to reset the audience to a baseline of relative quiet before it hits you with the next spike. That pullback is absolutely critical to the film's success - it's what lets each intense moment stand out as a discrete event. The feeling of constant intensity comes from the way it places a lot of those discrete events in rapid succession. If it were actually maximum volume all the time, the film would have been an incomprehensible mess where nothing stuck out because every intense moment blurred together with the other intense moments that came before and after. The thing that makes it one of the greatest action films ever made is its restraint.

#TODO Breathing room and Ma (active stillness). Related concepts, but distinct. Contrast matters. Surround your big moments with small ones. Slow the fuck down before and after something important happens to give it weight.

#TODO What the fuck I put this section here and within 24 hours a video conversation with Brennan and Matt drops where they talk about this same thing in context of encounter difficulty. Address that.

I have been running games for fifteen years at the time of writing this. I can clearly remember every instance during that time where I've decided "Do I really want to turn things all the way up right now?... yes, yes I do."

It has happened exactly once.

Pressure = Stakes × Complexity × Constraints

#TODO Stakes = scene stakes and dramatic question, why are we spending table time on this instead of something else. Complexity = number of moving parts in and surrounding the scene. Constraints = limits on effective player actions (there are always less of these than anyone thinks).

Progress Clocks

I'm about to spend a lot of words explaining why numeric progress tracks are a good idea and how to use them. This may sound obvious; it isn't. Clocks are among your most powerful tools for managing pressure, especially in systems that mandate complications on certain rolls (as opposed to just suggesting them, like this game usually does).

Blades In The Dark, which popularized the concept, suggests using clocks with no less than 4 and no more than 12 segments, with 8 as the go-to if you need to just pick a number and move on. Blades makes heavy use of clocks for many of its systems, but the most common and most relevant rules are:

• When the players don't want a clock to fill or empty, a mild complication increments it by 1 tick in whichever direction they're trying to avoid, a moderate complication increments it by 2, and a severe complication increments it by 3.
• When the players are trying to increase or decrease a clock, a failure increments it by 1 tick in the direction they want, a complicated or partial success increments it by 2, a clean success increments it by 3, and a critical success (success with style in this game) increments it by 5.

These rules work well in Blades, which features multiple degrees of complication severity and success baked into its ruleset (and can have any severity of consequence appear even on a complicated success in the right conditions). If your game is closer to a binary pass/fail system, or uses a different method of handling complications, you'll need to adjust them to find an implementation that works for your table.

There's some debate over whether or not clocks should be player-facing; I take a very strong stance of "it depends." Players seeing the clock is great for tension - it becomes a SWORD OF DAMOCLES hanging overhead, where they can see the rope holding it in place wearing away and getting closer and closer to snapping. Or, if the players actually want to fill the clock, it becomes a clear PROGRESS MARKER that lets them know exactly how close they are to their goal. This works well in situations that have obvious known time limits, stealth operations where the players' actions bring them closer and closer to detection, battles where the players are trying to escape before enemy reinforcements arrive (or hold out long enough for reinforcements of their own to turn the tide), scenarios where there's more options available than time available to complete them in and prioritization is required, and so on.

On the other hand, there's an argument to be made for players not seeing the clock in cases where their decision gets MORE INTERESTING (not just harder) with incomplete information. This works well for cases where it's obvious that tension is building up and something dramatic is about to happen, but nobody knows what will set off the powder keg or how long before it blows. For example, a boss charging up a devastating attack, a mob boss running out of patience during a tense standoff, or a villain working slowly behind the scenes and advancing a plan that, when it fires, will completely upend the overall situation of your campaign.

It's important to remember that CLOCKS GO BOTH WAYS - they should be another element of player decision making, not just a ticking doom timer. Players should be able to push back against dangerous clocks when it makes sense in the fiction. Going back through some of our previous examples, methods for the players to shift the clocks in their favor might include making a brief alarm activation seem like a false positive to lower suspicion, clearing out a wave of enemies efficiently enough that it creates a window to patch themselves up, and placating the mob boss's temper before he has someone demonstrate what hedge trimmers do to the human body.

#TODO Players should also be able to start clocks of their own to accomplish long-term or complex objectives.

Tick Tick Tick Tick Boom

#TODO sometimes advance per scene, sometimes advance with in-fiction time, sometimes advance only when there's a complication, sometimes when the players put in work to advance one, sometimes all, sometimes none

In games that have mandatory complications on some rolls, clocks can serve as a hidden way for GMs to bleed off excess pressure. When the dice demand something go wrong, but the scene, session, or even the game as a whole would suffer from a complication happening right now, you can defer that complication by starting or ticking a clock. #TODO build up to pressure cascade later, clocks are discretionary and if you don't really want to you don't have to ever advance ones that aren't serving a purpose, also works if you just can't think of anything in the moment and need to kick the can down the road. But do this sparingly - if your players learn not to expect negative clocks to actually fill, they cease to be a meaningful complication and become "the GM is letting us off easy again."

Difficulty

#TODO multiple types of difficulty that have nothing to do with each other. Setting difficulty numbers and opposition targets is a system-dependent GM-facing math problem - how often should the players be succeeding or failing? Other type is fictional difficulty - how many obstacles are standing between the players and their goal? These are separate faders. You can have a scenario with very high mechanical difficulty and relatively low fictional difficulty (e.g. a gladiator fight where all you need to do is sign up and win). Or, you can have the opposite, a situation with little mechanical challenge that still requires extensive problem-solving and the completion of multiple steps to achieve the players' goal (e.g. riddles, puzzles, going to the DMV.)

The Golden Path

"Golden path" is a game design term for the absolute MINIMUM REQUIRED STEPS to complete an objective. These steps flow out of the fiction as a natural extension of whatever objective the players are currently working towards.

This isn't the same thing as planning a story sequence the players must follow. When you determine a golden path, you're not trying to decide what the players will do. Instead, you're examining the situation to figure out how it answers this question: "if the players decide to do this, what steps would it take for them to succeed?"

It doesn't matter how these steps are dealt with, only that they are dealt with. The story of your game comes from your players supplying the "how." Often, that "how" will come with its own prerequisites. Through this process of solutions to existing problems generating new sub-problems, a single objective can become the center of an entire complex scenario that's still driven by proactive player action.

It's important to think through whether something is truly a REQUIRED PREREQUISITE for the fictional objective, or just an OPTIONAL OBSTACLE - some things that might sound required at first glance actually turn out to be optional when you examine them closely. For example, if the players want to steal a painting, and the painting is currently hanging on the wall of a museum and protected by an advanced security system, then the "get into the museum" is a necessary prerequisite for "steal the painting." It doesn't matter how the players get into the museum, but unless they have some way to get at the painting without being anywhere near it, they do need to get into the museum somehow, so that's part of the golden path. "Deal with the security system," however, is not a mandatory step - the players can steal the painting without addressing the security; that just means the security will activate and they'll have to suffer the consequences.

If you realize the situation you're setting up lacks enough prerequisites to be interesting, you can adjust the unestablished facts to add more. Continuing the painting example, you could decide that instead of being on the wall in plain view and ripe for the plucking, the painting is currently in the museum's secure archives, and it's in some kind of crate nestled among many other crates and containers. Now, the prerequisites have expanded from "get into the museum, steal the painting, and get out" to "get into the museum, get into the archives, locate the painting inside the archives, get the painting, get out." Notably, the order of this is as flexible as the methods are: there are any number of ways the players could locate the painting in the archives, many of which could theoretically be completed well before actually entering the museum to perform the heist, or after entering the museum but before breaching the vault itself. You can also insert additional optional obstacles the same way - you'll get a sense for your preferred balance of how tough to make these obstacles and how many to use as you get more experience.

Sometimes, the players will surprise you with creative solutions that bypass prerequisites you didn't expect to be bypassable. ENCOURAGE THIS. If they pull this off, then an easier task is their reward for engaging with the fiction deeply, treating it as true, and creatively using the tools you've given them - they've earned it. For me, these moments of unexpected player brilliance are the some of the greatest joys I ever experience behind the GM screen, and a huge part of why I love running RPGs.

Transparency

#TODO How much do the players know about the situation and the factors influencing / effects of their decision? Maximum transparency: full explanation of everything going on and what will happen if you do X before you commit to X. Minimum transparency: Tomb of Horrors.