Dice: Beautiful and Terrible as the Dawn (Whatever THAT Means)

Print Friendly, PDF & Email

June 26, 2019

Something like ten freaking years ago, I wrote this article about how to actually run a role-playing game. I didn’t explain the rules, mind you. It told you how to actually use the rules to run a game. Something I would have thought that every RPG rulebook should have included. But, I guess I’m the crazy one for thinking that might be important. Thing is, I explained that the rules were just part of the game. Because the game was really about the GM setting the stage and inviting the players to act and the players taking actions and the GM determining and describing the outcomes. And the rules really only existed for that last part. The determining and describing actions. Which I called Action Adjudication.

Well, it turns out that I was right and the game designers were wrong. Because everyone loved that article. And it became the basis for a decade of farting out advice building on that basic framework. It started with general rules for when to use the dice and when not to use the dice. It grew into how to string actions into encounters and string encounters into adventures and all sorts of other crap. And I even managed to polish the same damned advice and repackage it in book form.

Recently, though, I turned all of that upside down when I started talking about another, completely different way to adjudicate actions. One that wasn’t based at all on a nice, systematic approach. One that was based on the idea that every action is a proposition bet in which the players put something on the line in return for a chance at a desired outcome. And I promised that, eventually, I’d present a better approach to adjudicating actions. One that takes into account the systematic Lawful approach I presented a long time ago and the wild Chaotic approach of just tossing a gamble at the players and seeing if they take it. That is to say, the Lawful approach where you use the tools and the Chaotic approach where you throw the rules out the window and say “well, maybe that’ll work, but what’s it worth to you?”

This isn’t that article. Sorry. It’s coming very soon. But there’s something we need to address first. Because there’s something at the heart of role-playing games that everyone uses, that everyone accepts, that everyone embraces, that everyone loves, that everyone can’t get enough of, and that nobody but me seems to actually freaking understand. Because it’s something both great and terrible. Something that seems utterly Chaotic but something that is actually heartlessly, cruelly Lawful.

It’s dice. Dice are great, but they are terrible. And I’m the only one who gets them. And since one of the most important parts of INTELLIGENT, REASONABLE, RATIONAL action adjudication is knowing when to use the dice and when not to use the dice, well, you just can’t be a good GM if you don’t understand the dice. And this is going to be central to Angry’s All-Encompassing Action Adjudication Axiom. Which is apparently what I’m calling the big theory I’m building toward.

What the Dice Represent

Normally, I’d start by talking about why the dice exist in the game and what they do. But I want to start by addressing an idea that several readers managed to slam out with their keyboards and send to me despite the fact that they are obviously missing key brain lobes. Because it was a stupid idea. It was in response to my two recent articles about action adjudication. The one about flexibility with ability scores and the one about gambling and Chaotic GMing. And the idea runs like this:

“When a player rolls – for example – a Strength check, the result on the die represents the actual physical force they bring to bear on the task. Nothing more. Nothing less.”

I had several different versions of that mental diarrhea splattered across my inbox, using different examples and stated in various different ways. I chose that because it was the simplest, the shortest, and the clearest. And the person who stuck that in my comments section should not be proud that I’m using their comment. Because I’ve used it to appoint them the ambassador of Moronistan solely by virtue of the fact they seem to be the most literate of all their countrymen. Which is not a high bar by any stretch. That’s like electing one particular fourteen-year-old as the community representative because he’s got the most social skills of anyone on Fortnite.

Anyway, the idea is that, for every task that requires a die roll, there is a certain amount of “work” or “effort” required to accomplish the task. A door requires precisely 120 foot-pound-newton-joules of power to break down for example. Or dodging the fireball requires an absolute reaction time of 0.3 seconds and an acceleration of the body of 25 feet per second to get clear of the blast. That effort, that work, that bullshit is what goes into calculating the DC. When the player rolls the dice, they determine exactly how much of that effort, that work, that they generate. They generate some random amount of effort, modified by their physical and mental capabilities. And if that effort exceeds the amount of effort required, well, then the door breaks or the fireball is avoided or whatever.

Now, here’s the problem with that idea. It’s utterly stupid. It completely misunderstands not just randomness and game design and the how D&D works, it also misunderstands how the actual world works. Like, I’m not sure the people who hold that idea have ever actually tried to do anything that required any actual effort before. And if they did, it’s a wonder they survived.

Let me explain what the dice represent. They represent nothing. Effing nothing. And everything. They represent everything and nothing. And they never represent the same thing twice. They are a complete abstraction. And trying to explain them is like trying to explain the science behind the magic in the Harry Potter universe. Not only is it pointless and impossible, but it also completely misses the effing point and ruins everything fun about the Harry Potter universe.

Explaining what the dice represent is like taking a spiritual, magical, all-encompassing force of the universe in a beloved franchise that was evocative precisely because it was beyond the realm of the physical laws and it was completely unexplainable and explaining at it as microcellular organisms. And establishing a blood test for that crap.

What the dice really represent is the limits of human knowledge, human perception, and determinism. Look, when you try to break down a door, in theory, with enough scientific equipment, you COULD determine exactly how much for you need to deliver to exactly which spot on the door to pop the door open. But you’d have to do a lot of measuring. And, if the door is outside of hermetically controlled laboratory conditions, you’d better work fast because all of your data is going to become useless pretty quickly. I mean, the air temperature affects both the elasticity of the door and the contact force between it and the door frame because things expand and contract in the heat. Oh, and the humidity plays a part too. And a thousand other factors as well. This is, by the way, the essence of chaos theory.

And then, too, you’d have to deliver that force very precisely and to a very precise spot. Super precisely. And humans just aren’t that precise. Some are more precise than others. Some can control their force and accuracy very well. Within a certain margin of error, of course. And that margin of error is actually pretty broad. And beyond that, all sorts of factors will affect your accuracy, your speed, your power. Did you get enough sleep last night? How long have you been walking today? Are your muscles fatigued? Building up lactic acid in there? How’s your sugar level? And did you notice the slick patch on the floor in the poor lighting before you did your run up? Because that’s going to through you off balance. And if you have a beta carotene deficiency that’s affecting your eyes’ ability to adjust to different light levels, it might even be harder to see it.

The idea that the DC and the ability score represent anything precise, anything more than a rough estimate of the odds that this particular person might accomplish this particular task roughly two-thirds of the time assuming all conditions are perfect and that the dice represent anything other than “nothing is certain, everything is chaos, who the hell knows how things will turn out?” That idea is stupid. It’s stupid to try to explain what the dice are. Because they are everything. Everything that might be overlooked. Everything that can’t be predicted. Everything that can’t be perceived. Every tiny variation. And that’s before we remember that this is also a world where magical fields permeate the universe, where karmic justice is a property of the universe, and sometimes the gods get pissed off and screw with you for no good reason.

If you’re not coming into action adjudication with that mindset – that attributes represent potential, that DCs represent approximations for how difficult a task is for a reasonable person under reasonable conditions, and that the dice exist because the universe is unpredictable – well, you can’t run a good game. Your brain is not up to the task of running a game. Sorry.

With that long rant out of the way – and that’s what it was, a rant, because shit like that just pisses me off, especially when people speak with such conviction and say “well, actually, your entire thesis is wrong because this is absolutely what dice must represent because I can conceive of no other thing,” and then my blood starts to boil and I just want to beat someone to death with a copy of Jensen and Sibani’s Stochastic Dynamics of Complex Systems

Sorry.

With that long rant out the way, let’s talk about dice. Why dice are a part of the game, why they’re great, and why they’re awful so that you – a GOOD GM – have some useful criteria for deciding when to solve a problem with dice and when not to.

Why Dice at All

Here’s an interesting question very few people ask. Or ask properly. Why do RPGs use dice? The reason I say no one asks this properly is that the usual answer is “they add tension,” or “they add a random element.” But those aren’t good answers. Sorry. They are technically true. But they don’t answer the question.

Now, this is where I’m going to make a statement and pretty much just present it as a fact. I’m going to gloss over the long explanation. And you’re just going to have to damned well trust me. And if you don’t trust me, well, that’s your problem. Because I’m right. The thing is, this particular piece of information is important, but it’s only tangentially related to what I’m talking about and if I try to explain it in detail, I will ramble on for five thousand words just on that one subject. So, here’s the fact:

Human beings need to be surprised. We need to be caught off guard. And so, we have this hardwired tendency to find surprises – to find the unpredictable – exciting. Within limits. See, it’s actually a very complex interplay of forces in our brain. But, in effect, being surprised is a way of training our brain to exist in the universe we occupy. Because the universe we occupy is utterly unpredictable. And our greatest strength is our ability to reason, plan, and imagine. To understand hypothetical causes and effects beyond the moment we exist. And we enjoy playing games – on a deep, fundamental, biological level – because it trains our brains to exist in the universe.

Long story short – games need to have an element of surprise, they need to have an element of the unknown, they need to have an unpredictable element – to be fun. Well, that’s one of the things they need.

“Aha,” says you. “See, Angry, you answered your own question. That’s why games have dice.”

“Okay, then,” says I, “where’s the dice in chess or Pokemon: the Gathering?”

See, games do not need to dice – or even randomness – to surprise us. Games can surprise us simply by hiding information from us. We can guess what our opponent in chess might do in response to this or that particular move, but we don’t know for sure, because we can’t read minds. We have to guess. And maybe we are right. In which case feel smart for making a good choice. Or maybe we are wrong, in which case we have an exciting scramble to adapt to a different outcome from the one planned. In Pokemon, we might know exactly what cards we have in our deck. Maybe we even know the cards in our opponent’s deck if we’d played them before or if they just download a list of all the cards in the latest champion’s deck like every other freaking player. But the cards are face down and randomized. We don’t know what’s in our opponent’s hand. We don’t know if he has a useless Mr. Mime or a 10/10 Leviathan. And we don’t know if we’re going to pull another swamp on our next turn or whether we’ll finally get the Tuner Monster we need to Synchro Summon our Red Nova Dragon.

Randomness isn’t the only way of being surprised. Hidden information is also a great way to be surprised.

If you’re a player in a role-playing game, pretty much the entire game is about being caught off guard by hidden information. You have no idea what’s waiting around the next corner, you don’t know if the chest is trapped, you don’t know if your friend the cleric is going to heal you or if they are going waste that spell slot on Guiding Bolt again, you don’t know what the monsters are going to do, you don’t know how your next attack is going to work out, you don’t know where the treasure is, and you don’t know how many more encounters you have to survive today with the resources you have left. Everything in the game is a surprise.

By the way, this is NOT the case if you’re a GM. There’s only one bit of hidden information at the table for GMs. The GM does not know what the players are going to do. But that’s actually a very minor bit of information most of the time. The fact is that in most situations, the players’ actions are pretty predictable on the large-scale. And the small-scale choices don’t matter so much since the game is stacked so heavily in favor of the players. And THAT is why so many GMs have urges to add random elements to the game. To generate encounters from random tables, for example. Or even to build random dungeons. And that’s also why games in which the GM doesn’t get to roll any dice are bullshit. Because the GM needs to play too. And that’s also why Gary’s remark that the only reason GM roll dice is because they like the sound they make was a lot more insightful than anyone thought. Metaphorically. The GM needs dice more than the players because they give the GM more opportunities to be surprised. I just don’t think Gary understood that.

But that’s beside my point. My point is that RPGs already have a crap ton of hidden information for the players and the players are constantly being caught by surprise. So, why the hell do you need dice? What do dice do – what does randomness do – that hidden information can’t do.

Well, for that you have to understand some other things about human nature. Remember how I said there’s this complex interplay between the randomness of the universe and the human understanding of cause and effect? Well, believe it or not, it’s that second part that makes hidden information a big problem. See, we need to believe that things are ultimately fair. That, if we make the right choices, we’ll get the right outcomes. I mean, that’s the point of making choices. It’s about pursuing the best outcome.

The problem is, things don’t always work that way. Sometimes, we make the best choices – and I want to remind you that this is just about game design theory and has nothing to do with functioning in real life – sometimes, we make the best choices we can, and we still get sucky outcomes. And that’s down to random chance. That’s what we mean when we say, “life isn’t fair.” And we don’t like it. But we accept it. Because that’s the universe we live in. And, hell, everything in our brains that allows us to adapt to the unexpected is about surviving in a universe that isn’t fair. That is beyond our control.

But imagine if the outcome was controlled not by the universe, but by another human being. Imagine if you made the right choices, did everything wrong, and got a sucky result because some other person just decided you don’t get the right result. Like, let’s say you bet $50 on a Yugioh tournament and you won the tournament and your opponent refuses to pay up. And refuses to even admit you won. He just says, “that game didn’t count and I’m not paying you. Screw the bet, I have the money.” That’s unfair. And it makes us angry.

Now, imagine how you’d feel if every outcome of every action in, say, a fantasy role-playing adventure game, was determined by the GM. Imagine the GM just suddenly announced, “a goblin popped out of nowhere, rolled a critical hit, and killed you dead.” Pretty unfair, huh?

And THAT is why a role-playing game has dice. Because the dice are impartial. You can say a lot of things about the dice, but you can’t say they hold a grudge. You can’t say they play favorites. Well, some people DO say that. And all of these complex mental forces I’m talking about are precisely WHY people DO say those things. And why it’s even comforting sometimes to think the dice just hate you after you have a bad run of luck. And why it feels good to switch your dice. Because your brain only stays sane as long as you maintain the illusion that you actually have some control over your life. And that extends to simulations of life.

Anyway…

If you’re exploring a goblin cave and a goblin was hiding in ambush and you failed a perception check and then rolled a sucky initiative roll and then the goblin attacked and the goblin rolled a crit and then rolled for maximum damage and your character died, well, that’s just the way it goes. You might still be mad – and we’ll talk about that in a moment – but you can’t really accuse the GM of being unfair. Everyone saw the dice. Everyone knew the odds. Mostly.

And that’s why RPGs use dice. Because they are impartial, random, and fair. Everyone can see them. And, within the laws of probability, everyone knows how the work. The outcomes are random, but they are predictably random. They are random in a way that everyone at the table can see and understand.

And that’s also why dice are terrible.

Nobody Wants Fairness

It might seem like dice are the perfect solution to the problem of adding fair, impartial unpredictability to the game. And, for the most part, they actually are. They are a great solution. They are so good at what they do that I don’t understand why GMs insist on hiding as much information as they do. Because, if you read everything I said above very carefully – and you aren’t one of those wrong people who think I’m incorrect – you’ll understand why it’s actually a good thing for the players to know what their target’s AC is and what the DC they are rolling against is. Most of the time. Not every time. But most of the time. And GMs should be a lot less obsessed with hiding absolutely every damned number ever. But I don’t want to get into another rant here.

Back to the dice. Here’s the problem with the dice: they are utterly fair. They make no judgments, they have no ill will, they play no favorites, they have no biases. They just generate a random number which is fed into the game engine in a completely predictable and understandable way to determine the outcome. And that’s why people hate them. Because people actually hate fairness.

As I noted above, our brains have to cope with two opposing ideas. The first is that the universe is random and beyond our control and absolutely everything that happens around is so complex and so beyond our understanding that, despite the fact that we live in a deterministic universe, absolutely nothing outside of spherical chickens in a vacuum is remotely predictable. The second is that our brains are the only useful biological advantage we have and if can apply reason and logic and make the right choice in every situation, everything will work out fine. And because those ideas are utterly contradictory, we don’t like to think too much about that first idea, but we do have to be ready to cope with it. Which is why we hone ourselves with unpredictability.

Dice represent that first idea. The idea that everything is completely unpredictable. The best we can ever do is compute the odds and even if the odds are really strongly in our favor, all we have to do is roll a 1 and we’re screwed.

Look at that goblin thing. Imagine this: you made your first D&D character ever. It’s your first game. You’re excited. The game starts with a simple adventure into a goblin cave. You’re ambushed. You roll a ‘1’ on your initiative. The first roll of the game. Your first die roll ever as a D&D player. The GM announces that the goblin attacks you. The GM rolls a 20. Then rolls max damage on all the dice. And you’re killed dead. Game over. User wins.

How does THAT feel?

It sucks. Because there was literally nothing you could have done to prevent that. And the dice don’t effing care. The dice killed you. Because that’s what the dice do. Whereas a compassionate human would realize that’s a really sucky way to start off your first gaming experience ever and might fudge things, the dice just generate random numbers. And laugh. Because dice hate you.

That is the problem with dice. Dice serve as a reminder of the most terrifying truth humans have to learn to cope with. Again, game design, not life advice. And when they do something like that, well, it’s hard to pretend you’re in control. And it’s hard not to get angry.

The truth of the matter is we don’t want fairness, we want the world to be unfair in our favor. But that’s a very cynical thing to say. And I’ve been angry and bitter enough. So I’ll say this: we don’t want fairness, we want justice. We don’t want the rules to just apply equally to everyone, so everyone is equally screwed some of the time. We want the outcomes we think we deserve and no chance of being screwed if we didn’t do anything to deserve the screwjob. That’s so ingrained that people set up some very complex delusions to convince themselves they deserve things they didn’t earn and then scream that the world is unfair and unjust when they don’t get them. But, game design, not life advice.

And that’s why people get worked up sometimes about some of the weirder aspects of the game that arise from the random elements. For example, the idea that a fighter with a Strength of 18 who rolls a 1 can fail to break down a door but a wizard with a Strength of 8 who rolls a 20 can break down the same door, leaving the fighter to say “well, I loosened it for you.” That’s unfair. The fighter DESERVES to break down the door. He chose to focus all his points on being able to hit things hard instead of being smart enough to rewrite the fundamental laws of reality. At least he should be able to break down a damned door without the smartass wizard showing him up. And, obviously, the game is broken because it allows that outcome.

And maybe the game IS broken. But those situations don’t happen very often. And mature people can laugh them off and keep playing the game. So, they probably aren’t worth fixing. But they do illustrate the problem with dice.

Mitigating the Dice

Here’s what this all boils down to: we need unpredictability in games to add tension and excitement. Hidden information isn’t sufficient. We need to know the unpredictability isn’t going to be unfair or biased or unscrupulous. So, we use dice. But we hate the dice. Because they can’t be controlled. They are utterly, heartlessly, mercilessly, cruelly fair. They are random to a fault. No matter how much the odds are in our favor, things can still go wrong. And so, we – the players – do everything in our power to limit the influence of the dice. Or influence the odds. That’s called strategy. And that’s how players earn their victories.

Flank an opponent, get a bonus to attack. Use a crowbar to open that chest, get a bonus. Work together as a team, get a bonus. Use a magical item, get a bonus. Benefit from a blessing, get a bonus. Meanwhile, move to get a clear shot, avoid a penalty. Stand back and prod the chest with a ten-foot pole, hopefully stay out of reach of the trap. Gain temporary hit points, soak up some extra damage. Fight on the defensive, decrease the odds of getting hit. And on it goes. That’s the game. And the GM who complains about players doing that crap – throwing their strengths against obstacles, seeking bonuses, avoiding penalties, and avoiding danger – that GM is actually complaining that the players don’t want to live or die on coin tosses. Heads you win, tails you get to make a new character, no there’s nothing you can do to change the odds. Suck it up. Embrace failure. Failure is when the game gets fun.

What a load of crap.

Players love and hate the dice. They want to roll the dice because the dice are exciting and surprising and impartial. But they are forever fighting the power of the dice. They trust the dice implicitly. More than they trust the GM. Until the moment the dice turn against them. Then, they hate the dice. It’s that interplay again: randomness vs. determinism, cause, and effect vs. pure chance, law vs. chaos.

What Can You Do With This?

So, what’s the point? What can you – as a GM – do with this information? With this understanding? Why is it so important? Well, for now, the best thing you can do is keep it in your head. Because if you really want to be a master of advanced action adjudication, all of this crap about randomness is just another thing that will influence your decisions. When you’re deciding whether or not you need to use the dice to determine the outcome, you have to be aware of the dynamic that using dice creates at the table. And the potential perils. Because, yes, sometimes, even if there is a chance of success and a chance of failure and a risk or cost, sometimes, you don’t want to roll the dice. I know. It’s like there are no actual rules. Just making shit up as you go.

This – the effect of randomness on the game and the way players interact with the random element – is the third of three major factors to consider when adjudicating actions. When we add that into the rules we discussed long, long ago and the idea of action adjudication as a proposition bet between the players and the GM, we’ll find the real art of action adjudication.

But meanwhile, I will leave you with a couple of general bits of advice you can start using right now.

Just Because You Have to Roll the Dice Doesn’t Mean You Have to Roll the Dice

I know I gave you simple rules for determining when to roll the dice: roll only when there is (a) a chance the action could succeed, (b) a chance the action could fail, and (c) a risk or cost associated with the attempt. And I stand by those rules. Except when I don’t. Like when the best outcome of the die is an unjust outcome, don’t roll the dice. Just narrate the result.

For example, after the fighter fails to break down the door and the wizard steps up to give it a try, just tell the wizard he fails too because, you know what, the door is just too damned sturdy and the party is going to have to find another way through. Amazing, right? Such a simple solution. And yet it occurs to none of the hundreds of GMs and players online pissing and moaning about how ability check ruin games because they allow wizards to break down doors after fighters fail.

The Do-Over: When One Roll Isn’t Good Enough

One small, extra rule I spelled out years ago is that you should never require multiple die rolls to resolve an action unless there’s some kind of ticking time bomb or something which affects the choice to keep trying. Basically, I said, “one roll is usually good enough.”

But, when the stakes are really high – say, a character lives or dies on the roll – or the players have taken a lot of steps to mitigate the risk, one die roll may be enough to resolve the action, but it’s too few rolls to feel just. Notice that the stakes on most die rolls in D&D are actually pretty small. It’s rare for one die roll to kill you outright. Characters have to be whittled down. And, although I’ll argue to my dying day about the utility of properly used save-or-die effects, D&D 5E was right to dump the magical effects that can destroy a character based on one die roll. It takes three failed death saves to kill a character. And so on.

So, when the stakes are really high and the dice do something really cruel, like kill a character outright, it’s okay to find a way to offer a “mulligan.” I don’t mean just letting the player reroll. That’s stupid and the players will see right through that crap. It won’t feel good. It’ll feel like pity. Like charity. It’ll make them feel bad.

But, say, they miss the critical jump across the ravine. Offer a saving throw to catch themselves. Pretend like that was how you had it written all along. “You soar through the air, your toes strain to catch the foothold on the far side, and you miss. You come up just short. You start to fall and flail wildly at the ledge as you fail. Make a saving throw to catch yourself.”

Don’t overdo it. Keep yourself to one do-over. And don’t let the do-over mitigate everything. The players should still be in a terrible place. But, one extra die roll to soften the stakes when the stakes are super high or when the players have done everything in their power to mitigate the chance of failure, one extra roll is okay. And often, that’s all it takes. Succeed or fail, it tends to placate most players. Not all of them. But you can’t please everyone. And if you take away all of the risk and all of the surprise, well, it’s not a game anymore. It’s not worth playing or running.

Just remember: on the one hand, the dice are exciting, surprising, and impartial. They make the game fun. On the other hand, they are the embodiment of an existential terror that lurks in the darkest recesses of the mind of every human being in this universe. So, you know, maybe don’t overuse them.


Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

52 thoughts on “Dice: Beautiful and Terrible as the Dawn (Whatever THAT Means)

  1. “For example, after the fighter fails to break down the door and the wizard steps up to give it a try, just tell the wizard he fails too because, you know what, the door is just too damned sturdy and the party is going to have to find another way through”

    I apologise Angry. I don’t know whether it was me you quoted, but I was definitely involved in the conversation, and apologise for the irritation it caused you.

    On the flip side, the line above is exactly what I was looking for when I mentioned the fighter, wizard and door. I just didn’t know it till I saw it. Thanks for that, I guess I just wanted some sort of permission to rule things logically in place of the rules.

  2. Tangent: Chess is a terrible “game” specifically because there is no randomness. At high levels of play, it’s merely about how many decision trees your meat-based computer is able to remember. The only way to have fun with chess is to have two players who don’t know what perfect play looks like. That’s why we’ve been losing to our silicon-based superiors for over twenty years now.

    • “The only way to have fun with chess is to have two players who don’t know what perfect play looks like.”

      Isn’t this every game of chess? No person (and no computer) has calculated the perfect game of chess.

    • Poker involves randomness as well but one can reliably come out ahead over time – or even all the time. It’s a game about understanding opponents and predicting their behavior

      I think RPGs are great because their goal is ultimately to have fun – sometimes that’s by following the rules/rolls precisely and sometimes that’s by throwing the rules out the window. Remember, Calvinball is ALSO a game

      Truly random games are addictive but, I think, dissatisfying (for example, slot machines, craps, etc). They certainly fulfill the desires of some people but I personally can’t stand them. A computer will never reliably beat a human at a truly random game – because player skill has no bearing on outcome.

        • Fun isn’t about winning, it’s about a sense of accomplishment. I could create a game I can “win” every time, but there’d be no satisfaction in that.

          In RPGs, this corresponds to players who make deliberately sub-optimal characters. If the character is still played to the best of the player’s ability during encounters, that’s likely to be a lot of fun for that player. I personally dislike players who deliberately play their characters poorly because it’s “what the character would do”.

          “And the GM who complains about players … seeking bonuses, avoiding penalties, and avoiding danger – that GM is actually complaining that the players don’t want to live or die on coin tosses”

          Angry clearly agrees that gaming should be a balance of skill in decision making and chance. True randomness is boring, as are solvable games.

      • I cannot imagine having my head lodged so firmly in my ass that I would call a game that has endured for 1500 years and is played by more than 600 million people a year objectively terrible. I mean, not only would I have to assume I was smarter than that many people, I’d also have to misunderstand the subjective concept of fun really badly to think there was some sort objective pronouncement like that. I think if I ever made a claim that, I’d be embarrassed to ever show my face in a serious discussion about game design ever again.

        Likewise, I’d feel equally embarrassed if I ever got caught claiming that the 2 billion people who play random casino-type games must be addicted as that’s the only way anyone could find that sort of thing. And I’d never make that claim on a website run by a guy who has, himself, professed a thorough and genuine enjoyment of casino games, gambling, and especially craps and has zero problem walking away from said games when his purposefully set aside budget for such fun has been depleted.

        Here’s the thing kids: fun is not objective. And there are many factors that contribute to the things that people find fun. While there are some overall patterns – for example, a mix of unpredictability and strategy – the actual mixture is greater than and different from the sum of its parts. And thinking you know better than anyone else about what they find fun and why is no way to talk seriously about game design. There is no good that comes of sneering at things other people find fun because you, personally, can’t fathom that crap. Especially because your own opinions on what’s fun are PROVABLY irrational and all of your explanations are made up after the fact by a conscious brain that is constantly scrambling to explain your emotional responses.

        It’s okay for you to have an opinion of what’s fun FOR YOU or not. And to share it. But if you’re going to speak for anyone other than yourself with authority, you’d sure as hell better be sure you actually do know what you’re talking about. And that you’re being descriptive rather than prescriptive.

        • I am embarrassed that I defined fun incorrectly and that I implied addiction is the main reason to take part in games of chance.

          It only distracted from what I was trying to say – computers will never be better than people with a truly random game of chance. I enjoy RPGs because they don’t rely on strictly adherence to RAW to be fun.

      • Poker isn’t random (in the sense that dice are) – between them the players know what all the cards in play are. The skill of poker is in finding out what they know without letting them know what you know. Poker is t a game of chance it’s a game of empathy.

        • But who gets what cards and the order the cards end up in the hands of the players is random. That’s why shuffling is called randomizing. You can argue that the skill and strategy aspects of Poker vastly outweigh the random element and you can also state that the skill and strategy involves social skills, deception, and intuition, but to say it isn’t at all random just tells me you either don’t know what the word random is or you have never actually played poker.

          Either way, this entire thread starting with the statement of chess being terrible is now officially a dumpster fire. Consider this thread closed. Further comments in this thread will be deleted and replaced with insults.

    • Also, chess clocks. Expert chess players can handle a lot of those decision trees given the time to think them through, so we give them a time limit in competitive play to hopefully pressure them into making a few mistakes, to make the game more interesting.

  3. Interesting article. In the example you give about a fighter and a wizard breaking down a door, shouldn’t the fighter break it down wothout a role? If the wizard can easily attempt a second try, the fighter should be able to to. If thats the case there should not be a role at all, if i read your articles on action adjudication right. have a nice day.

    • Having been one of the annoying commenters in the original discussion, there are many situations where a roll to break a door may be relevant. If the players are captive behind a door for instance, or if they’re fleeing an enemy.

      • Then why doesn’t the Str 18 char, who has a greater chance of success, try again? Something has to have changed to make it so they cannot continue. And anyway, the Str 8 char is making a separate attempt in that case, with potentially different variables, because things that matter are happening as time passes

        The reason this gets brought up as an example is because folks get into a mindset that each char gets a a single one-and-done chance to try, either simultaneously or one after the other, and the results for the party will be the same no matter who succeeds. The other common version of this mindset is ‘Knowledge’ checks, which is far easier for more people to envision a scenario in which they believe each char should get one and only one check, individually, and that any success will be the same result for the entire party.

        • I suspect it crops up because a second player will often state their attempt to break the door down before the first player has fully recovered from their failed attempt.

          That is, saying “I try that too” takes less response time than saying “I try that again”.
          Also, there is probably some element of subconscious spotlight-sharing, mixed with a hesitance to simply repeat a failed action until they succeed.

          Essentially, it’s not necessarily that they can’t try again, it’s that a different player is more likely to try second, and by definition the second choice is likely to be the less logical choice.

        • The dice roll, especially out of initiative, should not be treated as a single effort, but as an abstraction of all efforts to do a thing. “I try to break down the door…I rolled a 6” “You fail” “I want to try again” “You already did try again and the final outcome of all your efforts was failure”
          In real life, when you are attempting a feat of strength, like removing the stuck lid of a jelly jar, you are initially met with resistance. Then you apply more force, and more force, sweat beading on your brow; up until the point at which you apply enough force to actually get the top off, you are failing, but we only count the final result. Were you able to finally get it off or not? You don’t actually fail until you give up.
          Asking for another roll is like saying “I have not yet given up”, but the GM should be able to say “no. you gave it your all, the dice have served their purpose in abstracting the sum of all your tries, and we are moving on”

          • If you’re playing 5e of 3e, that’s not the case. 5e explicitly has a DMG rule for trying again until you succeed, and 3e has take 20 to cover the same situation.

        • While I accept there are few likely times this exact scenario would occur with the limitations implied, the purpose was to illustrate the problem with an easy to understand example. The likelihood of this exact scenario is irrelevant, as long as either it or a similar situation are possible.

          Limitations to trying again could include a trap triggered by attempting to open the door, time pressure where one try is attempted before moving on, turn rounds during combat or limitless others I haven’t considered.

          It’s common for people to suggest the wizard help’s the fighter instead for better odds. However, this feels less immersive in a time pressured scenario.

    • I do this all the time in my game. If success is the obvious conclusion to their action, then they will succeed. HOWEVER, players (my players…most players?) really like the rolling of the dice. It’s not just the GMs who like the sound they make. So, as the GM, I already know I’m going to allow their proposed action, but if I also see the dice in their hand and that gleam in their eye, I tell them to roll anyway. If they happen to roll a one, it’s always fun to have them fail-successfully (fail forward).

      The same concept goes for critical information. I don’t know why so many GMs insist on hiding the most critical, key, important pieces of information their players need to advance the game behind DCs. If your game literally stops if they don’t gain xyz info, then the DC is 0. Useful ADDITIONAL information can hide behind the DC.

    • The fighter’s role is to be strong and do stuff like break down the door. The door’s role is to be either open, or in this case, shut. The wizard’s role is to cast spells and stuff. This is a discussion about role playing games after all, so it’s hard to imagine anything happening without a role.

  4. >”For example, after the fighter fails to break down the door and the wizard steps up to give it a try, just tell the wizard he fails too because, you know what, the door is just too damned sturdy and the party is going to have to find another way through.”

    I thought that was one of those examples where, once the first roll was failed, whether they break down the door or not no longer has stakes. The fighter failed to surprise whoever’s behind the door and, after that, just breaks it down without a roll. Would there really ever be a situation where, after the fighter fails, the GM thinks there is still a cost to failure, and allows the wizard to step up and roll? And in that situation (like the room is filling up with water or something), wouldn’t it still make more sense to give that roll to the fighter again because they’re the strongest (and have the biggest bonus)?

    • If a round is a finite period of time during which everyone (including adversaries) gets to do something, and also there’s time pressure, then it’s more optimal for every character to do something rather than nothing. Skipping a given character’s turn doesn’t allow the fighter to go faster relative to whatever is creating the time pressure (escapee, hunter, water, etc).

      Angry’s scenario gives the wizard a much more compelling reason to use knock or something similar.

  5. “The truth of the matter is we don’t want fairness, we want the world to be unfair in our favor.”

    Great line, and a great paragraph. Reminded me of Sid Meier’s speech regarding the Civ games, and the players’ belief that they should never lose a battle in which they have 2:1 strength ratio over their opponent, even though they still have a 33% chance of losing.

    I’ve always thought of dice, cards, etc. in games like D&D as encompassing all the myriad remaining factors not already covered in the rules somewhere. They don’t add randomness; they wrap up all the other potential factors into a tiny piece of plastic, to speed the game along.

  6. What I expected from Angry: Rambly explanations of core RPG elements.

    What I did not expect from Angry: Recommendations on probability textbooks. Is Angry by any chance familiar with Nassim Taleb? Come to think of it, I’ve never seen them in a room together.

  7. One thing I also find myself doing is that when something isn’t critical, a dice roll can be used to define the environment, also offering up alternative ways to address a problem. The strong barbarian rolls a horrible strength check, maybe it’s because the door is stuck from years of misuse and the intelligent wizard instead of making a follow up strength roll can instead roll investigation, possibly finding the source of the jam and allowing the door to swing open freely with an arrogant eyebrow raise to the barbarian still rubbing his aching shoulder…

    • That can be a good way to handle it, offering a thematic way of avoiding the situation where everyone in the party tries, one at a time, and the GM avoids multiple rolls and the dreaded “if you’re going to just keep trying until you open it, I might as well determine how long that’s likely to take and move on,” and at the same time avoiding having to explain why “only one of you gets to try, so decide before anyone rolls” is fair and not immersion-breaking.

  8. RE: Fighters And Doors; I have started telling my players “One player tries to do this, if the rest of you want to help, feel free. Tell me who tries and who helps before you roll” for a LOT of things – Doors, Diplomacy, Gathering Food. They do get to make their Aid roll, but no “Then the wizard rolled 20!” situations.

    • I’ve considered implementing a rule where multiple attempts at the same task (whether on multiple occasions or by multiple characters, such as group stealth) are represented by the same natural die roll.
      If the bonus has changed (either because a different character is being used or because circumstances have changed), then the attempt can succeed where it previously failed, however the natural die roll remains constant.

      So for stealth or perception, where everyone succeeds or fails as a group, the effect becomes very similar to Angry’s rule about group checks (one roll using the highest or lowest bonus), but firstly it can be applied to a wider variety of situations, and secondly it can be applied retroactively once a second attempt is made.

      For example, the Fighter tries to break down a door (DC 15) and rolls a natural 9 with a +3 Str for a total of 12, which fails.
      The Barbarian then gives it a try with their +5 Str for a total of 14, which still fails.
      The Barbarian rages to gain advantage, allowing them to roll a second die. Unfortunately the second die turns out to be a natural 3 (for a total of 8) and thus still fails.
      The Cleric then remembers they can cast Guidance to add 1d4 to the check, rolling a 3 for a total of 17, finally allowing the Barbarian to smash down the door.
      Success!

      I have not yet tested this system, but I do think it has potential.

      • (The following needs to be re-evaluated after reading this article, but here was my thought process in case anyone’s curious)

        The above rule could also apply to actions involving multiple skills.

        If a task could be accomplished using a mixture of Athletics or Acrobatics, but either one of them would be sufficient (for example by scaling a wall by Jackie Chan-ing your way up a corner), then you could make two ability checks, one using each skill, but both using the same d20 roll.
        Effectively the same result as saying “roll highest of Athl/Acro” (or “roll either”, because most sane people would choose their highest).

        On the other hand, if a task requires two skills, and the lack of either one could cause the task to fail (such as Perception to notice a clue and Investigation to realise its significance), then you again make two ability checks using the same roll.
        Effectively the same result as saying “roll lowest of Perc/Invs”, meaning you need to be good at both to have a good chance of success.

        Of course, all of this is just an idea, and it might just not be worth the trouble of explaining the process to confused players (most groups I know still ask for skill checks instead of ability checks, either because they don’t know any better or because players assume it works that way and it’s too much hassle to correct them).

  9. Holy moly, Angry! About midway into your article I wanted to pick up my Pop-O-Matic and smash it against the wall, but then I remembered I had already done that…years ago after a particularly dismal game of Trouble.

  10. Angry,

    Has my original GM written in to you because your example with the Goblin crit was scary. Got invited by a college roommate to try out D&D for the first time. I was given a pregen character that was trying to arrest one of the player characters. During our intro conversation we were ambushed to force us to work together. I turned around and got off one attack. Then, since I was facing the other guy and they attacked from behind, the three ambushers focused on me and killed me before I or anyone else in the party got another turn. Then I offended the GM by spilling the secret that I had backup by asking if they could save me. They couldn’t. So I ended up with nothing else to do that session and didn’t end up coming back.

    I did join my roommate next semester when he started with a new group.

  11. Why do ability and skill checks, saves and attack rolls use the die with the most sides, when they are the elements of the game that probably depend least on luck or chance? Why do I not roll a D4 to see if I hit the goblin, then a D20 to see if I sever an arm or scratch a bicep?
    In other words, why is the fighter’s muscle, training and willpower such a tiny factor compared to the randomness thrown in by the D20? Why am I rolling that die and not one of the littler ones?

    • Two things.
      1. In 3.5, you could easy getting an Attack bonus greater than a d20 could roll on its own. And AC? 40 isn’t unreasonable. This gives impossible challenges a place to be, but also creates a mechanical divide in content (the opposite is 5e, where context is a big part on what is difficult)
      2. Crits. While reliability is nice, when randomness strikes, DC and Bonuses work together to form the actual % chance of success. But crits? Crits would need a rework, and to many, they’re ingrained in the rolling process. Cranking it up to 20% “Your modifier is irrelevant” is too much. This isn’t a huge problem, but it wouldn’t feel very D&D to the layman without crits. Especially in the era of Seducing Dragons.

    • I’ve often considered some kind of system for varying the die size for actions involving more or less uncertainty, but I think you might need to overhaul the whole d20 system to make that work smoothly.

    • How would rolling a d4 be more interesting for the rolls you make most often? If any results are failure then the only granularity you have to adjust is increments of 25%. Fewer outcomes doesn’t make them less random.

      More sides on the die allows for more modifiers (and the decisions involved in applying those modifiers), as you can shift outcome chances by small amounts.

      If you want more predictability, then roll multiple dice; to a first-order approximation you can swap 1d20 with 3d6. The bell curve moves expected results towards the center, keeping the capacity for unpredictable swings but making the static modifiers more important.

      Alternatively, don’t do binary decision rolls. Damage rolls are a great model for incremental progress towards an end, in a way that can’t be perfectly simulated and skipped.

      • I think the implication is that the modifiers remain as they are.
        The simplest example would be on an ability contest, because you don’t have to worry about scaling the DCs:
        d10+5 vs d10+2 is much more likely to succeed than d20+5 vs d20+2.
        d20 rolls have a lot of swing and the contest could easily go either way, but d4+5 vs d4+2 is basically impossible to lose (unless you can lose on a tie for whatever reason).

    • The actual answer is Core Mechanic. A D20 is the core mechanic used in D&D / Pathfinder, meaning that it is used as widely and broadly as possible for consistency and simplicity. I believe there are other systems that use smaller dice to show less skill.

    • Clearly you need to go back and reread the article. The dice do not represent luck or chance. I wish there were a way to test people for reading and comprehending an article before they were allowed to comment.

  12. So concerning the Do-Over:

    Call of Cthulhu 7 has a rule I quite like: Every player can repeat any (non-attack) roll they like, but if they also fail that one, the result will be basically a criticall fail.

    It both helps with “unjust” die-rolls at the times where you tried and failed to play to your speciality and it adds tension to exactly the right scenes, as it’s usually used for important rolls.

    • Pushing Skills is one of my favorite things. It works really well in COC because of how a lot of rolls there tend to have stakes that can seriously influence the game, and there’s a certain fun to it. Additionally, the fact you get a ‘free’ reroll is offset that any failure is automatically a super failure.
      It’s like Wild Magic. Either something fun and useful happens, or the GM/Keeper gets to have a field day and everyone laughs (or shivers in fear)

  13. At the risk of being called a moron, this reminds me of a question I’ve had for a long time. Thus sayeth Angry: “If there’s no consequence for failure or penalty to try again, don’t roll dice.” Fair enough. But also: “Some skills, by their nature, can’t be tried again after a failure (particularly knowledge checks). Consequently, when there’s a chance of failure, you always have to roll these.”

    Now I’m reading: “Rolling encompasses a whole kaleidoscope of factors that go into success or failure, and it’s not just a measure of how much skill the player brings to bear on the problem.” To me this suggests that most checks include a small chance of permanent failure, such that you should have to roll for everything. Some of the multitude of factors will be transient, of the sort that you could fix by trying again, like whether the rogue fumbles the lockpick while trying to open the door. But some of them could just as easily be un-retry-able, like whether the rogue knows how to circumvent the security mechanisms in the lock (acknowledging that the DC generally describes the difficulty of knowing how to pick the lock, but also that there still could be multiple variations of locks within the same DC, just as there are multiple pieces of information that are equally difficult to know). But if there’s always a chance of permanent failure, you’d have to roll for everything and forget the big important rule about action adjudication.

    I would suggest a solution where, if a player fails several retries in a row, it means that one of these permanent failure modes is in play, and the GM prevents the player from trying again. The character gave it their best shot, but they didn’t have enough skill, or they managed to jam the door even further than it was originally stuck. Or even just that the character got frustrated and tired and can’t perform at their best anymore. But I know Angry disapproves of that sort of thing, and who am I to say I know better?

    • An interesting mechanic I noted in the nWoD rules is using the “Resolve” stat as your limit on retries until your character gives up in frustration.
      I’m not saying it’s a perfect solution, but it’s an elegant use of an existing stat. 🙂

  14. It occurs to me that this model actually offers an explanation for some of the behaviors angry considers problematic.
    Consider the phrase “we had so much fun, we didn’t touch the dice the whole night.” Much maligned by our host. Rephrase it in thos model, and you get the (inarticulate) “The revealed information and character decisions of this episode were so surprising, no one felt the need to increase the potential for surprise with dice.” The real mistake then is a confusion of cause and effect.
    Secondly, the gm who becomes annoyed and tries to thwart player attempts to limit chance. “Gms need surprise too.” The gms complaint, (again, inarticulate) is “My player’s decision making doesn’t surprise me enough to make up for how they have removed surprise from my experience of the dice.”
    Although I’m curious, why, in your opinion, are the die rolls of the players insufficiently surprising in games that are player-roll only? Is there some psychological element to rolling the dice yourself?

  15. I would add an additional factor when determining whether dice should be rolled: (d) if the consequence of failure is interesting.

    • YOU might. But that’s why I’m the one with the website. “Is it interesting” is a terrible criteria and completely non-actionable. Like “is it fun?”

      Besides, maybe I’m missing the point of being a GM – and I’m not – but I always thought it was the GM’s job to MAKE it interesting. Literally any outcome or consequence can be interesting. Or boring. You make it so. At least, I make it so. Don’t know what you do.

      • The players’ duty to entertain me is just as important as my duty to entertain them (arguably more important since we’re talking about me). Otherwise I’m just an unpaid actor. Besides, I have a limited amount of time and/or creativity in a session to make every failed Animal Handling check be interesting.

        I have never seen the failure to research a clue in a book be interesting. Robin Laws created an entire game system because that’s the case.

        • Bwahahahahahahaha! Holy crap, that’s comedy gold. “The player’s duty to entertain…” You are a funny guy.

          Also, it’s a shame you can’t make your game interesting. Maybe stop taking lessons from Laws. I mean, he’s such a sucky GM that he had to write around failure because he can only tell a story that goes exactly the way he already decided it has and has no room for failure. He should just write novels.

          • Laws’ Gumshoe system doesn’t automatically grant success in solving the mystery, only in finding the clue. You still have to decide to search for the clue (using a skill you’re trained in) and then you have to interpret the clue and solve the mystery. The player decision to search and the process of solving the puzzle is way more interesting than just having failure be decided by a 20-sided dice. You can still fail to solve the mystery if you don’t use the information properly.

            I mean who really cares if the character who spent points to train in photography can automatically use that skill to determine if a photograph has been altered?

            D&D players seem to have this weird stigma that every skill must have a chance of failure, but are also fine with the mysterious forces of magic just automatically working. The trained blacksmith can’t always forge a simple weapon reliably, but those complex wizarding gestures and precise magic words to produce that factory-standard fly or locate creature spell? Nah, those ALWAYS work, no die roll needed.

  16. I’m curious as to Angry’s take on fudging rolls – whether he rolls out in the open, or behind a screen. Allowing ‘one last saving throw’ , or fudging the dice are very similar in what they achieve, except in one the GM takes control and in the other he, well, still takes control, but leaves the actual result up to fate.

    Personally, I’ve used both but I giving one last saving throw is one of my favorites. I often tell players. I tell them they should be dead, but such and such a situation exists in-game that they could feasibly survive. Here’s the DC, if they don’t meet it they die/fail/whatever. Those are the moments of greatest triumph in my games, because everybody becomes SO invested.

    Also, side point:
    I recently watched video on YouTube about the history of war gaming by Invicta. I knew D&D was inspired by wargaming, but it was interesting in the video when the ideas of an umpire and the usage of dice were introduced.

  17. Others have mentioned granularity, but it may be helpful to put some numbers to it and see what falls out. In D&D 5E, your average 1st level character with standard array attributes has a +5 bonus to their best skills, while they have -1 to their worst. Now throw some sort of challenge requiring an attribute check at a group of average characters. If we’re throwing a D20, a fairly easy challenge gets a DC of 10. The non-proficient character needs to roll an 11, so they get a 50% chance to succeed – below the 66% or so chance Angry has noted is the general assumption in D&D and fairly poor for an easy challenge. The proficient character needs to roll a 5, giving them an 80% chance to succeed and clearing marking them out as proficient. But importantly, both characters can succeed or fail, their chance just changes depending on their stats and skills.

    Now look at it with a D4. What DC do we assign this easy skill check? The non-proficient character has a -1 modifier, so 3 is the absolute maximum they can possibly succeed at. The proficient character has a +5 modifier, so 7 is the absolute minimum they can possibly fail at. In other words, it’s literally impossible for an easy skill check to exist. Either it’s so easy it’s impossible for relatively competent character to ever fail, or it’s so difficult it’s impossible for anyone else to ever succeed.

    It should be fairly easy to see that the range of allowed modifiers has to correspond to the range of possible die rolls. If skilled and unskilled characters have modifiers differing by more than 4, rolling a D4 can never bridge the gap between them. If you instead use a D20, you can have modifiers differing by 10 or more that still allow everyone a chance at success while making it much easier for some characters than others.

    Finally, it’s not about depending on luck or chance. If you have to roll a 3 or more on a D4, that’s identical to needing to roll 11 or more on a D20. The size of the die does not affect the chance of success, it only sets a restriction on what it’s possible for that chance to be. A D4 can give you 75%, 100%, and nothing in between. A D20 can let you have 75, 80, 85, 90, 95 or 100% chance. If the DC of a challenge gives you a 75% chance to succeed, the die you roll is irrelevant. But choosing the correct size of die to roll can give challenges a much wider variety of available chances.

Leave a F$&%ing Comment (Limit: 2,500 Characters)

This site uses Akismet to reduce spam. Learn how your comment data is processed.