GM fiat - an illustration

[soviet]

I don't think these questions are at all reasonable. You are just dissecting ideas most gamers will easily grasp (and you are adding things in, like agenda) that are separate considerations and concepts: and I don't think the core books should get into GNS at all in that respect---don't find agenda a useful idea). This is just one example of the kind of language. I signed off on it because someone raised it as an alternative to a more involved lexicon. I don't think ti would be the entirety of the break down. But I do think most people know when a GM says 70 percent combat 30 percent roleplaying, that means you are in a campaign with a lot more focus on combat than other aspects of play.

I think taking 70 percent combat, 30 percent role-play to mean you don't role-play during combat is not how most people would interpret this language, and I don't think it requires we set down rigid definitions of terms. Most people get this kind of language.

I think you are exaggerating how hard this is to understand, and minimizing how difficult it is to understand some of the language being floated in this thread.

I wasn't talking about creative agenda in a GNS sense, just agenda as is commonly understood. What's the point of play? What are we as a group of people seeking to do?

It seems to me that you are unable to describe how you play beyond woolly platitudes that describe everyone's play, or no-one's. When more specific descriptions are presented you reject them.

 

[Bedrockgames]

I wasn't talking about creative agenda in a GNS sense, just agenda as is commonly understood. What's the point of play? What are we as a group of people seeking to do?

It seems to me that you are unable to describe how you play beyond woolly platitudes that describe everyone's play, or no-one's. When more specific descriptions are presented you reject them.

I don't know @soviet, I think I have given a pretty clear idea of how I play beyond platitudes.

 

[Crimson Longinus]

The crux of the point is the anti-black box dynamics of the game engine (which includes structure, always-on and always- stable resolution procedures, transparency, overt principles, robust currencies that are always reliable, as well as "playing the fiction") of Torchbearer fundamentally shuts down the paradigm presented above. It has never happened (and by god have I GMed a metric eff-ton of TB) because it fundamentally_cannot_happen. The gameable space is always robust and always stable, sequence-in sequence-out.

Yes, it indeed could happen that a GM overrides a rule in manner you do not like. Now in a game where everyone is on the same page this is unlikely to happen, but still. But what it also means, that the GM can override a rule to allow your character to do something that would make sense in the fiction, but not by the rules. Granted, ideally games are built so that such disconnects do not commonly happen, but there is no such thing as perfection.

In any case, I am not saying that consistent rules are bad thing, but that gameable space doesn't rely on the rules. We can have gameable space purely via fiction.

 

[robertsconley]

@pemerton
This conversation started with our differing comparisons between the Alarm spell and Aetherial Premonition, my analysis was rooted in how I run campaigns using World in Motion alongside classic D&D and D&D 5e rules. As a result, the discussion naturally broadened into an exploration of the differences between Torchbearer and World in Motion play, differences that underscore the distinct ways Alarm and Aetherial Premonition function in their respective systems.

Throughout this discussion, I’ve argued that the key distinction between Torchbearer 2e and World in Motion-style play lies not in whether prep exists, but in how resolution interacts with established circumstances. In World in Motion play, events, factions, and geography are already in motion before the referee calls for a roll. The purpose of resolution is to determine whether the players intersect with those elements, not to create the event from scratch.

In contrast, Torchbearer often uses resolution-first procedures, most notably the Camp Event and Town Event tables, where the system mandates that something happens, and only afterward is the fiction constructed to fit the result. While Torchbearer includes prep tools and setting elements, they are not what directly drive those procedures. The system assumes adversity will occur and uses mechanics to create it.

This results in a fundamentally different structure of play: Torchbearer puts thematic tone and systemic tension into the foreground, while World in Motion prioritizes world continuity and causality. That structural difference, not the mere presence of random tables or factions, is what I’ve emphasized throughout this thread.

I think we’ve each clearly marked out our positions. For anyone reading along, I trust they can draw their own conclusions from the information we’ve both provided.

I'm certainly willing to accept that tone, and the themes that tone can support, aren't important to you in the way you present the setting to the players. In practice, I would expect that to produce play in which the PCs are largely amoral, and mostly motivated by instrumental concerns.

That is a good comment, which I will reply to in a separate post and I will also reply to your comment on Middle Earth as well.

 

[FrogReaver]

Tagging you here @FrogReaver .

One GM might determine that the game theory deployed by the peasant class vs law enforcement or the military class (LEoMC) would be to renege on promised cover to the Folk Hero should they be pressed by the LEoMC because of the "not risk their lives for you" clause and then they would black-box in the LEoMC pressing itself via some black-box decision-making/scheme); extrapolation.

Another GM might determine that the LEoMC would either have already bribed the peasant class to turn in the Folk Hero or will do so now (deploying the GM's infinite reservoir of black-box currency to facilitate this); extrapolation.

Another GM might determine that adventurers are inherently dangerous, thereby initiating the "shown yourself to be a danger to them" clause, thus unknowingly to the player crippling the Rustic Hospitality Trait at the PC build stage before play even begins (!); extrapolation.

Another GM might decide this is the perfect opportunity to mainline an encounter onto play where some essential, GM Storyteller plot node with the LEoMC is made manifest; GM Storyteller.

5 x other GMs might decide that it doesn't work by player fiat, but rather to devise varying in-situ, black-box fortune resolution mechanics to randomize the results with a spread from 20/33/50/67/85 % chance to resolve in @hawkeyefan 's favor; behind-the-screen randomization with GM-decided upon odds in which neither PC build nor currency muster interacts with.

5 x other GMs might decide that it doesn't work by player fiat, but rather the PC in question has to make a Cha (Persuasion) roll with widely varying DCs; actual action resolution, but who knows what DC is being set + who knows if this player has invested any PC build resources into Cha or Persuasion to facilitate this + how is the player to know that they should save an Advantage for such black box action resolution which is almost surely on the tail-end of other action resolution to get to the farmer's barn in the first place (some form of chase or traversal/exploration challenge).

* Circling back a hair, again this sequence at the farmer's house/barn would be on the tail-end of a decision to make a runner for that farm in the first place. If the goal is "achieve safe haven from danger" and the initial action declaration is "let's make a run across town for the farmlands...the peasants there will shield us from the law searching for us (it says it right there in the fine print!)." But between that decision to make that runner is however many other stages of action resolution/"playing the fiction" (who knows how many?). Maybe players spend currency (Advantages, Spells) or gain attrition in the form of Hit Point loss and gained Exhaustion Levels...all for a plan that was either (a) a dead-end in the first place before they even set out or (b) an unknown/unknowable low % or remote possibility in the first place? If they had actually not been working under the dynamics of a black-box, they would have made an entirely different strategic decision (!); "hey, forget the runner across town to the farmers...they'll might/most likely/definitely will give us up to the law...let's just either (i) pick a defensible position and rain hell on the LEoMC or (ii) make a runner to the forest and attempt to disappear under its dark/haunted boughs (and yet again...who knows if the GM will have them pursue into the dark/haunted forest and how that black box will resolve)."

---

The crux of the point is the anti-black box dynamics of the game engine (which includes structure, always-on and always- stable resolution procedures, transparency, overt principles, robust currencies that are always reliable, as well as "playing the fiction") of Torchbearer fundamentally shuts down the paradigm presented above. It has never happened (and by god have I GMed a metric eff-ton of TB) because it fundamentally_cannot_happen. The gameable space is always robust and always stable, sequence-in sequence-out.

IMO the crux of this issue is treating the gameable space of 5e D&D as having some kind of reliable, "I can hide out with commoners no matter the circumstance" ability. That's not an ability in the gameable space of 5e D&D. What 5e does offer in it's gameable space is a reliable ability to hide out with commoners provided they they don't need to risk their lives and you haven't shown yourself a danger to them.

5e D&D rules tell us exactly how to resolve the rustic hospitality ability. It's just not a very great ability in the system because the kind of fictional people that the PC's would typically be hiding out from in most campaigns wouldn't hesitate to use violence against those hiding them. I'd even go as far as saying that it's nearly approaching trap option territory, but I wouldn't call it a complete trap because I think there's a few niche use cases where it can still be useful.

 

[robertsconley]

How do you track all this? Is it by faction? By region? If you do a situation report to update things based on the last few sessions, do you just list changes and reference the appropriate game elements (nations, organizations, NPCs, etc.)?

All of the above and none of them. I adapt the content of the report to what the situation requires. The basic principles I follow to decide what to include are not unlike those the military uses for sitreps.

One or more of the following.
1) Who is involved
2) How are they organized
3) What resources exist
4) What has happened
5) Upcoming plans
6) Where is this taking place.

Is this all in a spiral notebook or a word doc or what have you?

Spiral Notebooks and binders up until 1990, word processing files since then. I was sold on word processors after I used a late 80s UNIX office suite in my last two years of college.

For example some of my Majestic Wilderlands folders and files.

Some of the oldest files in my archive

I don’t think there is a big difference between the two, honestly. I think they are both true.

Again… what’s plausible in most instances allows for a range of outcomes. So you narrow it down to the most plausible options… let’s say three, and then you decide which of those three it will be… you are absolutely directing the outcome.

And… you’re basing your extrapolations… your consideration of what’s plausible… based on background details and other setting information that you’ve decided in the first place. So your extrapolations are you considering what decisions to make about the decisions you’ve already made.

That’s a lot of GM control going on. Again, this may be good or bad depending on expectations and desires for play… but it seems very GM driven and focused.

Yes, the referee decides what is plausible; however, plausibility is a standard that players can use in a discussion with the referee outside of the campaign. Just as an author of an alternate history story decides what is plausible, but is ultimately judged by their readers. For referees, like myself, using World In Motion this is an important part of the process of how we improve our campaigns over time.

During the campaign, there is only so much I am willing to talk about. Afterwards, I generally have a debriefing, for lack of a better term, where players and I ask questions of each other about the events of the campaign and why they happened. Sometimes this also happens during the campaign, but it's rare because nobody wants spoilers about things their character wouldn't know.

Since you and I have talked about these issues before, I would like to be clear about something here. Over the history of the hobby, there has been a lot of criticism of GM fiat, and a great deal of creative energy has gone into developing systems to address those concerns. What I don’t see enough of is discussion about how to teach people to become better referees that doesn't rely on systems for RPGs.

That is where I focus my efforts. I try to help people improve as referees within the time they have for a hobby, and I aim to give them practical tools to support that process. I am not interested in debating the merits of system-first versus person-first approaches. My approach is person-first.

I am always happy to explain the details of what I do when people ask specific questions, as you have been.

 

[robertsconley]

I'm certainly willing to accept that tone, and the themes that tone can support, aren't important to you in the way you present the setting to the players. In practice, I would expect that to produce play in which the PCs are largely amoral, and mostly motivated by instrumental concerns.

These things are not in contrast. In other threads you've talked about playing a Middle Earth game: JRRT's setting is 100% intended to have a certain tone. That doesn't meant that there is no presentation of a world.

That’s a fair concern, but it doesn’t match how World in Motion play actually unfolds at my table.

First, I strongly emphasize first-person roleplaying. My players don’t interact with NPCs as abstract entities; they treat them as real people with motives, cultural beliefs, and histories. Most of the campaign’s tension and drama arise through these interactions. As a result, players naturally engage with moral and social concerns, whether through loyalty, faith, ambition, or survival, not because the system enforces them but because the world reacts to those decisions meaningfully.

Religion, culture, and personal history aren’t narrative backdrops, they shape outcomes. When players act in line with their character’s culture or faith, they often find new allies or open up opportunities they wouldn’t get through pure pragmatism. And when they violate those norms, consequences follow. So moral play does emerge, but it’s driven by circumstance, not theme.

As for Middle-earth: I didn’t want to run a campaign that felt like another Tolkien novel. I wanted to run a campaign that felt like being in Middle-earth. What made using World in Motion possible was how the Adventures in Middle-earth line handled setting detail, especially the Shadow. It’s not a narrative arc or mood dial its a supernatural corruption rooted in the metaphysics of the world, the Marring of Arda by Melkor. Like sanity in Call of Cthulhu, it emerges from how the setting works, not from narrative tone.

Likewise, their audience system, while not something I used mechanically, was invaluable as a guide to what different cultures and NPCs value. Because of that, my players' decisions in Middle-earth campaigns weren’t just tactical, they were moral and cultural. The system didn’t need to push them to act a certain way; the world made their actions matter.

 

[robertsconley]

The crux of the point is the anti-black box dynamics of the game engine (which includes structure, always-on and always- stable resolution procedures, transparency, overt principles, robust currencies that are always reliable, as well as "playing the fiction") of Torchbearer fundamentally shuts down the paradigm presented above. It has never happened (and by god have I GMed a metric eff-ton of TB) because it fundamentally_cannot_happen. The gameable space is always robust and always stable, sequence-in sequence-out.

Sorry to collapse your post, I know you raised several points and gave several example. However, after reading, I can see the issue you talk boil down to a single issue.

Failure of Initial Context.

What is the initial context?
From my Basic Rules for the Majestic Fantasy RPG

The Initial Context
One of the hard things about a campaign using these rules is what the players do at first. It is important to think about the initial context—the situation the players find their characters in when play commences. Sometimes the players are notably self-driven and the group has a specific idea of what they want to do. In these cases, the initial context can be minimal.

In most cases, the players will be unsure of the possibilities, so it is best to have three to five rumors, pieces of lore, or contacts prepared so the players have a choice of where to adventure. It is important that anything vital they would know is written up in a small handout and made available. Keep this handout as minimal as you can while still covering everything you deem important.

From my blog.

Initial Context
Most sandbox campaigns fail. Why? Because of the lack of a good initial context. Many mock character histories and background but if you going to get a sandbox campaign you are going to need a least a half page of specifics for each players and a half page of general information for the group as a whole.

Players who enjoy being plunked down in the middle of a blank map and told "Go forth and explore" are few and far between. About as common as players who enjoy playing GURPS with all the options in play at once. Most players want to feel their choices have meaning. Picking one of the six surrounding blank hexes is not a choice with meaning. So work on the initial situation so that it is interesting and give the players enough information to make some valid decision of what to do.

Torchbearer works well because it establishes its initial context clearly and reinforces it constantly. Right from the first page, it lays out the social position and assumptions about adventurers:

Adventurer is a dirty word. You’re a scoundrel, a villain, a wastrel, a vagabond, a criminal, a sword-for-hire, a cutthroat. Respectable people belong to guilds or the church or are born into nobility. Or barring all that, they’re salt of the earth and till the land for the rest of us.

Your problem is that you’re none of that. You’re a third child or worse. You can’t get into a guild—too many apprentices already. You’re sure as hell not nobility—even if you were, your older brothers and sisters have soaked up the inheritance. And if you’re cursed with visions from the Immortals, the temples won’t take you in. You question their authority and subvert their power, so you are outcast like the rest of us.

Players know from the start where they stand in the world, what people think of them, and what sorts of problems and responses to expect. That clarity makes the game playable without relying on the GM to interpret every situation in the moment.

But this isn’t because Torchbearer is “better” in some universal sense. It’s because the authors baked in the kind of initial context that I advocate for as a referee. You don’t need Torchbearer to do that, you can build it into any campaign or system.

The lesson here, whether you’re a referee or a game designer, is to explicitly spell out the assumptions and expectations of your setting. That’s how you avoid black-box adjudication. That’s how you enable meaningful player choice. And that’s how you keep the gameplay grounded in a world that’s consistent and knowable.

 

[pemerton]

In World in Motion play, events, factions, and geography are already in motion before the referee calls for a roll. The purpose of resolution is to determine whether the players intersect with those elements, not to create the event from scratch.

In Torchbearer 2e, events, factions and geography are already in motion before a Camp Events roll is made. This is how the GM determines the danger level.

Multiple examples have been given upthread, by @Manbearcat and in my quotes of my actual play reports.

In contrast, Torchbearer often uses resolution-first procedures, most notably the Camp Event and Town Event tables, where the system mandates that something happens, and only afterward is the fiction constructed to fit the result. While Torchbearer includes prep tools and setting elements, they are not what directly drive those procedures. The system assumes adversity will occur and uses mechanics to create it.

You continue to assert this. But you have not actually provided any account of how a Camp Event roll is different from a random encounter/event roll in classic D&D play.

For instance, the system does not assume that adversity will occur. I've posted two examples now: the 19 entry for Wilderness Camp Events; and the example of the Town Event from my actual play, where the PCs returned to town to learn that the alchemist they had sold stirges to had died of stirge poisoning while they were travelling through the foothills south.

 

[Manbearcat]

You continue to assert this. But you have not actually provided any account of how a Camp Event roll is different from a random encounter/event roll in classic D&D play.

For instance, the system does not assume that adversity will occur. I've posted two examples now: the 19 entry for Wilderness Camp Events; and the example of the Town Event from my actual play, where the PCs returned to town to learn that the alchemist they had sold stirges to had died of stirge poisoning while they were travelling through the foothills south.

I don't have time to engage with @robertsconley 's response to me, nor CL's, nor FR's. One thing I will say off the bat is that I'm very much of the opinion that (a) looking at what Mouse Guard does and then (b) examining the myriad ways Torchbearer diverges from MG to produce a very novel play experience is at least as fruitful (likely more) as comparing TB to D&D.

But for the record (as you know), the following are very much true:

* Like Mouse Guard, the setting (events, factions, geography) of TB very much generates an "active actor (in motion)" quality to play through the synthesis of map-making, Adventure design, Events tables, test-borne Twists (particularly conflicts and their resolution), and how players interact with the map and the relevant NPCs (particularly, their relevant NPCs either character-built or accumulated).

* Torchbearer isn't like PBtA or FitD in the way those two games rely upon the cascade of snowballing move resolution. Yes, TB play certainly bears out a duress-filled "spinning plates" quality of play via its interlocking parts, but it doesn't assume adversity in any given, consequential chunk of play. Of course, neither Town phase nor Camp phase features The Grind. The actual play I cited above was from many years ago. From recollection (I'm pretty much 100 % I'm right), that Camp Events table was 3d6+1 (-3 +4 = +1) which yielded an 11. An 11 on the Wilderness Camp table (I had a Bleakwood Table which used most of the Wilderness table but a few custom entries) = safe camp result. Further, the checks spent in that camp phase didn't yield any Twists. In fact, I'm pretty sure all the checks spent on Recovery were successful as well. I think it was a clean sweep of success in that Camp phase.

Finally, I've certainly GMed my share of Town phases where a gentle Halloo (from your neighbors) or Pleasant Smells from woodsmoke and kitchens or better hit on the Town Events tables + accommodations for recovery were able to be paid for via Resources + a PC stays at their Parents Home + a few simple Lifestyle moves to Build Kit and Do Research didn't turn up any trouble + Paying Your Bills went off without a hitch when town was left.

If I had to Scientific Wild Ass Guess it, my collective Camp phases probably turn up major trouble (from Events or significant Twists) at a remote clip of inside of 15 % and minor trouble in the form of some kind of a small Camp Event hit or a consequential Twist perhaps 30 % of the time (typically the worst part of Camp is a gut-punch failed recovery test on a nagging condition). Towns are probably double both of those; 30 and 60.