Sometimes, I have excellent intuition. I know exactly when I’m getting myself into trouble. Like that Christmas adventure article. I knew – I KNEW – it was going to make me miserable. That’s why I set it to hold all comments for moderation. And that’s why I put firm warnings in there. And that’s why comments are disabled on it now. And as much as I want to blame things like moral relativism, a general hatred of traditional values, anti-religious sentiments among intellectuals, and all that crap, I’m not going to bother speculating. Because the one thing I’ve discovered, the great paradox of gaming, is that gamers just hate fun. They hate people enjoying things. Especially things they don’t enjoy. But whatever. I knew there was going to be some issues with that article. And I was prepared for it. And, by the way, if you have opinions about the way I’m opening this Long, Rambling Introduction™ and want to disagree with this paragraph… don’t. I don’t care. See, I hate moderating, but I WILL do it. Especially when I feel like my fun is being ruined. If there is fun to be ruined, I will do the ruining. This is MY website.
ANYWAY…
My point is that I can usually see a response coming. I’ve been at this for a decade now. And I’ve learned to recognize the patterns. Which is why I actually do spend time pre-empting arguments. Because they are actually arguments I’ve had. I don’t work in hypotheticals nearly as often as you think I do. But sometimes, a response catches me totally by surprise. Like, I didn’t expect a simple – albeit long – speculation about looking at dice mechanics from a narrative rather than a mechanical standpoint would elicit such a response. That thing already has over 130 comments. And it has sparked several chats in my Discord, a few private message exchanges, two different e-mail exchanges, and one fistfight. Well, a tangentially related fistfight. It might also have started a Reddit fight. I don’t know. I haven’t checked to see if anyone posted a link on Reddit. And I’m banned from most of the RPG related subreddits anyway.
I’m not complaining though. Unlike people trying to literally ruin the one time of year that actually manages to make me happy, all of the discussions have been interesting, thoughtful, insightful. And very impassioned. Which is why I don’t want to harp on the crap about Christmas. Because most of my fans and readers are – through careful, brutal training and a bit of Stockholm Syndrome – actually smart, analytical, and interesting to talk to. And passionate. And they can take a punch. Which is important. Angry don’t truck with no pansies who can’t take a hit.
Whenever an article has that much feedback and sparks that many conversations – and controversy – I like to follow it up. Partly because the conversation and controversy help either strengthen my arguments or change the way I’m looking at things. Partly because I like yelling at people who misunderstand me. And partly because it’s easier than coming up with a whole different topic to write about. Once you’ve got a good milking cow, you keep squeezing until there’s nothing but dust coming out. I assume that’s how dairy farming works.
Thing is, though, that, while there are three topical teats, I want to squeeze dry, none of them really warrant an article by themselves. Even with the longest of Long, Rambling Introductions™, they’d be too short. Useful. But short. And while it’d be really easy to round out December with three, short crappy follow-ups to a crappy, BS piece of gaming dice gazing and then buzz off early for Christmas vacation, I’d actually feel kind of bad doing that. Because Santa is watching and I want him to bring me a Nintendo Switch. Besides, I have this thing about loot systems and crafting that I have to get out.
So, I’ve decided to take a page from one of my favorite episodes of House, M.D. and cram three short articles into one. So, here’s a quick blitz on The Nature of Randomness and Tension, Subjectivity and Fairness, and on How to Reveal Information to Players.
Angry Mini Feature: Dice, Tension, and Randomness
Sometimes, I have excellent intuition. I know what’s going to… just kidding. I’m not going to start each of these with its own Long, Rambling Introduction™. But I really did think about doing that. Just to screw with you all. Because I’m a mean-spirited SOB sometimes. But not this time. Let’s hit the ground running.
I’ve noticed that after I start using a word – like, say, “tension,” – suddenly, everyone is using and abusing the hell out of that word. And adding qualifiers. Since I posted that article, I’ve had people telling me that this thing or that thing increases tension or that more tension can be had by doing this or that. Some people have attached the word “dramatic” to the word “tension,” not realizing “dramatic tension” is an actual, specific narrative device that has nothing to do with dice. The problem is that people are equating uncertainty of outcome with tension. And, while they are related, all tension is not created equally.
So, for example, I’ve had people point out that confirming crits – because it “draws out the moment of highest tension” – increases tension. I’ve had people say that damage rolls do the same thing. Or tell me that resolving dice pools – moving dice around, counting symbols, and canceling out successes and failures – increases the tension. Well, yes, they do. Sort of. They make things more uncertain. And the make the moment of uncertainty last longer. The problem is that’s not a good thing. Tension is not a good thing. Not by itself.
Imagine, for example, that you’re wearing a collar that I can use to give you electric shocks. And, because I’m sadistic, I just hit the button at random sometimes. There’s no pattern. There’s no rhyme or reason. You just get painful electric shocks sometimes. In the middle of the night, while you’re eating, while you’re pooping, whatever. Sometimes, I give even just give the control to my cat to bat around as much as she wants. That is a highly tense, highly uncertain situation. And all it does is make you frantic. You’re constantly on edge. You’re wound up tight. And not just because of the pain. If the shock collar were replaced with a device that made you very “happy” in an adult way, you’d still be really tightly wound. I don’t think you’d want either collar. And even if you think you do, trust me, the novelty wears off quick and it gets very hard to live. I know.
Tension actually isn’t fun. Tension is anticipation. It’s a wind-up. The excitement comes in the release of tension. And I could continue the metaphor with “adult happy feelings” here, but you can all guess what I mean by that. You’re not actually after tension. You’re after specific patterns of tension. Moment to moment, tension should rise and fall. Over the course of a story, tension should trend upwards until the climax and then release. And it’s a fractal thing. If you graph the tension in a story, it’ll look like the graph of the tension in a scene, and in a single exchange. That was my point.
This is all deeply connected to psychology. It’s all connected to the way our brains analyze cause and effect and try to keep us alive. Which is also connected to the way humans formulate and tell stories.
Tension is not what you’re after. You’re after a specific pattern of rising and falling tension. Think of like shooting a bow. You pick your target, raise the bow, draw back the string, make sure your aim is good, and release. You can’t hold the bow drawn for too long. Otherwise, you get fatigued. And you can’t draw the bow, and then release it a little, then pull it back and check your aim again, then hold it, then release. I mean, you can, but that doesn’t come as naturally. It doesn’t feel right. And it’ll wear your arm down. You want a nice, smooth motion.
Now, let’s talk about randomness. Randomness is a terrible, awful thing that most people hate. Seriously. Randomness is a complete lack of control over the world around you. It’s horrible. It means you have no agency. Your choices don’t matter. For example, if a thunderstorm drops a tree through your car, it doesn’t matter what your plans for the day were. Because now your plans involve insurance claims and not going anywhere while you wait for the insurance company to approve a rental car.
Randomness is one of two ways that RPGs add uncertainty. Yes. One of the two ways. Remember that an action includes both an intent and an approach. I talked about this a long time ago. And if you want a refresher, you can check out my book . Coming very, very soon. It’s coming on a boat from China right now!
Anyway, the intent is what the player wants to accomplish. And it leads to the outcome. The player either accomplishes or doesn’t accomplish the thing. The approach is how the player wants to accomplish it. And it leads to consequences. The lasting repercussions of the choice. Basically, you can think of those as the short- and long-term effects of the choice. So, if a player wants to pick the lock on a door, the short-term outcome is that the door is open or it’s not open. The long-term effect is the players retain the element of surprise. Or that they can lock the door behind them. Breaking down the door with a battering ram leads to the same outcome. But the consequences are different. The party might attract attention. Or they might find themselves regretting that they can’t lock the door behind them later when they are being pursued.
The long-term consequences are determined by the GM and by the way the game plays out. Because the GM is omniscient – and also very good at improvisation – the GM can figure out the ripple effects from every choice. The GM knows whether there is anything around to hear the door being broken. And the consequences of the door being broken might only become important when the party is running from a horde of zombies later and could really use a barricade. That’s a sort of long-term uncertainty in the game. The unknown consequences of the players’ actions that can’t be predicted with perfect clarity by the players simply because they don’t know all the information they need. They can guess at some of the consequences. Consequences do follow logically. And that affects their choices. If the party wants to retain the element of surprise, they can decide to pick locks instead of breaking doors.
But the tension from consequences is a sort of slow-building dread-type-tension. There is also a short-term tension that rises and falls with every action and snowballs through the course of a scene until the scene is resolved. That’s the tension of outcome. The tension that comes from die rolling. And that’s the tension we’re talking about when it comes to randomness. It’s die-rolling tension.
Just like with long-term tension, short-term tension also comes from the players not being able to predict the outcome with perfect clarity. And to model that tension, we use die rolls. But, just like with long-term tension, the players do need to be able to predict the likely outcome. Otherwise, they aren’t making rational choices. By the way, this is why complaining that players are always trying to throw their strongest skills at a problem is stupid. That’s what they are supposed to do. The tension comes from the die roll. Of course, some situations increase the tension by not allowing the players to throw their best strengths at the problem. Or trade off long-term consequences for short-term certainty. Maybe the players can’t use their best social skill – Intimidation – against the king because he has an army with him. Or maybe the players can use Intimidation, but they also know the long-term consequences of pushing the king around might be very painful.
Not the point, though. The point is randomness is both necessary and terrible. It’s necessary because it adds short-term, moment-to-moment uncertainty and models the inability of people to predict the outcome of any action with perfect clarity even if they seem to have complete knowledge of the situation. But it’s terrible because true randomness robs players of their agency and makes their choices meaningless. That’s why players need ways to affect the outcome. And, to be honest, that’s one of the biggest weaknesses in non-combat action resolution in D&D. The only choice you can really make in D&D to affect the outcome of a non-combat action check is to pick which skill or ability to try to throw at the problem. And that turns the situation more into a puzzle than a choice. It’s not “how should I react to this guy,” as much as it’s “how can I use my physical strength to get this guy to do what I want because I’m stronger than I am personable?” But that’s also not the point.
What is the point? Well, there’s a few. First, tension and randomness are related. But they aren’t synonyms. There’s a direct relationship. More randomness equals more tension. But, more tension is not automatically a good thing because it’s the rise and fall of tension that’s important. And, more randomness also decreases agency and the meaningfulness of the players’ choices. And you can’t rely simply on the long-term consequences of those choices to give the narrative a proper tension curve.
Am I making sense?
Now, I am not going to argue that one die-rolling system is objectively better than another because this is all a complex interplay of forces. And I am not even going to say that the moment-to-moment tension curve that I was discussing – decide, deliberate, resolve – is important enough to override all other concerns about die-rolling. But I am going to say that, if your only argument for a die-rolling system is that “it adds more tension via uncertainty” or “stretches out the moment of uncertainty,” you are equating things that aren’t the same.
Angry Mini Feature: Stop Objectifying Your Brain
I have had many discussions since I posted that article about dice systems and tension. And this recurring theme kept… uh… recurring. It hit its absolute peak in an argument about revealing information to your players. And that’s what I’ll be talking about in the third Mini Feature below. But, we have to get this out of the way first.
I have had more GMs than I care to count outright tell me that they want OBJECTIVE RULES AND UNIVERSAL STANDARDS. For example, I watched an argument swirl for HOURS about what information to reveal to your players and when it should be revealed that was based around establishing a single, unifying rule about when to reveal what information. And then I stormed in and yelled at everyone. And then I told them the correct answer. It’s below. But the correct answer was predicated on the idea that there isn’t a single, unifying rule or objective standard you can apply. And I’m not saying any of that bullshit about how there’s no “one right way” and “every table is different” and “every GM should decide for themselves.” I mean quite literally that there is not one unifying rule or objective standard that a single GM can apply to all situations for anything in the game. If there were, there wouldn’t be GMs. And if you seek such an absolute standard, you’re wrecking your ability to run good games and missing the point of table-top RPGs.
The GM – the referee as he was once called – is a game mechanic. He’s part of the game rules. He’s an action resolution tool. He’s the first action resolution tool. And he’s the tool that provides the open-endedness that makes RPGs RPGs. It’s his ability to make judgment calls that make the game actually work. The rulebooks provide all sorts of objective standards and unifying rules and all that crap. They are the laws of the land. But just like you have judges who weigh the merits of individual cases, extenuating circumstances, resolve conflicts and contradictions in the laws, and figure out ways to apply the laws to situations that weren’t imagined by the lawmakers, you have GMs to do the same with the game’s rules.
Now, you can argue that all laws must apply equally to all people – and that’s true – but that doesn’t mean all laws apply to all situations as equally as all other laws. We have a law against murder, but we also have exceptions for self-defense, and we have principles for the minimum necessary force. All those laws and principles and exceptions have to be weighed in each specific case. Am I comparing GMing to being a judge in a life-or-death self-defense case? Yes. Absolutely. They are entirely the same thing.
Well, they may have slightly different moral weights.
Not the point.
GMs have this problem of seeing fairness in anything that isn’t a universally applied standard. That is, if you can’t spell out the exact rule that applies equally and absolutely to all cases, it can’t possibly be fair. And there’s two problems with that. First, fairness is somewhat overrated. Nothing is completely fair. But, second, it’s actually judgment calls that get you closer to fairness. As long as you consider all forms of fairness.
For example, in an RPG, there are three basic principles that need to be at the forefront of the GM’s mind: consistency, agency, and engagement. The players need agency to feel like their choices, they need consistency to predict the likely outcomes of their choices, and they need engagement to actually care about the consequences of their choices. Absolute fairness speaks to consistency, but that’s only one of the three pillars of the game. Imagine, for example, you have this absolute standard for delivering information that ruins all of your flavor text. Makes it boring. Well, the players won’t pay attention, they’ll miss important information, and their agency will suffer. And they’ll be bored. Which kind of hurts engagement.
I’ll give you an even more solid example from my own game. Character speed and counting squares. Sometimes, to speed movement on the battle grid, I fudge the movement. I let players squeeze an extra five foot of movement out because they are trying desperately to avoid an attack of opportunity and it’d bog the game down to let them keep counting out different paths, but it’d also be a bit of a screwjob to tell them they can’t take the time to figure out the best route when one might exist. So, I just say “okay, you get there, now what?”
Is that consistent? Is it an objective standard? Hell no. A PC’s Speed on their character sheet – and a monster’s – is not actually an accurate measure of how far they can move in all situations. It’s close enough. I mean, it’s not like I’d ever let a player cover two moves worth of distance with that sort of fudging. But it does introduce inconsistency. And it also gives the players a little more freedom to act with impunity in corner cases and prevents the most exciting parts of my game from getting bogged down and turning boring.
Is it fair? Absolutely it is. It ensures everyone is having a good time and it rarely matters enough to affect the outcome of the game in a serious way. And when it DOES matter enough to affect the outcome of the game in a serious way, you know what I do? I. COUNT. SQUARES. Yes, the minute the consistency outweighs the agency and the engagement, in my judgment, is the minute I change the rule.
I do not trust any GM who says “always” or “never.” Because a GM who “always” follows a certain rule or “never” allows a certain thing is on autopilot. They are not engaging their brain. They are just executing a program. And that means they are running a video game. Now, video games are fun, but they aren’t RPGs. And I don’t want a GM to run a video game for me if I’m sitting for an RPG. I want a GM who is constantly balancing the priorities to run a game that provides the best balance of consistency, agency, and engagement.
Well, that’s hypothetical. Because I don’t want any other GMs running games for me. I’m too clever and creative and talented and charismatic and handsome to be a player.
The point is, though, that objective standards and universal rules that “always” or “never” apply, move you AWAY from real fairness, not toward it. Once you consider all sides.
Angry Mini Feature: When to Let Your Players in on the Secrets
On the Angry Discord Channel – which my high-tier Patreon supporters get to hang out in – on the Angry Discord Channel, a big discussion exploded out of a very simple question. And man, did the discussion ever get contentious. Eventually, I had to step in and yell at everyone for being a bunch of… I can’t use the words I want to use because these people support the site and I am contractually obligated to not hate them at the very least. The discussion grew out of my very firm statement in that last article that I am now of the very strong opinion that hiding numbers like DC and AC from your players is bad and wrong and you should feel bad if you do it. I stand by that.
The question was: “okay, if Angry is right, when do you reveal the numerical information.” And then arguments went back and forth about incorporating numbers into the flavor text and using narrative clues and overwhelming the players with mechanics that ruin the flow of the game and all sorts of crap.
Meanwhile, I was also fielding my own little mini arguments from that statement because, while most people did admit I had some kind of a point, a few wanted to point out that the times when they didn’t reveal such information. Like, it would be foolish to reveal the DC needed to find a Secret Door that the party didn’t know was hidden when they made a roll. Or the DC needed to hide from a foe when their success or failure wouldn’t be evident until sometime later when the foe reacted to their presence. Stuff like that.
All of that is why it was necessary for me to point out that the idea of objective standards and absolute rules that you must “always follow” or “never violate” are anathema to the entire fricking point of role-playing games. And that’s sad. Because that shouldn’t have to be said.
First of all: obviously, you do not reveal all of the mechanics ever. Sometimes, you keep the mechanics hidden because the mechanics would reveal information to the players that their characters would have no way of gleaning about the world and would drastically change their behavior. Note, though, that that statement is very carefully worded to be vague. Because I know some mouth-breathers are going to jump in and say “characters have no way of knowing the exact AC of a foe and therefore, I violate my own rules if I reveal the AC.” But those people are stupid. Because the armor class of the VAST majority of foes comes from things that any individual trained in combat who has been through more than one fight would be able to assess. First, there’s equipment: armor, shields, weapons – because some weapons are better for defense than others – and so on. Second, there’s visible signs of training: like reflexes, build, carriage, speed, stamina, and reaction time. Any martial artist or fencer or boxer can tell you that those things are telegraphed to some extent. In fact, part of any combat training is learning to assess your opponent. So, if you have a proficiency bonus that applies to any weapon, you’ve learned that crap. Third, there’s physical traits of the enemy: thick hide, carapace, shell, being made out of solid rock. Fourth, there’s knowledge about the things that live in the world through hearsay, stories, pop-culture, and that kind of thing. Everyone knows dragons have impenetrable hides. Even I know that. And I have never fought a dragon. Nor do dragons even exist in this world. And most Greeks knew all sorts of things about sphinxes and manticores and chimeras because they had seen plays and hear stories. The one X-Factor someone MIGHT argue is supernatural stuff. Magical effects. That might not be visible. But it also might. There’s nothing to suggest a mage armor spell is visible OR invisible. So, you have to decide which it is, Mr. or Mrs. GM. And if you decide it’s invisible just so you can hide some numbers, you’re being a dick. Or dickette.
No, none of that crap is quantitative. The quantitative measure – AC – translates all of that information into something that we players who can’t see and hear the world of the PCs and don’t know what the characters know can talk about. That’s literally why the numbers exist.
And that’s why, as far as I am concerned, the argument that revealing numbers like ACs and DCs is metagaming because the characters wouldn’t know that stuff exactly, that argument is done. It’s over. It’s wrong. And I am no longer, after today, gratifying it with a response beyond “you’re a moron, shut up.” Seriously. That argument is over. I’m right. You’re wrong.
But there are times when the information contained in the DC actually reveals stuff that the characters would have no idea about. And which would make the characters behave substantially differently if they did know. For example, if someone decides to ransack a room in the hopes of finding some treasure or a secret door that may or may not be there, and you tell them “okay, you need a 19 to find the hidden gem and a 25 to find the secret door,” you’ve told them there’s stuff to find. The character was searching on the off chance something might be hidden. You’ve confirmed their suspicion simply by revealing those DCs. And if they fail the roll, any reasonable player would want to find ways to try again.
That’s what I mean about information which the characters could not glean in any way, not even qualitatively, that would drastically change their behavior. Yes. By all means. Hide that information. Hell, make the roll in secret.
But beyond that, there’s no good reason to hide numbers like AC. The characters should be able to pin down the AC of almost any opponent to a reasonably small range by using their senses. So, niggling over the exact number is just dumb. And no, it’s not “more fun” for the players to gradually figure out the number they need to hit. You can claim that because that’s what your players do, but until you’ve also tried it the other way and seen how both play out, you don’t know what’s “more fun.” You only know what your players do in response to the way you run your game. You don’t even know their motives.
If you disagree, you’re a moron. Shut up. Argument over.
So, now we have this Discord argument: WHEN do you reveal the information? The argument started because one of the people in my Discord chat was concerned that peppering all sorts of numbers throughout the flavor text would kind of wreck the flavor text. And players wouldn’t really hear the numbers. They wouldn’t remember them anyway. That’s true.
At the same time, though, if there is information you are willing to give your players that would affect their decisions – like the AC of the foe they are fighting – and you withhold that information for any length of time after it should be obvious, you’re screwing with your players. That’s also true.
Think about. If the players are trying to decide how to open the fight, and the cleric goes first and is trying to decide between buffing the fighter with a bonus to hit or a bonus to damage or a bonus to defense, those numbers would affect the decision. Holding them back until later means hearing the cleric saying, “oh, man, if I’d have known his AC was so high, I would have cast True Strike instead of Shield of Faith.”
And remember, we are no longer ever again arguing about whether the cleric gets to know the numbers at all and whether making decisions based on those numbers is bad playing. Because the argument is over. Shut up. Go waste Mercer’s time with that bullshit. Or Colville’s. Or Crawford’s.
So, what is the one rule? Do you ruin all of your flavor text forever or do you risk withholding vital information and screwing with the players’ decisions? What’s fair? What is the best universal axiom that always applies?
DO YOU SEE WHAT I’M SAYING?
Look, most of the statistics in D&D fall into a pretty small range of normal. And most players have a sense of what that normal is just from playing the game. A DC of 10 or 15 is normal. An AC of 12 or 14 or even 16 is normal. At least at low to middle levels. So, you’re not blindsiding a player by not revealing the exact number. A fighter who would happily attack a foe with an AC of 14 isn’t likely to quail at throwing himself at an AC 16. The exact number actually won’t change the players’ decision very much. If at all.
So, most of the time, you reveal the number the moment you actually need it. Like when you say, “make an attack roll; the foe has AC 18.” Or when you say, “make a Reflex save to hit the deck and avoid the fire breath; the DC is 15.” Or when you say, “Athletics check, DC 15, to climb that wall. Make the roll.” And that’s all I was really saying in that last article. When it comes time to resolve the mechanics, lay out all of the mechanics. There’s no good reason to keep them secret once the dice are flying. Just tell the players what they need to roll when they are rolling the dice.
And, if the player says, “whoa, wait, that DC check is way higher than I expected for this Climb check. Can I rethink that?” Let them rethink that. Don’t be all, “well, your character doesn’t know the exact DC so you’re committed; roll the check to not fall and die.” Just assume the character reassessed as they were getting ready for their climb and realized they were underestimating the difficulty of the task. You know, like a trained climber MIGHT ACTUAL DO.
And remember that all of the tension and uncertainty in an action comes from the die roll. Hidden information – like secret mechanics – is just a bunch of fake difficulty that pushes tension into frustration. And if you’re relying on that, maybe you want to win against your players just a little too hard. Yeah, I’M saying that about YOU. Think about what that means.
And if you think a number is well outside the normal range, be it AC or a DC or a damage rating or a save DC, telegraph that shit. Let the players know. Because they have a right to change their actions. Here’s the thing: the players getting surprised by something and losing a lot of HP? That’s not interesting. What choices the do players make to deal with something unexpected and out of the ordinary? That’s interesting. That’s role-playing.
P.S.: Monster HP totals don’t count. Never reveal those to the players. I don’t. And I shouldn’t have to even explain that.
>P.S.: Monster HP totals don’t count. Never reveal those to the players. I don’t. And I shouldn’t have to even explain that.
May I get an explanation anyway?
I’m going to take a stab at it:
1) HP are enough of an abstraction that it’s hard to tell in-game what someone’s HP total is (vs AC, where someone trained to fight should be able to generally estimate how hard it’s going to be to hit someone).
2) Knowing a foe’s HP at the beginning makes some choices too easy. Sure, a fighter’s going to go hit it with a stick until it stops hitting back, but a caster has to choose whether to break out the big guns (ie., the high-level spells), and a healer needs to decide whether to heal an ally or attack the foe. If the players know how many HP the enemy combatant has, those kinds of decisions become too easy: the trade-off between “kill it fast” and “conserve resources” is too easy to weigh.
Exactly. If the players know the numerical HP of an enemy, they switch their tactic from “cleric heals and buffs, wizard debuffs, fighter smacks” to “everyone hits that damn thing”. The worst is how the players will engage in full mathematics in order to defeat it within one turn. Its nothing but a box of wrenches in the gears of the game.
I think however, and other comments seems to point in that direction, that it doesn’t mean to give now information at all about how close to death is the monster. If an Ogre is at one-digit HP to be dead, a wizard should know she doesn’t have to burn a 3rd level spell slot to kill it: it would pretty much looks like a screwjob to me.
The idea, I think, isn’t to give exact numerical HP value of monsters, but more a rough idea of death progression: perfect health, slightly injured, seriously wounded, almost dying… However, the HP system makes this quite tricky, considering how abstract it is and the fact mechanically, a 1-HP Ancient Dragon is the same as a 500 HP one (I don’t say it’s a bad thing, only that this mechanic doesn’t have a way to handle the information to the players for you).
“How many HP does it have left?”
“How many does it have now?”
“How close are we to killing that thing?”
“Is it dead yet?”
“Are we even still having fun?”
EVERY.SINGLE.ROUND.
I revealed when a monster was bloodied (1/2 HP) by an attack when that was relevant in 4E. Players could have calculated exactly how many HP they needed to finish it off, but never really did.
One of the main things I’d say is because the actual number of HP is meaningless, from a narrative sense. AC abstracts what a character can know – that enemy is heavily/lightly armored, or slow/quick moving – into a standard range of values.
HP on the other hand is a massively abstracted number that has too much range to be narratively functional. An enemy has 9 hit points remaining. For a town guard or bandit, that means they’re fresh and rested. For a storm giant, it means they’re battered and bloody, labored in breath and inches from unconsciousness.
The other one of course (and I’m risking Angry’s wrath) is the metagame aspect – oh that thing looks strong and rested but it only has 30 hit points? Cool, no need to worry about powerful spells and damage features then, it’s like CR 1/4 and we can ignore any bluster it has about being more powerful than that. Knowing hit points tells you a lot about a creature’s actual power and fundamentally changes how you would approach it, based on something that you really can’t know in any reasonable way.
I began announcing “bloodied” (half HP) state in 4E as well, and have adopted it as a standard practice in my games independent of system in addition to visibly showing cumulative damage because I have found it helps my players make more informed choices in combat. “Oh crap, this thing’s taken X damage and isn’t even bloodied yet? We might need to rest/recover before we take this on” or “We’ve done almost twice its bloodied hp, I think I can finish it off this round and then it will be safe to stabilize you”. It also gives the players a visible sense of making progress on high-hp boss fights, similar to health bars in RPGs where you don’t know the final number, but you can see the how relatively damaged it is by color/fill.
I’d say it’s because some spells succeed or fail based on the HP of the target, so you’re taking away the inherent risk associated with the spell by telling the player whether they will succeed or fail before they take the action.
I’m AFB but I think there’s a death ray type spell, maybe finger of death or PW: kill; but also just sleep (the spell), for example.
Because there’s only two things to know about somethings HP. Whether it’s alive or dead.
One of the easiest ways to adjust the difficulty of combat on the fly is to change hitpoints. If the characters are having no trouble, add more; if the characters are dropping left and right, make the next hit a killing blow.
It is impossible to do this if you tell the players how many hit points the enemies have.
The answer is tension.
In video games, seeing a healthbar increases tension. Most videogame bosses can still kill you very quickly if you forget to dodge. You restart the fight when you die so dieing only costs you the progress you’ve made in that fight. The later in the fight the more you stand to lose. Tension builds over the fight and is released when it ends. Angry discusses the optimal tension graph in this article and this is the perfect tension graph.
In DnD it’s the opposite. A lot of combat is resolved via random chance. It’s very hard for a dnd boss to win while at 10% hp when the players are all still full health. Tension is high at the start and gradually decreases as the battle inexorable goes the way of the PCs. Exactly the wrong shape of a tension graph.
Also holy 6 years late batman. (i just discovered this website and have been bingeing the content on it)
Great article Angry. I feel bad for asking this since you said you shouldn’t even have to explain it, but why shouldn’t I reveal monster HP? In the same way that a warrior can access the AC of a plate-armour Knight with a shield, wouldn’t they be able to tell that a bloodied owl bear is far from 100%? I’m sorry, I really tried to think this through but I can’t see the difference between AC and DC in this circumstance.
AC and HP, stupid typo.
People keep talking about fudging hit points, but I never liked the implication of cheating in the word. You don’t cheat; you’re the fucking GM!
The reality is that hit points (which I call hp, not HP, because I write a lot and I’m lazy) are a tuning knob, a way for the GM to adjust a fight so that it is as satisfying as possible. Most of the time, it’s probably fine to use the hp as they are written down. Other times, the party is unexpectedly lucky, or a fight that you thought would be fun risks turning into a slog, and hp are a simple knob to adjust fights so they make sense and move the narrative forward.
NEVER let players know hp.
I think this might lead back to “making optimal decisions”. Decisions that have an optimal answer are math tests and not meaningful.
Attacks are varied enough that choosing one defense against another is inherently different in the outcome. Your sleep spell works differently to your fireball, so you don’t cast one over the other *just* because you are more likely to hit.
On the other hand, there’s so many ways to deal damage, that you might be inclined to calculate the optimal descision based on the options you have.
I know that Colville argues that HP is also the easiest thing to fudge, as it is less player-facing than AC, where you’d notice if an attack roll suddenly does hit, when it was a miss a few rounds ago. Although I’m not sure of Angry’s view on that topic, so this might not be it at all.
I’d like to throw my hat in as someone else who doesn’t get the difference between giving HP and AC? Maybe we’re all dumber than Angry gives us credit for?
The only difference I see is that you might decide to fudge HP for a more dramatic moment, or because you dramatically miscalculated the difficulty of an encounter?
Similarly, I’m curious whether Angry would advise telling the players which dice are rolled for monster/trap damage? I see the logic that players should be able to assess the deadliness of an enemy, but they should also be able to tell roughly how many hits it should take to bring something down?
you can give AC since its a measure of how hard the guy its to hit. As an adventurer you should be able to asses this stuff after a first hit, just the same way a martial artist can asses a rival’s skill.
More importantly, AC can be defined easily as a “the difficulty to hit or harm an individual”
HP is more loosely defined. HP can represent as much wounds as spirit, moral and vitality. HP 0 not necessarily means a dead guy or a wounded guy, it can easily meant a demoralized guy who has lost his will to fight.But again, it can meant wounds or a dead, unconscious or almost dead guy.
Giving a number in AC is just saying “you can hit it if you have the neccesary skill”. It takes no mistery, it doesnt give you any meaningful info, it only gives you the odds. You cant really say if you will win or lose based upon the AC.
But Giving the number of HP is saying exactly “This guy will die in X number of turns if you succeed in X hits of X damage each” and when you know how many turns you need to win, that takes something of the game, call it mistery, call it tension, call it whatever. You are giving more than the odds in the end.
However, you can give reasonable info without giving the HP number as such. Personally, when my players ask how much HP the enemy has left i simply say “it looks barely wounded (when HP its over half, it looks injured (when HP is at half), it looks badly wounded and “it looks nearly dead”.
Mind you, more than once a “nearly dead” enemy has turned the tables on its favor.
Not just those two reasons, there are lots of reasons to fudge HP. For example if the players knock an enemy down to single digits after a ten round fight, when all the significant resources that are going to get spent have already been spent, you might decide to just let it count as enough damage to get things over with and move on to the next scene. Or you might pull an “ambiguous death” against an enemy that wasn’t quite as dead as you thought it was.
But another consideration is that tension/release thing. Not knowing the exact HP means never knowing exactly WHICH hit is going to be the one that kills the enemy. This doesn’t matter as much against mooks, but against drawn-out battles or boss fights or the like, especially ones where the enemy is still dangerous even at 1 HP (ones where you WOULDN’T fudge this stuff like I said in the first paragraph), the players are going to be on the edge of their seats to figure out “Is this enough? Will this end it?”
If you’ve ever watched Critical Role, think of how crazy the reaction is when Mercer drops a “How do you want to do this” sometimes. That reaction comes from the release of the tension of not knowing exactly how much more punishment the enemy can take. And the enjoyment the players get from that final reveal that they’ve finally managed to deal the death blow justifies any irritation from not knowing exactly how much HP they’ve got.
I do personally give them general sensory indicators of an enemy’s health though. “Bloodied” as several people have mentioned, but also things like “it’s taken some bruises but it doesn’t look too worried about them,” or “it’s struggling for breath, but it’s not giving up the ghost yet,” or the like.
For the first section regarding confirming criticals I think the split comes from those who were raised with auto crits and those who were raised with confirming crits. For those who were raised with confirming crits it’s waiting to see if the game rewards you. For those of us who were raised with auto crits the moment of tension is a moment of tension waiting to see if the game screws you over and takes back what you rightfully won already.
I actually fought for confirmed crits for quite a few years (before third edition came out) since I found it odd that for unskilled opponents literally every time that they hit the near untouchable opponent they did double damage. It didn’t go over well but it wasn’t until I actually sat down and went through the progression of someones thought process on rolling a 20 that I saw the error in my thinking.
To counter hiding HP, it is a big element that can be fudged. I’ve run games with a visible health bar on a screen. It mimicked Dark Souls where a health bar that barely moves is really scary and its too hard to count percentages.
Here is an open HP system that is info that a player could guess.
Give a creature tokens equal to its number of hit die. On first hit, remove a token and roll its hit die (d6, d8 etc…), subtract damage from the die value and if it goes over remove a token and roll again.
If that’s too slow, just track damage publicly and reveal its number of hit die. So they know its upper range but don’t know when it will die.
I had argued for revealing AC and HP. In fact, I have the numbers out in the open.
Sure, the players are able to make more informed decisions, but I don’t think they are making substantially different decisions. They still:
1. Attack one foe at a time until they are down and/or
2. Take actions that optimize action economy for the team and/or
3. Play to win while economizing expendable resources.
I’d like to know why Angry won’t/hasn’t revealed HP at any point in a fight. I can see arguments for waiting to reveal HP after the party has hit it once, but never? I want to know when my sleep spell can be the finishing move on a high AC monster. If I can’t know, I waste an action and a low level spell slot or I burn an unnecessarily high spell slot to get the last 10 hp.
“P.S.: Monster HP totals don’t count. Never reveal those to the players. I don’t. And I shouldn’t have to even explain that.”
“My point is that I can usually see a response coming. I’ve been at this for a decade now. And I’ve learned to recognize the patterns.”
“I do not trust any GM who says “always” or “never.””
I see what you did there. Shame that (as of this comment) literally nobody else has. Which is weird, because that postscript ran so counter to the spirit of the rest of the article that I immediately assumed it wasn’t on the level.
Yes, but remember I am also never wrong and I would never intentionally post bad advice. And I’ve been contradicting myself across articles for years. So what now?
Well, that’s the point right? You’re not a machine and GM’s shouldn’t try to be.
“Never use a unifying” rule is a sort of unifying rule. Therefore, there must be exceptions to it.
On some specific cases we will end up with an answer that pretty much equals “never do this”. Revealing precise HP totals is one of such cases because it has the power to turn combat into an math puzzle.
The thing is, I had the idea “oh it’s there because he expects people to remark on it” in the back of my head from the start but… like, I don’t see a reason why Angry would _want_ this reaction. It’s not a pleasant reaction when many people ask the same question because of one throwaway line. It pollutes the comment section.
He probably has a purpose for it if this is the case but it’s so weird to me I made a short post asking anyway.
We also have a very different idea of what “pollutes” a comment section.
Well it doesn’t if you wanted many people asking the question and many people trying to answer the question. I would put an :eyes: emoji here but this comment system seems to delete them.
It’s an “I don’t understand why you want half your comments to be the same question and the other half trying to answer that question when you probably already know exactly know which answers it would produce” type of situation.
I am MOST LIKELY I GUESS BASED ON YOUR OWN RESPONSE AND ALL THAT just VASTLY OVERESTIMATING the wealth of answers to this debate/question you must have already seen in your life. If Erniemist is right and you actually wanted to stimulate debate on this particular topic that is…
But if that’s the case, why not just ask people to discuss? Seems more in good faith to me in a way that I can’t even properly articulate.
On a sidenote,I think I understand why some people actually got mad the Angry riddle now. I feel way more insulted than I rationally should be under the circumstances (like “Not at all” would be a reasonable level of insulted-ness). That is such a weird observation to make about myself.
PS: Is there a way to bold or underline or italicize or strike through words here? I don’t like resorting to all caps for emphasis.
At this point people know that angry is usually right, and blatantly disagreeing with him in his own comments section without a watertight argument won’t get you anywhere. Instead people are discussing why you might conceal hp and what’s different about it. This leads to thoughtful debate about the underlying principles and reasoning of hiding values as opposed to blindly following Angry’s decrees, which is the whole point.
I’m not seeing the don’t tell HP. I’ve tried multiple methods and more or less settled on saying 1/2, 1/4, and solid numbers under ~1/8th on creatures that the players are familiar with, basically if you can succeed on your “what monster am i fighting!” check (with the exception of things that don’t show obvious signs of damage, like ghosts) you get rough HP responses , and for creatures they aren’t familiar with they just get the 1/2 HP bloodied state. (And yes my 5e campaign has creatures with bloodied effects, it’s too good of a mechanic to just drop it).
When I did the tight lipped HP my players would get bored of longer fights with tough opponents because there’s no sense of progress during the fight, the fighter could crit and do 40+ damage in one attack and see… absolutely nothing happen from what should have been a cool moment for him, and now the monster with 1 HP left gets killed by magic missile or something. That just sucks for the fighter.
When I did full disclosure HP my players were just doing statistical analysis and the fights became math homework.
What i’m doing seems to work because 1- the players are rewarded in a meaningful way for doing their research and knowing monsters that they’re fighting. 2 – the players get a sense of progression through the fight. The fighter crits and sees the monster instantly go from full HP to bloodied, and that feels good. 3- you get good tension towards the end of fights. You know the monster has 10 HP left and you manage to hit it with your longsword for 1d8+4, so you know a 6 7 or 8 is victory, and a 1-5 is you getting your face eaten off and a TPK because you’re the last member of your party alive and you only have 4 health left.
My guess on the postscript:
Conflicts are not defined as “race to 0 HP.” Conflicts are different parties having conflicting goals. Combat is merely one method of resolving the conflict. And once it has become clear which side will achieve their goal, combat is no longer important.
Few enemy combatants will fight to death. They will concede or run away or beg or whatever first and thus the question of the conflict will be resolved. And the game proceeds.
Focusing on getting enemies to 0 HP is totally the wrong way to look at it.
This is the best, and most Angry-esque, answer I’ve found yet.
However, I generally find fights do go to the death, simply because once an enemy becomes damaged to the point of fleeing, by the time their turn comes round they’re either immobilised or dead. A mechanic allowing monsters to flee would be highly valuable.
Isn’t dropping your weapon or yelling “parley” still a free action in 5e?
Yelling “Parley” could work. I tried dropping weapons or turning to flee before, the party just decided to kill them while they could. I understand the reasoning, an enemy left alive can come back to bite you while a dead one is gone.
Devil’s Advocate: if combat in an encounter is really just about resolving who gets to achieve their goals, what’s the point of monsters having hitpoints at all?
Let me clarify, I ask that in the interests of being creative, as in, what can you use a hitpoint count for in a RPG to make the game better? To what use can we put this tool, just as AC and attack bonuses are tools?
Also notice I’m asking why *monsters*, not PCs, have hitpoints. For PCs, there’s at least one clear reason they get a HP count – it tells you how long you’ve got left before your character is going to be caught or killed and disavowed by the Secretary, etc. It tells you how much longer you can keep seducing Lady Luck.
(Could it be that monsters have hitpoints in order to keep the GM under control, tell *him* when it’s time to stop an encounter when all else fails? Maybe. Human beings suck at being impartial and unfair, and a lot of DMs also – especially those DMs in the low-teen to early-20s age demographic, who by dint of sheer biology don’t have a fully developed frontal lobe and therefore don’t have full adult judgement.)
I think perhaps HP can be a tool — *so long as you combine it with something else*. On its own, HP is just a number, and without more, players will use it, as said, as a measure of how long it will take to smash an opponent into the ground. Add the fact the monster hits with attack bonuses higher than the most heavily-armoured, most agile fighter has in AC, though, and the HP becomes something else: it becomes a warning to the players that this might be a curbstomper and they shouldn’t fight them. HP –combined with a Monster Manual that says the normal HP for this monster is considerably lower than this one’s — can also be used to hint to the players that something’s off here. HP could be phrased not as a measure of physical resilience, it could be characterised as the protection of the gods, or some magical effect (thus the stupidity of temporary hitpoints, which tries to objectify what is a DM measure to make the monster last a bit longer than it normally does.)
Some thoughts…
Regarding HP: Where AC is an abstraction of “how difficult is it to wound”, HP is (sort of) an abstraction of “how hard is it to kill”. AC is something an experienced fighter can figure out by looking at a monster, but HP is much harder to discern. You could guess based on how large a monster is, and how tough it looks, but you have no way of knowing for sure. Technically, you would need to be present for the monster’s entire life (between meeting a player, and dying) before you could reasonably determine a number.
That said, I think there’s something important that you lose by revealing a monster’s HP number to the players. The longer-term tension of combat ends when the monsters are defeated. If the players know when monsters will probably be dead (or beaten within an inch of their lives), it takes a lot of the tension out of combat. There is no more uncertainty in “will this damage kill it?”, because the answer is there in front of them the entire time.
That said, maybe there’s room for a medic- or necromancer-type player character, who could determine such things with greater speed and accuracy? Characters who are experienced with life and death would potentially have better insight into these things, though it would be limited to their personal range of expertise (human or other intelligent race(s)). Maybe.
To the issue raised in the first mini, which was something like skill resolution turns more into a puzzle than choice. A solution I use is to make an encounter where the NPC is immune to a skill or has a pre-programmed with a response. Not so much like INTERaction, but giving your NPCs more depth than falling to a player’s favorite skill check. If the town guard is a grizzled veteran he may respond to ANY attempt at intimidation by smirking and blowing his horn while drawing his sword–he has the rest of the watch to back him up after all. He may also pee his pants and run away while blowing his horn. Yes, it is still a kind of puzzle but at least it prevents every Arnold Stalone barbarian type from flexing at every NPC in town and forces him to interact.
If it’s a locked chest, eh, the rogue is gonna pick it. (Although after a six Orc combat, why does no one just smash the chest open? Glass vials full of gaseous “gaseous form” potion?)
Angry, thank you for helping me and my players enjoy role playing more. Compliments of the season and Merry Christmas. I hope Santa brings you a Nintendo.
Sneaky way to promote some discussion with that post-script. Is this how you plan on doing the “engage the community” thing, like with that martial archetypes thread?
My way of dealing with HP and their presenting to players
– I try to always tell numbers when they are our of ordinary and obvious – that zombies are so low AC that blessing has no effect, players autohit them anyway, or that the foe is in full plate with shield, so his AC is at least 20. For second example, players should knoe second part from first, but reality is different, there are always few who are surprised by such high AC.
I now always reveal AC after first attack against target. It come gradually but now it is a rule which works and make game smoother in the end. Same goes with vulnerability, reduction and vulnerability. I always tell it that as simple rule statement first. Then add flavor if I am able. “Skeleton is immune to ranged piercing damage. You see you arrows passing through or glancing from bones. Maybe on crit you can do damage. Melee piercing would maybe be only resistant, as you can use some leverage to do minor damage.”
I am not more and more often revealing information about HP. They are abstract. And they are not at the same time. They have direct impact on game. So I revel them more in general terms on following situation:
Characters spent time studying target
– then I tell HP in general intervals, like “weakling, less then 10”, “10-20 hp”, “more then 20 and less then 50”. I do not have precise ranges set and use my intuition there.
Target is damaged for first time
– here I say more precise HP, rounding down to tens. “His max HP is a bit above 120”
Target is below half
Players with one strike removes half or more of its remaining HP – “One more such strike and he is surely down.”
I did not do so in past, but I see it leading to more informed decisions and strategy. On the other hand it is taxing, so I experiment with othe r strategies, like letting players count HP of enemies. Gives me more free reign to handle the rest and keeping that information totally revealed to them so far did not lead to any game detriment.
5e D&D defines hit points as:
“Hit points represent a combination of physical and mental durability, the will to live, and luck. Creatures with more hit points are more difficult to kill. Those with fewer hit points are more fragile.
A creature’s current hit points (usually just called hit points) can be any number from the creature’s hit point maximum down to 0. This number changes frequently as a creature takes damage or receives healing.
Whenever a creature takes damage, that damage is subtracted from its hit points. The loss of hit points has no effect on a creature’s capabilities until the creature drops to 0 hit points.”
Said Angry:
“Sometimes, you keep the mechanics hidden because the mechanics would reveal information to the players that their characters would have no way of gleaning about the world and would drastically change their behavior. Note, though, that that statement is very carefully worded to be vague.”
Maybe the physical durability, mental durability, will to live, and luck that hp represent constitute info that characters have no way of gleening. Luck and mental durability in particular – how does a creature telegraph mental durability or luckiness? Characters also couldn’t estimate hitpoints based off of change in monsters capabilities, since hit point loss does not affect monster capabilities until hp reaches 0.
Fudging HP to properly shape the tension in the narrative is an attractive answer too. I seem to remember angry suggesting that you could just tell the party that they succeed in defeating the dregs of the enemy force once the dramatic question of a combat encounter has been resolved.
I argue for revealing HP precisely because it is so abstract and so critical. I once suffered a TPK because the sole survivor didn’t know he just needed to put 2 more hitpoints on the final boss and instead opted to use the -5 to hit +10 damage to hurt it more.
He also said that the outcome of most combats should pretty much be known after three rounds, which I have found to be true, assuming all participants in the battle have been revealed to the party. Much the same way is HP.
I am not completely sure I agree with the statement ‘more randomness equals more tension […] and, more randomness also decreases agency and the meaningfulness of the players’ choices’. I think, tension is more a combination (product) of the likelihood of failure and the size of the potential loss when the player fails the roll. Knowing the likelihood of failure (e.g. knowing the DC) can therefore increase the tension, which could be another reason to let the players known the DC.
Furthermore, I think, think that some randomness increases player agency and not decreases it. Without the randomness of the dice the outcome is either completely deterministic, or completely up to the judgment of the GM. The first option is only possible if the rules would cover everything (which is impossible as you mention) and even then it feels wrong to me probably because the lack of tension; the game becomes more like a puzzle (assuming that the GM has made a successful resolution possible which they can calculate as everything is deterministic). In case of the second option, it becomes difficult for the players to determine the outcome of their actions; it matters less how strong a PC is in the skill they use and more if the outcome aligns with the direction the GM wants the game to go. With dice, after setting the DC (the exact DC doesn’t matter that much) the result is out of the hand of the GM and in the hand of players. The consequences are still mainly in the hand of the GM.
One of the key differences between HP and AC is range. It makes AC descriptive of the world while HP an under the hood detail of the world. Almost all ACs fall between a 20 point range approx 8 to 28 this means when a character is looking at something and determining if it can hit it it only has to classify it within 5% to be able to say the AC is correct and they would only be wrong by 1 if they had it grouped into the correct 15%. It is even easier when the enemies are a large homogeneous identically outfitted group. I can see some argument in keeping some details secret it maybe difficult to determine dex/dodge bonus on someone you’ve never seen try to evade an attack or the natural armor of something you’ve never seen pieced before. opening with something like their initial estimate, but after a round of combat they should be pretty sure what the AC is. All the “non-attack” attacks that are supposed to go on during the round in the background would really show what the opponents abilities are. All this is really only if the creature is particularly unusual or deceptive. You could probably argue that characters would often be off by 1, but I don’t think that nuance is worth the extra book keeping
HP on the other hand easily run to over 100. It is a nuance of the world like how many candies are in a jar. I do think gross health levels should be determinable Thus I like conditions like undamaged, damaged, bloodied and the idea of announcing them. Of course nothing stops players from just looking in the MM and keeping a damage tally. Something I’ve wondered about is to roll the last two Hit Dice of the monsters HP, Like modify the HP by +2d12-13. This would prevent knowing exactly when the creature will die, but prevent huge swings in difficulty.
I sometimes feel like you could start your own religion. Judging by the comments, you’d have a lot of followers believing everything you say. I mean, it’s good stuff and it’s always so insightful.
Implying the Angry Cult of Angry isn’t amassing its army on the Patreon Discord as we speak, poised to take over the TRPG industry with a wave of unreasonably useful gaming advice.
I’m the kind of GM who prefer handling the math and translate it in an immersive way to players, so players have a rough idea of difficulty without having to tell them “this opponent has an AC of 17”. However, I also see great value in clarity of information, and the way you propose to reveal numbers doesn’t seem too invasive.
Think now I’m going to give it a try and see how it works out at my table.