The Alexandrian

The Art of Rulings

March 29th, 2011

Banksy - The Grin Reaper

In response to the Update from the Crypt of Luan Phien, Poe asked me:

Out of curiosity, do you rely solely on the players expertise when developing maps or do you use some form of skill checks to indicate a character expertise in determining what is going on with an unusual map situation like this one?

In the case of that particular map, it was primarily player expertise that crunched out the workings of the crypt. But Poe’s question got me thinking about the wider question of how GMs make rulings while running a roleplaying game.

First, I prefer to use systems which offer broad mechanical support for GM rulings. Some people prefer pure GM fiat, but I like having a mechanical base to make rulings from because:

  1. It allows me to make a ruling of uncertainty. (Instead of saying “that definitely happens” or “that definitely doesn’t happen”, I can say “that sounds likely, let’s see if it happens”.)
  2. It allows for varying character capacity to have a meaningful impact on events.
  3. It can provide guidance when I’m not certain how to rule.
  4. The mechanical outcome is an improv opportunity, often spurring me to create things which I would not have created otherwise.
  5. It provides a consistency to similar rulings over time.

And so forth. In general, badly designed rules act as unreasonable straitjackets. Good rules, on the other hand, enable new forms of play and expand the scope of the game.

With all that being said, my general approach to making rulings as a GM basically looks like this:

  1. Passive observation of the world is automatically triggered.
  2. Player expertise activates character expertise.
  3. Player expertise can trump character expertise.


Passive observation may include stuff that’s obvious to everybody (like walking into a room with a giant ball of flame hovering in the middle of it), but it might also include reactive mechanics for determining whether or not characters notice something that isn’t automatically apparent. In OD&D this would include surprise tests. In 3rd Edition, this would include Listen, Spot, and Knowledge checks (although these skills can also be used in non-reactive ways.)

(Why am I including Knowledge skills here? Imagine that the characters walk into a room with a large heraldic shield painted on the wall. Do the characters recognize that as the archaic heraldry of King Negut III of Yrkathia? If they do, the players shouldn’t have to ask if they recognize it — they just recognize it.)

But now we get to the real heart of the matter. This is where a player says, “I want to do X.” And you need to making a ruling about how to resolve the outcome of X.


What I mean by this is that the characters don’t play themselves. With the exception of purely passive observation of the game world, players have to call for an action which requires a skill check in order for the skill to be activated.

If we consider a simple example, like:

Player: I check the chest for traps.

GM: Make a Search check.

This may seem self-evident to most of us. On the other hand, I have seen games where GMs will respond to “I open the chest” by calling for the Search check. You may have also heard players say things like, “My character is a 12th level rogue. She would have known better than to open a chest without checking it for traps first!”

I can see the potential legitimacy of the philosophical question being raised. Regardless of which approach we take, there is a point at which the player’s control of the character seems to stop. Consider that simple Search check again: The player decides to check the chest for traps, but then we’re allowing the mechanics to determine how, based on the character’s expertise, that search happens. But could we not, with equal validity, say that when the player decides to open the chest, we should allow the mechanics to determine how, based on the character’s expertise, that happens?

In general, however, I would point out that an integral part of roleplaying is, in fact, playing your role — i.e., making choices as if you were your character. When you turn meaningful choices over to the game mechanics instead of making them yourself, I would argue that you are no longer roleplaying.

Of course, on the other end of the spectrum you have a variety of pixel-bitching. Here you’re never allowed to turn the resolution of an action over to your character and searching the chest becomes a litany of detail:

Player: I check the chest for traps.
GM: How do you do that?
Player: I check under the chest for a pressure plate.
GM: How do you do that?
Player: I run my fingers around the perimeter, looking for any edges. Then I’ll pour some water around to see if it sinks into any sort of depression. Then I’ll very carefully lift one corner of the chest just high enough that I can slide a piece of parchment under there and see if it strikes any sort of spring-loaded trigger that’s rising with the chest.

(Or, if you’re playing with GM bastardy: “I check under the chest for a pressure plate.” “You lift the chest to look, triggering the pressure plate!”)

There is certainly some art in figuring out where the “sweet spot” is for activating the character’s expertise. But 99 times out of 100 it’s going to be fairly self-evident. When in doubt, look for the meaningful choice. Or, rather, never assume that a character is doing anything which requires a meaningful choice unless the player makes that choice.


On the other hand, you also don’t want to negate meaningful choices by insisting that certain actions must be handed off to the character’s expertise. That’s why I say that player expertise can trump character expertise.

Sticking with our chest-searching motif, consider a scenario in which there is a hidden compartment in a chest which can be accessed by lifting out the bottom of the chest. We’ve determined that the hidden compartment requires a DC 17 Search check to discover. The player says:

  • “I search the chest.”
  • “I check the bottom of the chest for hidden compartments.”
  • “I take my axe and smash open the bottom of the chest.”
  • “I check the chest for traps before opening it.”
  • “I check the lid of the chest for hidden compartments”.

The first example is vanilla. You’ve handed the resolution over to the mechanic and you get a flat Search check against the DC of the hidden compartment. A success could generate a number of different responses (ranging from “yup, there’s a hidden compartment” to “you notice that the exterior of the chest is several inches deeper than the interior of the chest”).

The second example is more specific (and happens to coincide with what’s actually there to be found). I would tend to grant something like a +2 circumstance bonus to the Search. If there were other things to be found in the chest, I might also allow the Search check to find them (but since you’re specifically looking for something else, such a check would have a penalty applied to it).

In the third example you’ve taken an action which would automatically find the compartment. No Search check is required. (Player skill has completely trumped the mechanic.)

The fourth and fifth examples demonstrate that trumping character skill isn’t always a good thing. In the fourth example, you have no chance of finding things hidden inside the chest if you’re limiting your search to the exterior. Similarly, in the fifth example you’re specifically looking in the wrong place.

As another example, consider a hallway with a pit trap in it. The pit trap has a 50% chance of activating whenever someone steps on it and it requires a DC 17 Search check to find it. The player says:

  • “I search the hall for traps.”
  • “I proceed carefully down the hall, tapping ahead of me with my 10-foot pole.”
  • “I summon a celestial badger and have it walk down the hall in front of me.”
  • “I pour a waterskin onto the floor to see if it runs down any seams or gaps.”

The first example is, once again, a straight forward Search check. The second and third examples bypass the Search mechanics, and instead grant a 50% chance that the pole or summoned creature will trigger the trap.

The fourth example, on the other hand, could be handled in several ways. One could easily rule that such a technique would automatically find the trap (particularly if the player specifies exactly which section of corridor they’re checking). I’d probably grant a hefty bonus (say, +10) to the Search check for using an appropriate technique.

In many ways, this comes back to meaningful choice: We assumed before that characters don’t take actions which require meaningful choice unless the player makes that choice. Here we assume that any choice the player makes is probably meaningful and take the specificity of those choices into account when we make our ruling.


Let’s consider the specific example of mapping the Crypt of Luan Phien, a segmented dungeon in which each section periodically rotates independently in order to change the layout of the dungeon.

At the high end, we can imagine a player saying, “I make a Knowledge (dungeoneering) check to make a map of the dungeon.” To which my answer would be, “No.” (They’ve failed to achieve the necessary specificity to activate their character’s expertise.)

In actual play, the mapping of the dungeon was solved almost entirely through player expertise. They simply observed which rooms connected to each other and slowly built up an understanding of the possible configurations of the dungeon. (The sole exception would be late in the process, when they started checking corridors to find the wall seams which would confirm their understanding of where the breaks between segments lay.)

But I can think of several ways that they could have activated their character’s skills:

  • Check the curvature of the stone walls that periodically blocked various hallways in order to determine (at least roughly) the circular diameter of each section.
  • Try to determine the direction in which each section was rotating.
  • Try to figure out how much stone would slide past an open corridor between sections in order to determine how far each section was rotating.
  • Use a compass or spell to determine orientation before and after a shift.

And so forth.

Go to Part 2

Part 2: Intention and Method
Part 3: The Fiction-Mechanics Cycle
Part 4: Default to Yes
Part 5: Skill and Difficulty
Part 6: Fictional Cleromancy
Part 7: Vectors
Part 8: Let It Ride
Part 9: Narrating Outcomes
Part 10: Fortune Positioning
Part 11: Narrative vs. Action Resolution
Part 12: Hidden vs. Open Difficulty Numbers
Part 13: Hidden vs. Open Stakes
Part 14: Group Actions

Addendum: Let It Ride on the Death Star
Rulings in Practice 1: Gathering Information

The Art of Pacing
The Art of the Key
Gamemastery 101

Share on TumblrTweet about this on TwitterShare on StumbleUponShare on FacebookShare on RedditShare on Google+Digg this

11 Responses to “The Art of Rulings”

  1. mister k says:

    As a player I hate accidentally making my character better than they should be. For example, during a recent Exalted game I deduced something that, on reflection, my character probably wouldn’t have, but was unable to retract it having said it in character. Obviously intelligence is a difficult stat to roleplay if its not the extremes: “grah, ogre smash” or “elementary my dear water elemental”, as deciding what your character should or shouldn’t know is difficult. Its also something that the GM should usually avoid policing, unless the brainless imbecile ogre keeps explaining the villains plan to the other players.

  2. Brad says:

    In regards to the “rulings” topic: Our group has always been a fan of the GM fiat. In situations where we’d be forced to look up rules that are less than simple, we get a group consensus and handwave it with that. We make a physical (non-mental) note to look up the rule later for future reference. Typically, this ruling is looked up by the player involved as soon as they have the opportunity (during the next player’s turn, if during combat, for example). Because we have a large group of players (we’ve had 11 people before), expediting unclear rules like this is essential or we’d never get anything done.

    In regards to characters, taking actions, and being descriptive: We usually stick very solidly to the rules but allow the player to be as descriptive as he wants. What I mean by this is, if he says he searches the chest for hidden compartments, but doesn’t specify that he also looks for traps, according to the rules, all he is doing is making a search check. According to the rules, success on the search check indicates that he finds both. Thus, we go with the rules approach, rather than the poor wording used by the player and the GM says something like “while looking for compartments, you find one and happen to notice a trip wire in the process”.

    We’re also pretty strict when it comes to being explicit with your actions and inactions. We handle this in a way that is similar to when you can add action dice to a roll which is “only before the GM reveals the result of that roll”. If the player doesn’t explicitly specify that his character searches the chest for traps, he can change his actions all he wants until the trap goes off in his face. If he didn’t say it before then, there’s no going back and there are no exceptions to this rule. Being so strict about this rule has lead to players doing a better job of thinking things through and making better decisions overall. Occasionally, if the situation is dire enough, we’ll occasionally give hints that the player may want to consider his options one final time before committing to a specific course of action. When we do this, we give a simple “Is that your final answer?”


  3. Andrew says:

    I handle “Intelligence” as “the ability to see patterns.” The argument there is that very intelligent people lack common sense, and some very common sense wise people are not able to do math quickly or well, put together puzzles quickly, or deduce riddles and such. Moving Intelligence into a task-based approach frees up the role playing and reduces passing judgment on player capacity.

  4. -C says:


    This is how we play, ideally. With the check (FT) for backup. Is there something wrong with this detailed approach?

    The corollary is that it’s not just random crap. It’s vague crap that gets more specific as you take a closer look. Chests themselves are markers of the specific crap – being able to detail all of these things is the point of my interesting treasure/tricks, empty rooms, and trap design documents.

    This is something I’ve given a lot of thought to, and if the game isn’t in the tactical combat, I’ve sort of come to the conclusion that it is in the very pixel-bitching you deride.

    Thanks for your analysis, btw. Like I said it’s something I thought a lot about and your perspective is eloquent and something I would recommend for most games.

  5. Justin Alexander says:

    I think there is a point at which everyone will say, “Enough is enough.”

    “I put the key into the lock.”
    “With how much force?”
    “As gently as possible. Then I turn it.”
    “Which way?”
    “It doesn’t turn left.”
    “Okay, then I try to the right.”
    “How far?”
    “Let’s start with 10 degrees.”
    “Nothing happens.”
    “Let’s try another 10 degrees.”
    “You feel some slight resistance.”
    “I’ll just a little more force, but just enough to keep the key turning.”
    “How far?”
    “Another 10 degrees.”
    “Okay, you begin to feel the tumblers turning.”
    “Okay, I turn it the rest of the way to a full 90 degrees.”
    “You hear the lock click open.”
    “I leave the key in the lock and begin examining…”

    Is the question of, “How far do you turn the key?” meaningful or interesting to you? If so, then you shouldn’t skip past it. If not, then being required to dwell on that question every single time you want to use a key is probably not going to result in entertaining play.

    (I suspect we’re actually closer in our agreement on this issue than you think.)

    My approach has largely evolved out of trial and error, but as I put it to analysis I suspect one of the reasons it works is because it tends to naturally find the right “sweet spot” not only for everybody involved, but for the particular situation under consideration. As a GM I set a relatively high threshold of “that’s enough specificity for me to make a ruling”, but if the players want more specificity than that — if there are more detailed choices that they feel are meaningful to the situation or to their character — then they are free to make those choices. (At which point I will assume they’re meaningful and take those details into account.)

    So for a typical chest they might just say, “I search it for traps.” And I say, “Okay, that’s enough specificity. Let’s roll it.” But later on there might be another chest that they’re feeling particularly paranoid about, and so they say, “I’m going to start by very carefully checking the floor around and under the chest for pressure plates.” or “I’m doing a thorough visual inspection of the chest before I ever touch it.”

    The one place where the system falls down is when I, as the GM, want more specificity than the players want to give. This is actually a problem I’ve lightly run into while running OD&D: Because there are frequently no mechanics for me to turn particular actions over to, I’m forced to ask for more details in order to make any kind of informed ruling.

    There are many people who would say that’s a feature. (Details are good! You’re making them really think about the game world!) But when it forces players to focus on choices that they don’t find interesting or meaningful (like whether to check the left side or the right side of the chest first), the experience is just frustrating.

  6. Andrew says:

    Though it is true that when there is any doubt about whether it is safe or not, it is always amusing to ask “Which hand do you touch it with?”

  7. Charlie says:

    I think the most important issue is the one you mention above in the comment section: the meaningful decision and the focus of the game at hand.
    For one group of players searching the chest or traps in a number of ways is the fun and the focus of the game, and in those cases, bypassing the entire thing with a Search check or a simple dice roll makes the game “less fun” for them, because it is a key element of that gaming style.

    For others, the chest and the traps might as well be just an accesory, and therefore they decide to fasten the process with a dice roll. Or apply a mix bag like this great article shows. But the important decision is whether or not the task at hand is an important part of the game. Is convincing the dragon to let you pass an important moment of the session? Then a straight Diplomacy roll isn’t appropiate, the players are supposed to roleplay that and then apply a bonus to the roll.

    As for myself, I usually apply the rules as the article, the player must do something in order to activate the skill, if he/she has a very good idea, then I go from a bonus to an automatic success. HARP has a very handful set of rules to solve all those things, for example there’s a skill for Knowing about traps, so a succesful roll adds a bonus to the Disarming Trap skill.

  8. Dan Dare says:

    “5. It provides a consistency to similar rulings over time.”

    I ran into an interesting problem with this when running Traveller.
    – Me: You need 9+ to succeed at that.
    – Player: But last time it was only 7+!

    The problem is that in the instance the task was harder for reasons the character was unaware of. I realised that telling the players what the success threshold is had some values and drawbacks.

    It gives them an idea of odds, so they can decide to attempt it or not.
    It allows them to make the roll and assess the outcome, leaving the GM free to multitask a bit.
    But its also dissociative, the character doesn’t actually know what the real odds are.

  9. Byron says:

    In the example from play, a theoretical player wants to create a map using their dungeoneering skill and as the GM, you would say no due to their lack of specificity.

    Other than the examples of player knowledge being acceptable, what would you need to allow the player to make a map? Would mentioning using the compass and other associated tools be enough for a player who doesn’t know much of dungeoneering?

  10. GM DON’T #3 : Résolution Différée – quefaitesvous says:

    […] d’actions suivantes puisse être proprement déclaré. (Les seules exceptions sont quand le MJ sent qu’ils n’ont pas assez d’information pour résoudre l’actio… ou si l’action déclarée semble se baser sur une mauvaise compréhension de la situation. […]

  11. L’Art du Rythme – partie 1 – quefaitesvous says:

    […] l’Art de trancher j’ai écrit « En cas de doute, cherchez un choix significatif ». Et le […]

Leave a Reply



Recent Posts

Recent Comments