I don’t have a Long, Rambling Introduction™ today. Sorry. I considered several rants that might fit here. I thought about pissing and moaning about modern dice designs and dice accessories like dice towers and dice trays and how the people who use that s$&% need to die in a fire. I also thought about whining, again, about how modern RPGs are great at telling you the rules and terrible at telling you how to actually play the games. But, none of it really fit and I don’t like to force this s$&%.
So…
Dice. How to Dice.
It’s Not the Size of Your Bag, But How You Shake Your Dice
Today’s all about dice. How to use them. How to actually f$&%ing use them.
“But Angry,” I hear you whining, “isn’t that something I can learn from the rulebooks? And isn’t that something I already learned from you better? Like ten years ago?”
No. Yes. Sort of. Shut up.
I realize you think you know to use dice because you know to roll a d20 and add an ability modifier and compare the result to a Difficulty Class. But that’s just rules. Woefully incomplete, crappily written rules. I mean, who rolls the checks? You? The players? How do you decide? Should the roll be hidden? When should you use opposed rolls? What do you do when everyone’s all cooperating to do one thing? Do you even know when people are really cooperating?
Don’t answer. I know you don’t know any of this s$%& because I read your e-mails and I read your comments and I read your s$&% in my supporter-only Discord server. And the stupid s$&% you all keep saying just proves you have no f$&%ing clue how to use dice.
I mean, some of you are still spouting inane bulls$&% like “resolving things with Passive Scores isn’t challenging at all because I, the GM, already know the outcome and so I, therefore, increase the DCs whenever I use them.”
Do you want to give me a f$&%ing aneurism? Because that’s what you’re going to do. And that’s why we’re doing this s$&% today. To prevent my untimely demise by brain bleed. And, to a much lesser extent, to make you a less worse GM.
But before I can teach you how to really use your dice right, I’ve got to make sure you really understand dice rolling. As a concept. What it is and why it sucks.
Players Can Be Stupid; You Can’t
The problem’s not really that you don’t know how to use dice right. The problem’s that lots of you have some very wrong, very dumb assumptions about dice cluttering up that cauliflower you’ve got between your ears.
That ends now.
You’re a GM. You’ve got a job to do. That job’s to run an engaging, satisfying roleplaying gameplay experience. An actual good game. Not a fun game. Not a collaborative storytelling bulls$&% session. An engaging, satisfying roleplaying gameplay experience.
You’re a Game Master, not a Clown.
I ain’t talking to clowns anymore.
Players are children. They want all sorts of stupid crap. And they think stupid thoughts. They do stupid things. If you feed their wants, however much they enjoy it today, they’re going to end up bored and miserable and insufferable tomorrow. Your job’s to give them a good game that’s good for them. They’ll love you for it. Tomorrow. Even if they hate you today.
People — Clown GMs — will say there’s no wrong way to run an RPG. But they’re Clowns. While there’s lots of good ways to run an RPG — and your job’s to figure out the right good way for you and your players — while there’s lots of good ways to run an RPG, there’s also wrong ways to run an RPG. And one of the worst wrong ways is to try to be your players’ Fun Uncle instead of your player’s Daddy. Or their Fun Auntie instead of their Mommy. To spoil them. To be their friend. And to blow off your GMing responsibilities the minute they get hard.
Of course, if you’re still here after last week’s Clown screed — and I doubt there’s many of you left — I assume you’re stepping up and taking it seriously. Being a Mommy or Daddy. Trying to make the right GMing decisions. But decisions are based on assumptions and reason. No matter how good your intentions are, if you apply crappy reason to faulty assumptions, you’ll get terrible decisions.
So flush everything you know about dice right out of your brain. I mean, not literally. Because I’m not going to actually explain that dice are plastic polyhedrals used to generate random numbers or any s$&% like that. I’m just going to tell the real truth about what dice are for in RPGs.
The Dice Are Your Tool
Dice are your tool. You — the GM — use them to find out whether actions succeed or fail.
RPGs work like this, right? You — the GM — describe the situation. Then your players tell you what their characters do in that situation. Then you — the GM — determine the outcome of the characters’ actions and describe it.
Notice how there’s no dice anywhere in there? That’s because dice have nothing to do with roleplaying games. They ain’t how RPGs are played.
The problem’s that, sometimes, there’s some question about whether an action succeeds or fails. Sometimes players choose actions that might work or might not. And sometimes those actions carry some kind of meaningful risk or cost. And when that happens, you — the GM — use a die roll to determine the outcome.
You — the GM — decide a die roll’s needed. Because you — the GM — have decided that a dice roll’s the correct way to determine the outcome.
Die rolling’s a tool you use to help you determine the outcomes of actions you can’t adjudicate using your brain alone.
There’s no other reason to use dice. Not a single, valid one. If you’re using dice for anything else, you’re wrong. You’re making wrong GMing decisions. And if I had the power, I’d shred your GMing screen and stuff your dice bag down your throat!
Do dice provide other benefits? Yes. Are they fun to roll? Yes. Do they grant an illusion of control? Yes. Do they create a sense of unbiased fairness people are often unwilling to assume of supposedly impartial judges? Yes. Do they elevate the medium so we’re not just a bunch of kids on a playground playing Let’s Pretend We’re Elves? Yes. Do they legitimize a game about pretend elves by adding actual math? Yes. Are any of those good reasons to use dice.
F$&% no.
That s$%&’s gravy. And you don’t drink gravy by itself. Gravy’s a thing you put on top of delicious, savory meat. Meat is rolling dice for good f$&%ing reasons. Gravy’s the extra flavor dice add the game when they’re used for good f$&%ing reasons.
If you let your players roll dice even though you already know the outcome, you’re wrong. F$&%ing wrong. I don’t care if you’re doing to it give the players a sense of control or fairness. You just served them a plate of gravy with nothing under it. You moron.
If you roll dice for an action you can easily adjudicate inside your own brain just because you like to surprise yourself with random numbers? Well, you just sucked down a gravy smoothie. Dumba$&.
You roll dice to resolve actions you can’t resolve any other way. And only if those actions have some kind of meaningful impact on the game. If you roll dice for any other reason, you’re a f$&%ing Clown.
Tip: Use Die Rolls to Add Loading Screens to Your Favorite TTRPG
Imagine you’re playing your favorite video game. Maybe something like the terribly designed, inaccessible, and probably racist masterpiece that is Elden Ring. Or maybe, if you hate fun, you’re playing Horizon: Zero Clue instead.
Let’s say you fire your futuristic compound bow or normal-a$& longbow at a robot dinosaur or at a, uh, dinosaur in a dog collar. Suddenly, the game freezes. A message window pops up. “Calculating…” And you have to wait while a little d20-shaped icon spins away before you get to see if the arrow hit the dinosaur.
And let’s say that happens every f$&%ing time you do anything.
That’s dice in TTRPGs. If you disagree, you’re a Clown. Or worse, you’re a player.
Look, rolling dice is tense and exciting and fun. Shaking the dice, tossing them, watching them hit the table and bounce around, waiting for them to settle on a number that’ll literally determine whether your character lives or dies? That’s awesome as f$&%.
You know what’s not awesome? Checking your character sheet first and trying to remember where your longbow skill is. Adding up cover penalties and range penalties and ability modifiers. Adding all that crap to the die result. Waiting for the GM to compare that number to some different, secret number and then describe the actual result. Being reminded you forgot to add some other number because of this or that special effect. Readding the results. Deciding to use an ability to reroll the die and start all over. Arguing over whether you have to declare your reroll ability before the GM describes the result. And checking the f$&%ing book to find out.
Rolling dice is fun. But it’s never just rolling dice. All that other s$&% that comes with rolling dice is a speedbump to gameplay. You may think it’s not. You may think your favorite system does it better because there’s less math or something. But you’re wrong. It is. And better isn’t good. It’s just less bad.
And that’s why you never, ever roll dice you don’t need to roll. There’s plenty of times you need to roll dice. Plenty of times when dice are the only way to resolve things. The players will get plenty of gravy. And so you will. But every dollop of gravy is also a f$&%ing drag chute on an exciting, well-paced, well-narrated game. Don’t add more.
Dice Are the Opposite of Challenge
Nothing is challenging about rolling dice. Dice are the opposite of challenge.
What you have to understand is that challenge isn’t the same as uncertainty. Nor is it the same as tension. Nor is it the same as difficulty.
A challenge is something that tests you. Tests your abilities. Physical, mental, psychological, whatever. When you’re challenged, you have to use your abilities to succeed.
So, here’s a d6. You’ve got one chance to roll a 6. Is that a challenge? No! It’s just statistically unlikely. Short of cheating, there’s nothing you can do to affect the outcome. If I challenged a hundred people with that same challenge, every one of those hundred people has exactly the same odds of success. That ain’t a challenge.
RPGs are a challenge because the players’ decisions determine what dice they roll. Or if they have to roll dice at all. Players who come up with clever plans end up with lots of bonuses. Players who come up with really clever plans don’t have to roll dice at all.
Either you come up with a brilliant way to handle this situation or else I’m going to flip this coin to see if you live or die.
That’s a roleplaying game!
There’s GMs who make their players roll checks specifically because they think letting the players win without a die roll isn’t a challenge. It’s not an earned victory. There’s GMs out there who don’t use Passive Scores because they think there’s no challenge if the outcome’s already certain. And there’s GMs out there who think that you can challenge a non-existent pile of math of paperwork. Those GMs are f$%&wits. At least, they are now. Maybe they were just ignorant before. Maybe they didn’t know better. But they’ve read this. So if they keep doing that s$&% anyway, they’re willful f$&%wits. Which is the worst kind of f$&%wit.
Don’t be a f$&%wit.
Rolling Dice: The Basics
Now you really get dice. You understand the dice are yours to use when you need them. They’re only for resolving actions. They’re a drag chute on your game. And they’re the opposite of challenge. Whatever fun and tension and psychological trickery dice bring to the table, they are to challenge and pace what a crowbar is to a knee cap. And just knowing that means you’ll make better GMing choices. You’ll be looking to minimize dice rolling — to do it only when it’s the only solution — and you’ll be looking to mitigate their effects on pacing and challenge whenever you can. Good assumptions. Good reasoning. That means probably good GMing decisions.
All this s$&%’s why, ten years ago, I told you how to use dice rolls properly. And now, I’m going to review that s$&%. Because some of you weren’t here ten years ago. And some of you refused to learn it ten years ago. Some of you still get this s$&% wrong. And I’m going to add a few notes and details that’ll only make sense because you really get dice now.
Whenever a player declares an action, you must be totally, absolutely, one-hundred-thousand percent sure that a die roll is really necessary. Which means, before you let anyone at the table touch a f$&%ing die, you’ve got to know everything about the roll they’re about to make.
After a player declares an action — not after a player asks for a die roll, by the way; players aren’t allowed to ask for die rolls, they must describe actions — after a player declares an action, you — the GM — must determine what the character’s trying to accomplish with that action and how they’re trying to accomplish it. Which I’ve explained about seventeen billion f$%&ing times now. I’m not explaining that further. This site’s got a search bar. Use it.
Once you — the GM — know what the character’s trying to accomplish and how, you — the GM — must decide whether the action described has an actual, reasonable, practical chance of bringing about the desired outcome. Or making progress toward it. If your answer is, “well, technically, in an infinite universe, anything is possible,” the answer is “no, this thing is not possible.”
If an action can’t succeed, it doesn’t. Don’t roll dice. Impossible things are impossible. Describe the outcome. Or warn the player they’re about to burn resources on something that any reasonable person in the world should recognize as impossible.
Now, you — the GM — must decide whether the action can actually fail. Is there an actual, reasonable, practical chance the action might not lead to the desired outcome? Or at least some progress toward it? Again, if the answer’s, “well, anything might fail, no matter how unlikely it seems,” the answer is, “that’s a sure thing, failure is literally not an option.”
If an action can’t fail, it doesn’t. Don’t roll dice. Certain things are certain. Describe the outcome.
Once you — the GM — have decided there’s some reasonable, practical amount of uncertainty in the action, you must decide whether a failure changes anything about the situation that would keep the character from just trying again. And this is another case of actual, reasonable, practicality. If the answer’s, “they’d waste ten minutes” and ten minutes doesn’t actually change anything in the game because you don’t use Tension Pools or Random Encounters and there’s no time bomb about to go off, nothing actually changes.
If failure doesn’t mean anything, the character will try over and over again until they succeed. Which means the action can’t fail.
If you get to this point, you’ve got an action worth rolling dice for. If you haven’t gotten to this point and let anyone touch any dice, you’re a Clown. Get off my website. If you’re not resolving an action and the action’s not reasonably likely to succeed and reasonably likely to fail and failure doesn’t change anything enough that a reasonable person would stop trying, no one gets to touch any dice.
Unless the game’s telling you to roll a saving throw or damage roll. But those are stupid rules. The designers are stupid. And there’s nothing I can do about that. Fortunately, you only roll damage rolls to resolve the outcomes of Attack actions in combat — because you’d never use a damage roll to determine what happens when the rogue sneaks up on an unaware guard and attempts to assassinate them, would you? — and you only use saving throws when the game mechanics specifically say to use saving throws.
But I digress…
Now, you — the GM — must decide what success looks like and what failure looks like. You know what the player wants to accomplish. You know how they’re trying to accomplish it. Success should look something like what the player expects it to. And if it won’t, you should warn the player that they can’t get the outcome they’re trying for with the action they’re trying. You know that failure’s got to cost something. What’s the cost? The risk? What changes if the PC fails.
Next, you — the GM — have to determine which single ability score has the most to say about the outcome. It doesn’t matter what the book says. Doesn’t matter what the rules say. All that matters is the action the player described and the ability score that you — the GM — think is most relevant.
Then, you — the GM — must determine whether training in any specific proficiency might help the actor.
And then you — the GM — need to classify the action as either Easy, Medium, Hard, Very Hard, Unpossible, or whatever and assign the appropriate DC. 5, 10, 15, 20, 25, etc.
After that, you — the GM — have to determine if anything about the character’s approach or the target or the situation or anything else affects the odds of success. In other words, you dole out bonuses or penalties or advantage or disadvantage.
And you — the GM — must do all that s$&% in the order described and you — the GM — must do it all before you let anyone touch a die. Or touch one yourself. Before anyone so much as breathes on a die, you must know what success and failure look like and which ability scores and proficiencies are most relevant and what the DC is and whether there’s any modifiers.
What does that s$&% mean? It means there’s some things that you’re not allowed to do anymore.
You cannot call for a roll even if you know the action can’t succeed on the off chance a 20 comes up and you’re thereby forced to pull some ridiculous nonsense out of your a$& to turn the action into a success.
And you cannot call for a roll for an unfailable action in the hopes a 1 will turn up and fumbling hilarity will ensue.
Let me remind you, by the way, that in D&D 5E natural ones and natural twenties are not automatic failures or successes on any rolls except on attack rolls in combat. Which, as I already noted, does not encompass just any old attempt to injure, disable, or kill.
That natural ones and twenties rule is a stupid rule anyway. Even in combat.
You also cannot ask a player to roll a check then invent an outcome based on the number they roll. You must know what success and failure look like before you let a player roll a die. And you must set a DC.
You also cannot let the player dictate the ability score or proficiency used to resolve the check. Players can beg and plead all they want, strength is the primary thing that determines the length and height of your jumps and your ability to climb. Sorry. That’s a f$&%ing fact.
You can, however, adjust your perspective if a player makes a convincing argument or suggests something you hadn’t thought of or changes their action. But, if you make a habit of doing that, every action’s going to turn into a negotiation. And the game’s going to go from one in which the players are trying to figure out the best in-world way to handle the situation to one in which they’re just trying to make the strongest argument for throwing their strongest stats at every problem.
That game sucks.
Once you — the GM — decide how an action’s to be resolved, be strict and stick to your guns. Only acquiesce on very rare occasions to the players’ arguments. You’re the Mommy. You’re the Daddy. The players are your children. You know best.
Rolling Dice: The Advances
If you follow the above instructions — every f$&%ing time for every f$&%ing die roll — and you combine that procedure with the three fun facts I shared about how dice really work, you’ll mostly make good GMing decisions. About dice anyway. But based on the crapton of questions and comments I’ve received, there’s some advanced dice-rolling topics and techniques folks want some extra help with.
I’m going to run through a few of those topics now. Partly so you know how to handle this s$&% right — not that you’ll listen — and partly to show you how solid reasoning and good core assumptions can help you make the right calls without needing them explicitly spelled out for you. Which I will demonstrate by explicitly spelling them out.
Who Rolls Where
Let’s start with this question: when a player’s character takes an action, must that player be the one to actually roll the die? Or can the GM roll the die? And when should the GM hide the result from the player? Roll behind the screen?
There’s a lot of debate about this s$&%. And it’s all based, unsurprisingly, on stupid assumptions. Which ain’t a problem for you, is it?
You know now die rolling’s your thing. It’s your tool. While it’s fun for the players and creates an illusion of control and fairness and all that s$&%, none of that’s a good enough reason by itself to call for a die roll. That’s gravy stuff. Given that, it doesn’t actually matter one f$&%ing whit who rolls the dice and who sees them. If you — the GM — made every roll behind the screen and the players never even got to touch any dice, that’d be totally fine. And if you — the GM — never touched any dice at all and you let the players roll everything, that’d also be fine.
But…
That stuff may be gravy, but gravy’s nice. It’s tasty and people like it. And players are whiny little b$&%es. If you never let them roll any dice, they piss and moan. They become impossible to play with. And there’s also the fairness and trust thing. Players think dice are fairer than you. Should the players get over that s$&%? Yes. They should grow up. But they’re players. They won’t.
So, if it’s not a big deal, you should let the players roll their dice, right? Don’t invent unnecessary die rolls just to give the players checks to roll, of course. But when dice rolls are needed, let the players roll some of them. Unless doing so hurts the game.
When does it hurt the game?
First, when the die roll gives the players information that might change their decisions, that’s bad for the game. The dice don’t exist in the universe of the game. Therefore, they shouldn’t change what the characters do. But they can. For instance, imagine a player ransacks a room. She rolls a 2 on her Wisdom (Perception) check and turns up nothing. She’s going to assume she failed the roll and she’ll want to try again. On the other hand, if she rolled a 19 and turned up nothing, she’d assume there was nothing to find. She probably wouldn’t try again. The number on the die changed the player’s decisions about what the character would do.
If a player would end up with information that could change their decisions just because they rolled a check or saw the outcome, that’s when you roll the dice yourself. Away from prying player eyes.
I should mention though — because some of you take everything I say to absolute f$&%ing extremes — that sometimes die rolls convey information that the characters would be aware of. Say a player attacks an orc. He rolls a 19 on his attack roll and misses. That tells him his enemy has very high defensive stats. It’s armored or evasive. Whereas if the player rolled a 2 on the attack, he’d assume his character flubbed the attack. He stumbled or his sword shifted in his hand. Something. And if people can see that s$&%, it takes some pressure off — or reinforces — your narration. Everyone can intuit that s$&%. That’s why, generally, rolling in the open is good. Whether you roll or the players roll.
Second, dice rolls release tension. When a player decides their character acts, they’re tense. The outcome’s uncertain. Once they know — or can guess — the outcome, the tension dissipates. Usually, that’s okay because the outcome’s immediately obvious. When you search a room, you know when you find something good. When you stab an orc, you know if they die. But sometimes — with certain protracted actions — the outcome’s not immediately obvious. And sometimes — to maintain the game’s pace — you — the GM — ask for die rolls before you actually need them.
When a rogue’s sneaking down a hall, for example, she doesn’t know she’s failed until she’s been spotted. Unless, of course, she rolled the dice and got a 6 on her Dexterity (Stealth) check. She knows then she’s gonna fail. She’s just waiting for the moment. So whenever you want to preserve tension for an extended scene or withhold an action’s resolution, roll that s$&% behind the screen.
The third thing’s all about pacing. Every die roll’s a f$&%ing speed bump, right? Well, in the gamiest parts of the game — like combat — that’s no big deal. But some scenes can’t handle gamey speedbumps. Social interactions, for example, they gotta flow. Ever tried to have a conversation with a five- to ten-minute delay between everything anyone says? It sucks. So, whatever die rolls you need to run the scene, you roll them quickly and quietly behind the screen. Pick up a die, toss it, check the result, respond accordingly.
In case you’re wondering, all those things — hidden information, tension, and pacing — trump players like rolling dice and players trust dice more than you. Thanks for asking.
The Master List
Now, you didn’t ask, but I know you’re gonna. Because I’ve learned that s$&% that’s simple and obvious to me is actually galaxy-brain level GMing that only I’ve ever been smart enough to figure out.
How do you roll checks for the players’ characters without asking for their stats or taking their character sheets?
You keep a f$&%ing list. It’s like this: you — as a GM — must be able to roll any check for any creature at any time. Quickly and easily. So, you keep a list on which you’ve noted every character’s ability scores, proficiencies, proficiency bonus, armor class, and whatever other s$%& you think might be useful on there too. It’s a simple a$& spreadsheet any one of you could make on graph paper or in Microsoft Excel. Hell, you can fit that s$&% on an index card if you remember you don’t need to math out every skill roll.
If you’re a GM and you do not have that list in front of you at all times, you’re just wrong. You’re a bad GM and you should feel bad. And yes, it should be in front of you at all times. It must be self-contained and always visible. Having every character as a PDF in your computer open in a different tab is not good enough. You have to be able to see all the stats for everyone in one f$&%ing glance. Because pacing.
Let’s Fight About Passive Scores Again
Or rather, let’s not. At this point, if you’re not using Passive Scores the way I’m about to describe — which is, more or less the way the PHB says you should use them — you’re made a conscious decision to suck at running games. There’s no other reason. If you want to suck at running games, that’s fine. But please don’t share your suckage on my website. This is where people who want to not suck come for help.
You — the GM — use dice rolls to resolve actions. Actual, conscious, declarable, deliberate actions. An action’s something an audience could see and understand on a movie screen. “That character is searching the room for something,” they might say, or, “that character is trying to kill an orc.” And things that represent an internal decision the character made. “That character has decided there must be something hidden and has resolved to find it.”
Being aware of your surroundings? Noticing s$&% because your eyes, ears, and brain work? That’s not an action. Neither is “looking around really extra alertly.” Searching actively — poking, prodding, moving around to get a better view, tapping, touching, smelling — is an action. Noticing s$&%’s not. Same with recalling information because you saw something you recognized and your brain lit up. That’s not an action. Nor is intuiting s$&%. Nor is reaching conclusions based on visible clues.
Things that aren’t actions don’t get dice rolls.
The creators of D&D 5E — and I can’t believe I’m going to say this — the creators of D&D 5E were actually very smart about that. They invented the rules for Passive Scores. In any situation in which a character might notice, intuit, recall, reason, or conclude something without a deliberate, conscious action, resolve the outcome as if they’d rolled a natural 10 on an appropriate ability check.
If there’s a thing hidden in a room and the DC to find the thing is 15, any PC will notice the thing if they have a Passive Perception score of 15 or better. They just notice the thing is there. No roll needed. No search needed. The thing is not hidden well enough to escape the alert character’s notice.
That’s not just the right way to handle it, it’s the only right way. Using a die roll to add some challenge and unpredictability? That’s wrong. Changing the DC? That’s wrong. Doing any other thing than comparing a Passive Score — 10 plus the PC’s total modifier on the relevant check plus or minus 5 for advantage or disadvantage — is wrong. F$&%ing wrong.
You might think you have a valid counterargument. You don’t. I’m sick of having to read supposedly valid counterarguments. A valid argument must be based on valid assumptions. I already listed the only valid assumptions about die rolls that exist.
Guess what? That’s why it’s also correct to set a DC for s$&% like identifying monsters and knowing their weaknesses and then compare the DCs to Passive Scores like Passive Arcana and Passive Religion. Because recognizing things and recalling information about them is not a deliberate, conscious action. And because adding that s$&% to scene-setting narration improves the game’s pacing. And because rolling dice is the opposite of adding challenge.
Want to use a proficiency to get information that you don’t already have in your head? That’s called research. Or experimentation. Or study. Or interview. You can totally do s$&% like that. Go to the library, the lab, the temple, whatever. But what you can’t do is think really hard to try extra good to recognize and remember s$&%.
Rolling Against Opposition
Let’s talk about opposed rolls. Or, as the PHB calls them, Contests. F$&% do people overuse these things. Usually because they’re using them when they shouldn’t.
It’s not like the PHB isn’t straightforward about them. You use a contest when you’ve got two creatures actively working against one another to accomplish something when only one can succeed. Both creatures roll appropriate checks and the higher roll wins. And that “actively working against each other” s$&% is the super important bit that everyone misses.
Why? First, because two die rolls are worse than one die roll. If you’re trying to minimize the dice rolling, you don’t want to roll two dice for one thing. Second, because you only roll dice to resolve actions. So, for example, if one creature — say a PC — is sneaking and another creature — say a guard — might notice them, only one creature’s actually doing anything. The PC rolls a Dexterity (Stealth) check — though you probably roll it in secret — and the result’s compared to a static number. Say, the guard’s Passive Perception score.
Except not. I’ll come back to that.
Later, if the PC’s hidden and spying and the guard suddenly decides to do a perimeter search, the guard rolls a Wisdom (Perception) check against a static number. Because now the guard’s doing actual action s$&% and the PC is sitting still.
You only use contests when there’s two actions and only one can win. Like two creatures grabbing for the same item. Or tug-of-warring. Or arm-wrestling. You don’t use them when one creature’s actually doing action s$&% and the other’s just resisting it.
Now, let’s talk about that Passive Score crap. It’s perfectly fine when the PCs are doing the resisting because their exact stats and skills should affect the outcome. But when an NPC or object’s doing the resisting, you’re better off just assigning a difficulty — Easy, Medium, Hard, Really Hard, Impossibly Hard — and using the appropriate DC — 5, 10, 15, 20, 25, or whatever. Seriously. What is really the difference in the long run between checking the rogue’s Dexterity (Stealth) against a Passive Perception of 11 or just a DC of 10? Not enough to make it worth slowing the game for.
In short, contests should be really rare. They resolve action versus action. And that’s it. Otherwise, a roll against a Passive Score, an already rolled result, or a static DC works just fine. And, frankly, unless the resistance is a PC, just use a static DC and keep things moving.
Teamwork Makes the Dream Work
Finally, let’s talk about cooperative and collective actions and group rolls. Because, for some reason, this s$&% really f$%&’s with GM’s heads. Mainly because they don’t actually use their heads before game mechanics start coming out of their noise holes. Neither do players. Which is why you have stupid bulls$&% like players saying “I aid the Persuasion” and the GM saying “okay!”
So let’s talk this s$&% out.
First, there’s a few different kinds of group rolls. That is, there’s a few different ways where you’ve got multiple creatures — usually PCs — participating in an action. First, there’s cooperative tasks where the participants are directly assisting each other and succeed or fail as one. Two healers working on the same patient, for instance. Or four characters using a battering ram on one door. Those are situations where’s there literally one action and several people doing it together.
There’s limits to this s$&%. And the PHB is clear that you — the GM — should set limits. It’s hard for even two people to work together to break down a door without a battering ram or a long prybar or something. And there’s only so many cooks that can help cook a single pot of soup before they just ruin it. That’s why we have that saying.
Second, though, there’s times when people are all trying to accomplish the same thing, but they’re not really working together. Either they’re splitting a task between them, or they’re all trying to succeed and, at best, maybe they’re watching over each other and helping as best they can. For example, maybe two cooks can work together on the roast — one basting and spicing, the other managing the fire and turning the spit — while another cook’s roasting the potatoes and a fourth is making the pie for dessert.
When a group searches a room, they’re not really working together. Not really working as one. Instead, each is conducting their own little search. Maybe they’ve divided the room into sections to save time. Or maybe they’re all going over the same space thoroughly. But, either way, everyone’s doing their own thing. And while having a second set of eyes check what you checked can prevent you from overlooking something, your careless ransacking can actually make it harder for someone to find something you missed. And people are unconsciously less attentive to tasks they think other people have already done than on their own tasks.
Same’s true of stealth. Yeah, it helps to have some really good sneakers picking the path and giving hand signals and telling people when it’s safe to move and where to go, but everyone’s still doing their own thing. Same’s true, by the way, of picking through a swamp and avoiding quicksand and poison brambles and things. Which is the example the PHB uses for these sorts of group checks.
Yeah. There’s actually rules for this s$&% in the PHB. Good rules. Rules you should follow. How do you use them?
First, when you’ve got something that looks like teamwork, you — the GM — must decide whether you’re dealing with real cooperative teamwork wherein multiple people are taking one single action or whether you’ve got a situation wherein everyone’s doing their own thing while also doing their best to help each other with their things.
Cooperative teamwork’s resolved using the Working Together rules on PHB 175. The participant with the best modifier rolls a single check with advantage. Provided you — the GM — already decided teamwork’s useful and possible. And provided the task’s one the helper is capable of helping with.
Group actions are resolved using the Group Check rules on PHB 175. Everyone in the group makes an individual check and if at least half the checks are successes, everyone succeeds. Otherwise, there’s a failure to deal with. Either one of the dumba$&%es blundered into a pit full of quicksnakes or the inattentive f$&%ups have trashed the room in their search so thoroughly, an elven ranger named Matt Murdock couldn’t turn up anything.
Those rules work totally fine. Better than fine. They’re pretty damned good. And everyone forgets they exist. Hell, I forgot they existed and thought I was going to have to write my own group check rules for this s$&%. But it’s been a long time since I decided I wanted to stop running a good game and opened the D&D 5E rulebook.
But this article’s not about rules. It’s about dice. Rolling dice. If rolling one check’s a pain in the a$&, rolling five’s an utter disaster. At least in terms of pacing. So what do you do? How do you resolve it quickly and efficiently?
You do it yourself. That’s the only way. If you involve the players in this s$&%, it’ll take forever.
Look, you’re the GM. That means you’ve got a Master List. Everyone’s proficiencies and skill modifiers on one, single cheatsheet. And you’ve also got a fistful of dice. Use them. Grab a bunch of d20s — one for each PC — and throw ‘em all at once. Then, read them across the desk or table. Leftmost die’s a 17 and Alice has a +3? She got a 20. Nextmost die has a 12 and Bob has a -1? 11 for Bob. Run across the results, comparing each to the DC. Slide the dice around if you have to. Roll them right on your f$&%ing list, slide each die to someone’s column, move the successes up and the failures down. Do it a couple of times and you’ll get good at it.
I can roll a Dexterity (Stealth) check for six PCs in the time it takes one player to say, “wait, which one’s the twenty-sider again? This one? No that’s a twelve. Uh…”
Meanwhile, if you use a computer and an online die roller or you run games online, there’s a quick and easy way to roll your group checks and your secret die rolls efficiently. Print out a f$&%ing list, put it on the desk in front of you, and buy some actual dice!
It is way faster to quickly grab and toss dice than it is to move between documents, screens, and windows and click in multiple places. Especially after three sessions of practice. You can prove it to yourself with a stopwatch. Or don’t. Just believe me.
If you care about this s$&%, work analog whenever you can.
And you should care about this s$&%. Because you’re a mommy or daddy now. You don’t get to be stupid anymore.
As one of the Clowns: thank you for this article! I realize I’ve been overengineering my adventure to make up for my lack of confidence in picking the right ability, skill and DC while running the encounter. But if you never dare to make mistakes running a session, you’ll never learn how to do it right!
I swear to GAWD I thought the headline was a euphemism!
There are good arguments against rolling for initiative in here.
Recently I’ve just been keeping a list of PCs ranked by dexterity, with another list for NPCs. When I need an initiative order I put whoever instigates the fight at the top, then go down the lists alternating between PCs and NPCs. I’ve been quite pleased with how it has played out at the table so far.
FINALLY! VALIDATION! Angry says I’m not a bad DM!
“You keep a f$&%ing list. It’s like this: you — as a GM — must be able to roll any check for any creature at any time. Quickly and easily. So, you keep a list on which you’ve noted every character’s ability scores, proficiencies, proficiency bonus, armor class, and whatever other s$%& you think might be useful on there too.”
I have a master list, has it since 2015. I’m willing to share my version.
“I can roll a Dexterity (Stealth) check for six PCs in the time it takes one player to say, “wait, which one’s the twenty-sider again? This one? No that’s a twelve. Uh…”
I can do it even faster with an excel sheet with random d20 rolls on them to resolve groups and shorten the “waiting around to die” phase of combat.
I am disappointed because I was waiting for, hoping in fact, for Angry’s to blow up about dice trays and dice towers.
I completely share the frustrating with dice towers, but dice trays seem like not terrible inventions? I mean they remove the issue of dice rolls constantly going onto the floor which sucks up enough table time to be notable.
I imagine the answer is to learn how to roll dice without throwing them halfway across the room. Some tables/game rooms have ample space for trays, and I really like my own dice tray. It’s cheap, it stores my favorite sets securely, and it’s more convenient for travel than a lot of baggies. But they’re also things people buy because they’re dice goblins who put too much time, focus, and attention on the dice (note: guilty. I love fancy dice, though dice towers annoy me greatly). The point of the article – and the half-dozen Angry’s written on more-or-less the same subject before this – is that dice are a necessary evil. The more you muck about with Dice and Dice Accessories, the more you’re not mucking about with playing the game instead. Or at least, that would be my guess.
I am a proponent of the dice cup. Preferably leather. Into which I put a d20, rattle, turn over, and leave alone. I only lift the cup when DM calls for it or for initiative. In combat, I put my damage dice in as well so I don’t have to slow the game even more when I hit something.
Trays take up more table but are good, towers take up less table and are unnecessary evils brought upon us by dice goblins.
Also, I want you to throw the dice where I can see them because you take too long to do math and I can tell you at a glance most of the time whether you succeeded. If you roll in your own private little corner of the table, I can’t do that.
“Okay, so that’s… 11 plus… umm… 3… no… umm… 4… maybe, so…”
“The DC was 10. You succeed. Shut up.”
“…carry the 3…”
During a recent Patreon Live Chat, I did in fact blow up about dice towers and dice trays and crappily designed, unreadable dice with awful fonts and symbols on random faces and all sorts of dice-related crap.
I kid you not, I checked discord as I was reading this and saw someone say that the party wizard lifted a portcullis after the party ranger and fighter failed, and the very next comment was “was there time pressure?”.
Aside, I can’t help but feel that many GM’s probably wouldn’t call for pointless checks if the primary action resolution mechanic didn’t depend on a d20. If the modifiers only go up to +5, but the dice is anywhere from 1-20, then I think GM’s will call for ridiculous checks because the die matters the most in this equation. Obv this doesn’t apply as much to systems with bigger modifiers or different dice mechanics, so I’m mainly glaring at 5e DMs.
I had exactly this issue of the randomness far outweighing the skill modifiers while preparing my adventure, but it’s solved so nicely by the advanced tips in this guide. I can’t believe these are taken from the PHB when none of my DMs have ever used these rules. Though I must say Angry explained it far better here than the short section in the book did.
You know that little section at the beginning of every pen n paper PHB? The one that describes game flow. GM describes something, players respond, etc etc…
It should be a branching flow chart for a generic situation that might or might not lead to a die roll based on the example player’s decision.
As they are now (IIRC), they present the example play as always having dice and only as a binary outcome.
If only there was some kind of gaming genius with anger towards bad games that was making his own game that could include a better example of play. If only.
I know you’ve not specifically ruled it out, but I remain a big fan of ‘Roll a 20 and don’t get a 1 and you succeed’. ‘Very unlikely’ seems to come up much more often than ‘impossible’ – especially if your players get used to the idea that asking to seduce the dragon is going to get them eaten.
Yes, a trained assassin, having successfully sneaked up on a guard from behind, will slash that throat 95% of the time. A straight D20 roll with a DC of 2 is a great tool to have at hand – and with no modifiers or sheet checks, it doesn’t act as the drag parachute. Low risk, high impact.
It’s useful, sure, but it doesn’t serve the same function that most players/GMs think it does.
DC 2 is the equivalent of having a 95% chance of success. Those are XCOM odds. You should pull those out of your pocket whenever a PC is doing something “certain” within a tense and high risk situation, not when a character is casually pouring themselves a drink.
These are things like “short hops” over steep chasms, or shooting a sectoid with a pistol at point blank range. Anything that logically would seem like a sure thing, but that would still make your stomach drop for a second when you actually do it.
Yeah I disagree. I think that rolling the dice just to see if they get a 1 seems like you’re (as in, DMs that do this, not you in particular) looking for opportunities to screw over your player. Usually, the “roll a 1 and you fail” just leads to the player feeling ridiculous when their acrobatic rogue can’t jump 2 feet. It just leads to a wacky game rather than a suspenseful one.
From the article: Again, if the answer’s, “well, anything might fail, no matter how unlikely it seems,” the answer is, “that’s a sure thing, failure is literally not an option.”
A few problems here:
1) Not using a DC negates the choices a player made that would have given their character a bonus – class abilities, skills, equipment, etc. This is bad. No way around it. It will feel wrong to many players, it breaks the rules, (remember a crit fail on a skill check isn’t a thing) and it’s “unfair.”
2) In general, I’d say it’s rare to be confident of the odds to within 5%. I would think that in most of these cases, the actual odds are going to be much lower, and you’re forcing a larger rate of failure upon your players, who, as Melon pointed out, are likely to feel screwed. After all, if you’re dealing with odds this low, it’s probably an action the character is made for. (literally)
3) Even just picking up the die, rolling it, and reading it takes time. Not a lot, but some. It’s still a break in the pacing, however minor. Is a 5% chance really worth that? Even if you remove the other concerns, I think it’s usually not going to be.
All that said, at the end of the day it’s your game. if you and your players all enjoy it, go for it. I’m just some chump on the internet.
If the PC is searching the room for hidden doors, and there is no hidden door to find, that means success is not possible and only a Clown would call for a perception check, right?
…
F$&% me. You found a plot hole.
It’s honestly a very tricky issue. And part of the problem with these rules of thumb. The thing is, GMing is ultimately an intuitive process. It’s about making good decisions in the moment based on a whole host of incomparable factors that can’t be quantified.
Part of the issue is that you really can’t let the act of rolling dice or not rolling dice communicate information that would change players decisions. And there’s a definite difference between “you didn’t find anything” and “there’s nothing to find.” By not rolling dice in a situation wherein the characters tell the difference between a failure and an impossible situation, you’re differentiating the situations. Which means there’s no chance the party would waste a bunch of time continuing to ransack the room at greater risk to themselves of being discovered by some wandering beast because they’re convinced there’s something to find. I could do a bunch of word gymnastics and say that one key is to consider success and failure from the players’ perspective. Or from a neutral understanding of the platonic reality. But that isn’t just word gymnastics. It’s just an intuitive and different understanding. “Could this action – searching – reasonably turn up anything that might be hidden and could it also reasonably fail to turn up anything that might be hidden.”
Truth is, though, your game won’t get broken if you don’t roll search rolls when there’s nothing to find. And you’re actually the first one to notice that hole. Hell, I didn’t even notice it. Which has only deepened my understand about how I parse things like this at the table. Because I know my four rules are right and I know it’s right to roll hidden checks whenever not rolling would reveal information to the players the characters wouldn’t have. I either need to reexamine my rules of thumb or reconsider the priority order.
I’m not sure it’s such a big hole. You don’t actually need to do the whole rolling process to avoid giving the players the information, you just need some dice sounds to come from behind the screen.
After deciding the action is impossible you can ask yourself if the player should know it’s impossible. If the answer is no, roll a die, pretend to look at it, then tell the player they failed.
I agree. But it is still a place where the advice I gave doesn’t match what I do. And doesn’t lead to the intuitively best result.
Thanks to your unequivocal declaration that “Die rolling’s a tool you use to help you determine the outcomes of actions you can’t adjudicate using your brain alone” I’ve been a bit unsatisfied with my initial solution. Thinking about the problem some more I’ve decided that “Does the character find the thing” is generally not the right question when players are searching. The outcome the player is trying to achieve is probably “find the thing or determine that it isn’t there”. In general, a better question is “does the character stop searching before…” with the assumption that a character will stop searching when they either find the thing or are convinced the thing isn’t there.
Player wants to search a room looking for a non-existent secret door with no time pressure? No question, no roll.
Player wants to search a room looking for a non-existent secret door and there’s a basilisk wandering the dungeon? “Is the character convinced there are no secret doors before the basilisk wanders by?”
Player wants to search a room looking for the secret door they read about and there’s a basilisk wandering the dungeon? “Does the character find the door before the basilisk wanders by?”
Player wants to search a room looking for a secret door while breaking and entering and they declare they won’t spend more than a couple minutes for fear of being discovered? “Does the character find the secret door before they give up and get out of there?”
I can offer you a different answer in two words: Tension Pool.
My point is not that the game is better with time pressure. My point is that when a player wants to look for something the outcome they want is not “find something” it is “know whether something is there or not”. Recognizing that is important for adjudicating actions. It is impossible to find something that doesn’t exist, but it can be possible to determine whether or not something exists.
You can only prove that there is a secret door or that you haven’t found one yet. No amount of searching can prove that there is no secret door if it is sufficiently well hidden.
I can’t prove that unicorns aren’t real, but after a sufficient amount of research I can be convinced that unicorns aren’t real.
I occasionally roll random dice behind the screen when players are talking about their character’s actions, just so this doesn’t some up. They mean nothing, I give them no extra thought, but the players don’t know why I rolled and have no idea if my next bit of talking had anything to do with the dice or not
I feel there’s one additional layer to consider before touching the dice:
Will either result put the story in a fundamentally unacceptable place?
I’m assuming that you’ve talked with your table about the type of story you all want to have. Yet, within the fiction, there are always actions a character could plausibly attempt that, either by success or failure, will carry the narrative way into the badlands of “stuff we actually don’t want to have at our table”.
(“Yes, technically, you COULD overload the reactor of the party’s spaceship…”)
Sometimes, it’s an indication that a player doesn’t actually care (anymore) about the type of story you all agreed upon.
Sometimes, this isn’t even obvious to the players – maybe Greg just wants to tune up the reactor a bit for a tad more power, not knowing that an undiscovered sabotage in most imperial ship reactors is planned to be a major plot point a bit down the road. How you deal with that is up to your DM brain, but you SHOULD deal with it.
Generalized, make sure you and the players are on the same page about what the character is trying to accomplish and they have at least as much information as their character about how reasonable that attempt is.
If failure leads to a situation you do not want, then there should be no chance of failure thus no die roll. But there is a chance to rewrite your homebrew so you aren’t keeping essential information gated by a die roll. DMs gotta consider consequences of failure as part of this flowchart to dice rolling.
Sure, I mean, I’m literally talking about having a recurring table event of ‘roll a D20 and don’t get a 1 or you fuck it up in comedic fashion’. You pull that a few times, it becomes a thing and players know where they stand.
When that 1 comes up, no one complains as it’s an established thing. And who among us has not attempted to slash a throat IRL and just absolutely flubbed it? It’s rare among the pros but it still happens.
This comment got posted in the wrong place.
On the topic of giving players information based on the number they rolled, there’s probably subtle ways to telegraph this even when a roll is secret. Say, a secret stealth check – if they roll low, the character makes a mistake and gets detected, whereas if they roll well but the DC is just high, they get spotted despite not making blatant mistakes. Gives the players something more to go off (oh, these enemies are hard to sneak past, let’s try a different approach if we encounter more). Not necessarily something that needs to be strictly followed or codified, but it’s more fun to kill PCs because they had information and didn’t use it than because you made them stumble around blindly.
I used to be a big fan of opposed rolls. One day I realized exactly what they are and I have never used them since: an opposed roll just randomizes the DC.
It’s bad, stupid design. Just set a DC or use the opponent’s passive value (i.e. assume they rolled a 10).
Yeah, they’re not really doing anything *mechanically.* But from a UX perspective, opposed rolls feel like active contests in ways that rolling against a passive dc doesn’t.
Is this “gravy” (as Angry put it) worth it? Not often. But it’s still a useful tool to have for those few situations where the ‘feel’ of the game is worth the slowdown. (I.e. really pivotal moments.)
It didn’t come up, and I don’t think 5e uses it much, but either way I think I subconsciously apply degrees of success to dice rolls as well. I can usually think of a dozen ways something might resolve, either good or bad, along a spectrum of possibilities, and degrees of success help me identify appropriate results. It’s definitely more intuitive than mechanical. I tend to like the results. If a roll succeeds or fails a DC by a full step (5pts usually) you stand to gain or lose appropriately. It’s kind of like a dice version of “yes/no and/but”
I also like to describe things like the difference between dodging an attack or deflecting off armor, for instance. It’s not hard to put in the notes (or just remember) AC14 = 10 to hit, +2 from DEX (dodge), +2 from armor (deflect). It sounds like extra math, but once you get used to it, it becomes natural.
But out of game, in my own time, sometimes rolling a fistful of dice and adding them up is fun in and of itself. Everyone should try using 11d10-10 for d% once in a while
You might already know: 11d10-10 is not very similar to 1d100. The minimum and maximum are the same, but the odds are extremely different. If you don’t know, now you know.
Yeah if you’re rolling d% (d100 in a percentile system) then you don’t really want to be adding a fistful of dice instead of rolling one die (and yes, a d100 split up into two digits is mathematically a singular die).
However, I could think of a ton of really cool uses for 11d10-10. Probably in a video game, though, because computers are a lot faster at math than people are.
Fun math fact: 11d10-10 is pretty close to a normal distribution thanks to the central limit theorem. There are, in fact, a huge number of really cool uses for normal distributions, though computers have much more efficient ways of generating them than by simulating dice rolls.
Oh yeah, I know it’s different than a d%, and I’d never do it in game, and it circumvents the idea of a d% table in which I can redistribute probabilities anyway. But it is fun to roll a ton of dice and add them up every once in a while
How would you deal with abilities that modify checks on demand (eg. Bardic Inspiration) when rolling the checks as the DM? What about modifiers that happen on the first ability check after they are achieved (eg. Guidance)?
If Alice tries to persuade the guard and she has Bardic Inspiration from Bob, would you still roll the ability check behind the screen to keep the game moving, or would you make them roll so they can decide to use it (or just know they are using it in the case of Guidance)?
I’m realizing these features are like the mother of all drag-chutes when it comes to pacing, but they also seem very important to just ignore them.
There’s a central issue that makes me despise a lot of these, and the issue is that many of them – though not all – are really built around the perspective of the die roll being a thing. If you provide a bonus, for example, to someone through a conversation by making them more silvertongued or granting a blessing of karma, that’s one thing. But when the rules mention “on your next check” or “when someone makes a check,” that creates this sense that the check is a “thing” to be influenced. And while I realize it’s a matter of perspective to say, “well, it’s not a really a check, but an action,” the problem is that most “actions” are sort of arbitrarily separated. The one thing said to Persuade out of an extended conversation, for example, or the ten-minutes worth of searching a very large room when wandering monsters or Tension Pools do provide a ticking clock.
If the only way to decide to use the action is to acknowledge the game’s underlying mechanics, it’s a yucky area. That’s it.