The Role of GM Fiat?

[Quertus]

Oh god damn it. Don't muddle a clear-cut issue. It means arbitrary decisions made by a game master by authority of their position. Going on and on about different kinds of decisions is pointless, they are all fiat. Skip the semantic subcategorization and just talk about quality of decisions directly.

Ok, maybe I was confused about the term…

So, in the set “GM fiat”, we have many things, including

• Railroading
• Rulings
• House rules
• Please hit me with a clue-by-four
• Creating content

and, by a secondary definition, “a car that pulls the plot back on script”.

Is that a better definition? Anything I’ve missed?

If I’ve got it right this time… as with most things, I think buy-in is the key. For me, I find… hmmm… I was going to say, “I find GMs *making* things generally works much better than GMs *changing* things”. And, while true, I think I’ve missed my own point.

I think that the actual point is, anything - especially any fiat - that can be handed off *to the group* is more likely to be done in a way that the group enjoys when done by the group than when done by the GM.

Sure, sometimes, the GM has their fiat tow the starship back into Federation space. But, when you’ve got 5(+) people sitting around the table, the odds that the GM is the one with the cool, memorable idea? I wouldn’t bet dollars to donuts on it, and you shouldn’t, either.

That said,

Trees in the real world don't follow anything like a similar basic pattern. Variety and dissimilarity and fact-sensitivity are the rule; similarity is the exception. And the existence of similarity means there was already a decision from someone (ie fiat)--nothing exists in the fictional world that wasn't placed there by someone. It's decisions all the way down.

Personally, I find "the DC to pick all locks is 15 +- 2 unless I say otherwise" to be more jarring than "locks differ. Some locks will be trivial to pick; others nigh impossible. Yet others somewhere in between. The situation/fiction on the ground will determine which is which." And for those trivial/impossible ones, those are sliding scales depending on the situation. Given all the time in the world, a nigh-impossible one may be much easier. Given no time or equipment at all, a trivial one may pose a challenge worth rolling for.



IMO, that's the worst example to choose. Social stuff is intrinsically the most fact-bound part of games. In some cases, trying to persuade X to do Y may be literally trivial--they already were going to do it. In other cases, they have only a tiny barrier, so any attempt will work. No roll needed. In yet other cases, they're persuadable but the difficulty will depend on how you approach it[1]. In yet others, the best you can do is talk your way out of a catastrophe and into just a stinging rebuke (degrees of failure). NPCs are people, not stock elements with fixed dialogue trees.

And I find the idea that you always need to roll for everything to be horrible for gameplay. Rolls, like rules generally, are tools to be used when there is uncertainty to resolve. And when at least two of the possible outcomes are interesting (necessitating a choice between outcomes).

[1] persuading the corrupt guard to let you through will be trivial if you offer cash; impossible if you appeal to his honor. In contrast, persuading the paladin to let X slide may be possible if you can appeal to his desire for mercy, while a bribe will just get you in more trouble.

I think I agree with everything PhoenixPhyre said here, and that stating that agreement is one of the most efficient ways to clarify my position.

Consistency is not just “valuable”, it’s a requirement for intelligent thought. But, as the man said, “a foolish consistency is the hobgoblin of little minds”.

IME, a good game (ie, not Fatal) is likely to have less foolish consistency than one of my GMs would introduce foolish inconsistency. But, absolutely, “foolish consistency” like that described above should be avoided in any group with the capacity to produce something less foolish.

As that’s not something most GMs I’ve had are qualified to do, it’s really aided my perception that “the group” is superior to “the GM”.

[NichG]

In the context of this (and things about GM decides vs Group decides), I think it's only fair to ask: What is the role of 'Player Fiat'?

Maybe someone else in the group would have the cool idea about what one player's character could do in a situation, but generally there's a bias to prioritize a player's control of their own character versus having the group play the characters via committee. So there's a role for both group decision making and individual decision making. Understanding what is gained by not letting the group play everyone's character by committee seems like it'd be an important first step to understand why one might not just have the group democratically act as referee.

[icefractal]

Personally, I find "the DC to pick all locks is 15 +- 2 unless I say otherwise" to be more jarring than "locks differ. Some locks will be trivial to pick; others nigh impossible. Yet others somewhere in between. The situation/fiction on the ground will determine which is which." And for those trivial/impossible ones, those are sliding scales depending on the situation. Given all the time in the world, a nigh-impossible one may be much easier. Given no time or equipment at all, a trivial one may pose a challenge worth rolling for.

All locks being the same? No. But having an idea what kind of lock merits what kind of DC? Yes.

Like, if you have some baseline DCs for locks:
15: Low Quality
20: Simple
25: Average
30: High Quality
40: Amazing, a masterpiece of locksmithing

Then you have an idea what "Disable Device +10" means. It means you can open a simple lock in seconds if not distracted, and open most locks given a couple minutes to fiddle with them. But you're not good enough to break into an important vault, or to pick a good quality lock in unfavorable conditions. You're a good but not masterful lockpicker.

Yeah, there's situational modifiers - those are separate from the baseline. No tools, improvised tools, excellent tools, bad weather, distractions, all are modifiers, but still - a shoddy lock is easier to pick under adverse conditions than an excellent one is.

If the DC is "whatever the GM feels at the moment" - you can't really say anything accurate about how good you are at picking locks. Maybe you'll be able to pick that guild vault no problem, maybe a basic padlock on a shed will stymie you. Are you a good lockpicker with +10? Who can say?

Also - and this is not the only way to achieve this, but it's a useful side-benefit - having set DCs helps gently guide the GM away from the Oblivion-autoscaling treadmill. Just a little reality check - "Wait, I'm putting above-amazing locks on a normal warehouse? Does that seem legit? Maybe if [player] has put the resources in to become practically a demigod of lockpicking, I should just let him open this door without trouble."

[PhoenixPhyre]

All locks being the same? No. But having an idea what kind of lock merits what kind of DC? Yes.

Like, if you have some baseline DCs for locks:
15: Low Quality
20: Simple
25: Average
30: High Quality
40: Amazing, a masterpiece of locksmithing

Then you have an idea what "Disable Device +10" means. It means you can open a simple lock in seconds if not distracted, and open most locks given a couple minutes to fiddle with them. But you're not good enough to break into an important vault, or to pick a good quality lock in unfavorable conditions. You're a good but not masterful lockpicker.

Yeah, there's situational modifiers - those are separate from the baseline. No tools, improvised tools, excellent tools, bad weather, distractions, all are modifiers, but still - a shoddy lock is easier to pick under adverse conditions than an excellent one is.

If the DC is "whatever the GM feels at the moment" - you can't really say anything accurate about how good you are at picking locks. Maybe you'll be able to pick that guild vault no problem, maybe a basic padlock on a shed will stymie you. Are you a good lockpicker with +10? Who can say?

Also - and this is not the only way to achieve this, but it's a useful side-benefit - having set DCs helps gently guide the GM away from the Oblivion-autoscaling treadmill. Just a little reality check - "Wait, I'm putting above-amazing locks on a normal warehouse? Does that seem legit? Maybe if [player] has put the resources in to become practically a demigod of lockpicking, I should just let him open this door without trouble."

Alternatively (and much more simply and usefully), have a basic set of guidelines about what each DC (in buckets) means for any check. And set a fixed range of DCs. Oh wait...that's 5e. Specifically, you can play just fine with all DCs in the 10-20 range. You don't need specifics for each type of lock (which leave out way too much to be useful, while also being confining if taken as player-facing).

But the interesting part isn't what the DC is. That's fairly boring 99% of the time. No, the interesting thing (and the thing that triggers accusations of fiat) is that most of the time you shouldn't have to roll a check at all. But which times is intensely fact-sensitive, and so the book really can't help you much except in the vaguest of terms. So it has to be done "in software" by the DM.

[Telok]

Alternatively (and much more simply and usefully), have a basic set of guidelines about what each DC (in buckets) means for any check. And set a fixed range of DCs. Oh wait...that's 5e. Specifically, you can play just fine with all DCs in the 10-20 range. You don't need specifics for each type of lock (which leave out way too much to be useful, while also being confining if taken as player-facing).

And then we're back around to my having played 5e with a DM for whom climbing a rope was a DC 15+ check every time and the "jumping further with a check" didn't happen because his personal experience* told him those were really hard things to do and DC 15 was the average DC. My personal experience tells me that I can't convince people with blatantly false facts or fast talking, and I know that I'm an outlier because con artists exist. But when I started DMing many many years ago before the d20 system existed, you couldn't fast talk an NPC in my games no matter what resources or effort you'd sunk into it because my experiences told me you couldn't do that.

I made the mistakes of thinking that my personal biases, experiences, & assumptions should dictate what was 'easy' and 'hard' in a game. I learned from those mistakes, eventually, the hard ways. I've seen inexperienced DMs make the same mistakes (and eventually rage quit) in 5e with its lack of guidance, and I've seen those same DMs do better (not total amaze-balls, but definitely better) in Starfinder when they had a guideline for things like how difficult climbing a rope or jumping a ravine should be.

Rules are guidance & structure. My experience has been that novice DMs, who are often a bit twitchy about unsolicited DMing advice and don't sift forums & blogs for advice, benefit from more guidelines & structure. They have an easier time focusing on interacting with the plauers & trying to make the game fun if they don't have to keep making up rules stuff all the time.

As much as I love AD&D, its a rough game to start DMing because there is even less guidance & 'dc' type things than D&D 5e has on how to rule stuff. As much as I dislike D&D 4e, I think its a great game for new DMs to learn on because it does all the heavy rules lifting and lets those new DMs learn how to manage the game. As much as I like that an experienced DM can run a good game even in a sucky system, my locale seems to have a really super high rate of D&F DM burnout these last 5 years or so & nearly all games I don't DM (and I don't bother DMing 3e or later D&D any more) are run by a new/novice DMs.

Tie it back to DM fiat. I think its good when a DM can, but does not have to, resort to fiat in order to make the game system work. Since 'rule zero' is a thing that 'can' is always available.

* Overweight, a bad back, no head for numbers, thought 3 chances at 30% success were equal to a 90% success rate.

[Vahnavoi]

In the context of this (and things about GM decides vs Group decides), I think it's only fair to ask: What is the role of 'Player Fiat'?

Maybe someone else in the group would have the cool idea about what one player's character could do in a situation, but generally there's a bias to prioritize a player's control of their own character versus having the group play the characters via committee. So there's a role for both group decision making and individual decision making. Understanding what is gained by not letting the group play everyone's character by committee seems like it'd be an important first step to understand why one might not just have the group democratically act as referee.

I'll start from the end: why not just democratically vote on everything?

First answer: collecting and counting votes is slower than a person just deciding.

Second answer: involving every player in a decision regresses quality of said decision towards the average of the players' abilities. Corollary: players don't benefit equally from such process. Weaker players get more out of hearing stronger players than vice versa.

Proof: go watch "Twitch plays Pokemon" etc. It's fairly simple to observe that most people engaging in voting on the game moves could beat the game better and faster if playing the game normally as an individual. Similar dynamic is observable in "Chess master takes on the world" type challenges. Bottlenecks of decision making mean that only some types of problems benefit from having lot of people thrown at them; in some others, the group is only as capable as the most capable individual in that group. For yet others, you get "chain is only as strong as its weakest link" - the quality of decision is capped by least capable member.

[Quertus]

In the context of this (and things about GM decides vs Group decides), I think it's only fair to ask: What is the role of 'Player Fiat'?

Maybe someone else in the group would have the cool idea about what one player's character could do in a situation, but generally there's a bias to prioritize a player's control of their own character versus having the group play the characters via committee. So there's a role for both group decision making and individual decision making. Understanding what is gained by not letting the group play everyone's character by committee seems like it'd be an important first step to understand why one might not just have the group democratically act as referee.

Huh. Well, there are systems where you trade characters, or run them by committee. But, IMO, the reason most systems don't run that way is "role-playing", with a side of "hidden information" and "limited headspace".

Role-playing is making decisions for the character, in character.

"What should happen when I charge?" "Well, the rules are…" Anyone can say this - and, in many of the best groups I've been in, that's exactly how it works. Resolving decisions through mechanics is irrelevant to roleplay, and it's where any sufficiently savvy player's head should be at the point that you're discussing action resolution.

"What happens when I try to give the car a literal 'rocket boost'?" Again, anyone can respond to this. Odds are that the GM isn't the one with the best answer.

"What's over that hill?" The GM might know the answer, or they might not. They might not want to players to *know* whether they know the answer or not, because that might provide metagame clues about hidden information. My opinion on this matter should undoubtedly get copied into the "unpopular opinions" thread, but…

There are only 3 times that the GM should ask the players to create content that way. The first and most obvious is when they don't know the answer, and want (or don't mind) the players knowing that fact (it doesn't, say, ruin the mystery for the players to know that the GM doesn't know what's in the barn). The second reason is that it allows the GM to create an illusion that the *players'* belief will discourage them from seeing through. The third is in preparation for the second, convincing the players that the GM is just too lazy to create the content, or believes in the players' ability to do so, so that they don't notice anything unusual about it. Spanish Inquisition special, if the GM actually *is* lazy, or actually *does* believe that the players can make content.

So, what is the role of player fiat? To, by the power of numbers, by your powers combined, to statistically give better answers than a single person would give, in scenarios where the one person has no special qualifications (wrote the character, wrote the content) to give them obvious primacy. Examples of good scenarios for when crowd-sourcing should be the default include rules, rulings. Examples of when crowd-sourcing should not be the default are "what's over that hill?", role-playing. Examples of things where not only is crowd-sourcing superior, but can also be used to measure or get buy-in include dragging the game back on script.

no head for numbers, thought 3 chances at 30% success were equal to a 90% success rate.

Yeah, this is definitely one of the reasons why "give me rules over idiot GMs" is important. The rest of your story is another.

Think about it. If you're got 5 people at the table, the odds that the GM has the best answer on any given topic is only 20%. Yet many GMs will happily completely ruin their game rather than entertain the notion of even listening to a player.

You've got *probably* more than those 5 people writing the rules… which only gives the rules about a coin flip on being better than what your table comes up with.

However, with errata coming from responses from *every* table, thinking that *you* have a better answer than the world? That only makes sense if you have some special advantage for understanding the scenario, you're a genius with hubris as big as mine, or the system it written or edited by imbeciles. Sadly, at least one piece of one or more of those is terribly common.

[Quertus]

I'll start from the end: why not just democratically vote on everything?

First answer: collecting and counting votes is slower than a person just deciding.

Second answer: involving every player in a decision regresses quality of said decision towards the average of the players' abilities. Corollary: players don't benefit equally from such process. Weaker players get more out of hearing stronger players than vice versa.

Proof: go watch "Twitch plays Pokemon" etc. It's fairly simple to observe that most people engaging in voting on the game moves could beat the game better and faster if playing the game normally as an individual. Similar dynamic is observable in "Chess master takes on the world" type challenges. Bottlenecks of decision making mean that only some types of problems benefit from having lot of people thrown at them; in some others, the group is only as capable as the most capable individual in that group. For yet others, you get "chain is only as strong as its weakest link" - the quality of decision is capped by least capable member.

Voting, as you described, is of the ilk of "a foolish consistency".

Imagine, instead, "our ship is now 10 years away from Federation space".

Imagine several players coming up with ideas like "time skip" or "dig through Federation records for wormholes or tech" that are all "meh", none of which gets more traction then simply "retcon".

Until someone has the bright idea for as little green car (the GM's fiat) to tow the ship back.

Everyone loves it, "votes" enthusiastically that this is the best answer.

The GM shrugs, and, not having a better idea, accepts the will of the people on matters of Rule of Cool.

Point is, "what people find cool" or "what people find fun" or "what people hate" or "what people want to 'fade to black' for" isn't something that the GM can - or even should - answer. It takes true cluelessness and hubris bigger than mine for a GM to think that they can answer such questions for their players better than the players themselves can.

-----

Also, if you're running the homebrew game Paradox (think Rifts, but good), and you happen to not know the charging rules in that system? Don't just sit there trying to remember, and don't just make something up. Ask! Maybe someone at the table does know the answer.

Same goes for anything, like "what's today's date" or "what's the material components for Fireball"?

Note how most things of this type don't default to democratic voting for their resolution.

-----

I am reminded of a group I was in, that had one of the most elegant rules question resolution methods I've seen.

Whenever a question was brought up, the group had "5 minutes" (I never saw anyone check a timer) to produce a definitive answer.

A definitive answer could take the form, "it's X - here are the relevant rules" or "if Y, then Z - and nobody wants Z".

If no definitive answer was found, the GM would flip a coin (actually a duplex cookie, which I had brought, and thereafter continued bringing). White side up, it worked however was in the party's favor at the moment, *and was added to the house rules*. Black side up, it worked however was *not* in the party's favor at the moment, *and was added to the house rules*.

The GM understood that, things which can solved by reason should be solved by reason; things which cannot be solved by reason should be solved arbitrarily.

Note the lack of "democratic voting".

When you're dealing with reasonable people, you can vote on "what's cool" or "what's fun", not on "what's reasonable". When you're not dealing with reasonable people, it's time for a clue-by-four, not a vote.

[MoiMagnus]

However, with errata coming from responses from *every* table, thinking that *you* have a better answer than the world? That only makes sense if you have some special advantage for understanding the scenario, you're a genius with hubris as big as mine, or the system it written or edited by imbeciles. Sadly, at least one piece of one or more of those is terribly common.

I'd point it that there is a difference between
(1) thinking about having the best answer in an absolute manner,
(2) thinking about having the best answer relative to the current circumstances (using factors that the rules didn't accounted for)
(3) acknowledging that the rules might be the best for the gameplay/tone/etc the designers and playtesters wanted, but wanting yourself to achieve a gameplay/tone/etc fundamentally different [but not different enough to be worth searching through all the other RPGs systems... or different enough that using another system would be better but you didn't found one more adequate in your quick search]

[Quertus]

I'd point it that there is a difference between
(1) thinking about having the best answer in an absolute manner,
(2) thinking about having the best answer relative to the current circumstances (using factors that the rules didn't accounted for)
(3) acknowledging that the rules might be the best for the gameplay/tone/etc the designers and playtesters wanted, but wanting yourself to achieve a gameplay/tone/etc fundamentally different [but not different enough to be worth searching through all the other RPGs systems... or different enough that using another system would be better but you didn't found one more adequate in your quick search]

It's certainly fair to say that there being an "outside the box", that the rules not being complete, is a fairly fundamental attribute of at least most marginally Simulationist RPGs. So, yes, I expect *someone* will have to fill in the blanks, but that's different than "changing what's written".

Changing what's written can, indeed, mess up the tone/gameplay/etc. If the group wants different tone/gameplay/etc (without having to find a different system), the group is probably best suited to make those changes.

[HumanFighter]

I lean more towards trusting in the ruleset and system, using GM Fiat sparingly and only whereas needed. As a GM I don't fudge dice, or make anyone do anything, unless the rules say so (Confusion effect, for example. Roll to see what you do this turn, etc.) Rather, I give the players incentive to do things, hooks, as they're sometimes called. But I'm not making them pursue any particular plot. And still, the rules remain consistent and you usually know what to expect.

Why do I prefer this way? Well, I see that if a game relies solely on GM Fiat or leans too much in that direction, that game will feel exclusive and "clique-ish" only known to or accessible by that particular play group, leaving everyone outside of it clueless as to what is going on. I prefer ruleset because as long as everyone knows the rules well enough, everyone can understand basically what it is all about and it is more all-inclusive.
Also, at a GM Fiat-focused table it is too easy (and, in my experience always inevitably happens) for GMs and players to play favorites, and things can get quickly out of hand with the blatant unfairness here, leaving some players feeling left-out at best (and, at worst, humiliated). But I will acknowledge the flexibility and high potential for adaptability for these types of games, leaving it to be much more wild and "memorable." Memorable does not always mean good.

[Beleriphon]

I'm a big fan of the Fiat instructions in Mutants and Masterminds. Particularly 2E, where it was very clearly laid out the GM can do whatever they want, but they should be doing things to make the game fun. If they do an asspull like the bad guy gets away on a secret jet pack they need to give the players something in return. In the case of Mutants and Masterminds that's a hero point to be redeemed later.

[Jedaii]

GM Fiat drives games/story.

Players can get turned around and get confused by a game/story. The D/GM gets things back on track with a

level of "FIAT". While some may hate it GM Fiat drives games OUTSIDE of the rules as the rules can never

cover everything the players want to do. There is NO RPG that has a RULE FOR EVERYTHING. None.

That's where GM Fiat comes in to keep things moving when players find themselves in a place outside the

normal rules. This is why "Rules Lawyers" have a very specific place as they can identify and explain rules

as they apply to the PCs but again there are no rules for everything. In the space between a rule and an action

RPG players need a referee to determine RAI vs. RAW.

[kyoryu]

Then there are systems like Fate, where the mechanical bits are fairly well cut and dried and meant to be run pretty straight, while the results of actions is left almost entirely in the GM's hands outside of vague descriptors. So there is both GM Fiat and Not GM Fiat.

It's really an interesting question, then, of where GM judgement should and should not come into play, as well as the degree.

[Jay R]

Before role-playing games existed, I grew up with a version of GM fiat. My father would say, “No matter what the instructions say, no matter what I told you, don’t do nothin’ stupid.” That’s the essence of GM fiat done correctly.

Applying the rules is like eating food. That should always happen. Changing the rules during the game is like taking medicine; it's only a good idea if something is wrong.

[Easy e]

So, I had been toying around with a game that used a "meta-currency". In this system, what the GM can and can not do is very clear by using the meta-currency. There is very little GM Fiat. The GM does not roll any dice, only modifies the player's rolls.

After playing the game, the players told me that when the GM used their Meta-Currency to influence things; they felt like it was directed at them. Like the GM wanted to make their life harder. It was an antagonistic structure.

They actually said that they preferred GM Fiat adding mods, setting difficulty, and taking actions behind the scenes based on rules interpretations. They felt like this was "less targeted" than the GM using currency to add personal challenge. This surprised me as those personal challenges were strictly limited by the rules.

[Mordante]

True. But similar situations should be resolved similarly, I think. If a PC want to persuade an NPC to do something, "roll for a skill", "describe what you say" and "okay, you persuade them" are all acceptable responses from the GM, but I would prefer it if it was the same response every time (even if the specifics are a little different) and not seemingly arbitrarily varying depending on the GMs mood or whatever.

I disagree,

"Roll for a skill" When there is a chance it will fail but otherwise it is a reasonable course of action.
"describe what you say" When you try and persuade someone in a situation where it's really not applicable. Where the chances are very slim, or just an odd situation.
"okay, you persuade them" When there really is no need to persuade at all, or where the power/skill difference is big in the PC advantage.

[Batcathat]

I disagree,

"Roll for a skill" When there is a chance it will fail but otherwise it is a reasonable course of action.
"describe what you say" When you try and persuade someone in a situation where it's really not applicable. Where the chances are very slim, or just an odd situation.
"okay, you persuade them" When there really is no need to persuade at all, or where the power/skill difference is big in the PC advantage.

I see what you're saying, but I still prefer it to be handled more consistently (and note that I said that similar situations should be handled similarly, change enough parameters and it might make sense to change how it's handled).

Out of curiosity, would you feel the same way if the PC was trying to stab someone rather than persuade them?

[PhoenixPhyre]

If it's a sure thing? Absolutely.

[KorvinStarmast]

There is NO RPG that has a RULE FOR EVERYTHING. None.

Yep.

That's where GM Fiat comes in to keep things moving when players find themselves in a place outside the normal rules. This is why "Rules Lawyers" have a very specific place as they can identify and explain rules as they apply to the PCs but again there are no rules for everything.

Yes.

RPG players need a referee to determine RAI vs. RAW.

Yes.

GM-less games have their own resolution methods for that.

[Easy e]

For discussion, can you guys help me wrap my head around how to define GM Fiat vs Player Fiat?

GM fiat seems to be when the GM sets the boundaries of the world via when to roll, mods, DC/TNs, difficulty, etc.

Player Fiat seems to where the Players set the boundaries of the world with narration, defining action, narrating results, up to even calling for rolls, mods, or DCs/TNs?

Again, I am having a hard time articulating what I am driving at, but hopefully this gives you the general idea.

[Vahnavoi]

@Easy E: It is very easy: game master fiat is arbitrary decisions made by a game master by authority of their position. Player fiat is arbitrary decisions made by a player by authority of their position.

Which decisions are covered by fiat is purely game dependent - there are no universal rules about that across games. Same goes for which fiat decisions are domain of a game master instead of a player. In a game with the rule "game master has final say over game matters", game master fiat takes precedence over player fiat.