A word of warning…
This is my once-monthly Random Bullshit column. And that’s just what it sounds like. I’m just going to rant and ramble and I promise only that there will be words to read. Lots of words to read. If I’m lucky — and I usually am — those words will convey lots of interesting and useful ideas, but there are no guarantees here.
Standard disclaimer disclaimered, I have no idea how to proceed. Either with this Long, Rambling Introduction or with the actual column opener. These Bullshit columns are often a chore to start.
The thing is, I never write this Bullshit for just one reason. I’m never just responding to some stupid piece of crap design that WotC shat into the ongoing DBox One playtest. I’m never just calling attention to some YouTuber’s dumbass commentary or some idiot thread on Reddit misunderstanding or misrepresenting some simple game design concept. And I’m never just finally weighing in on some discussion that’s consumed my entire supporter Discord server for days.
Instead, I write Bullshit columns after a single concept or idea won’t stop coming up in all those places and I need to shut it all up once and for all.
Today’s bit of Baader-Meinhoff topical synchronicity? It’s Challenge. And Difficulty. And Attrition. And Microchallenge. And the whole point of designing tabletop roleplaying game systems.
Why is it So Hard to Understand Challenge?
You just don’t get Challenge. I don’t know how else to say it. And I don’t know how else to start this column. If I tried to list all the dumbass things I’ve heard y’all say about Challenge — and Difficulty — in terms of encounter, game, and system design, I’d never get around to the meat of the article. Don’t get me wrong, it’d be hilarious to read and fun to write. But it wouldn’t be useful.
So let me skip justifying myself for once and just cut right to the “everyone’s wrong but me.”
Whatever you think you know about Challenge and Difficulty? You’re wrong.
Especially if you think they’re the same thing or the one measures the other. Wrong.
Whatever rulebooks and game designers say about Challenge and Difficulty? They’re wrong.
Whenever your game of choice uses terms like Challenge Rating and Encounter Difficulty? Those terms are misleading and confusing and wrong.
Whyever you think the Attrition Model of Challenge is wrong or bad? Wrong!
And however much the dumbass designers at Wizards of the Coast justify stripping out their rest and resource mechanics because “most games only have one combat per in-game day?” Frigging wrong.
Everyone’s wrong but me. And I don’t have time to argue. Instead, I’m going to use my time constructively explaining this shit to you. Trust me: it’ll be important later. Like around October.
Challenge and Difficulty: What They Are and How They Ain’t the Same
I’m going to do you all a huge favor and just spell shit out for you. Clearly and concisely. I’m going to tell you what Challenge and Difficulty actually are. How they’re related and how they ain’t. Just like that.
Challenge and Difficulty — here — are game design concepts. And because I’m defining a pair of concepts, the definitions are going to be freaking conceptual. They ain’t precisely quantifiable, numerical things. And I say that for two reasons. First, because tabletop roleplaying games love including numerical stats with Challenge and Difficulty in the names. And second, because gamers think if you can’t define a thing with numbers, you can’t define it at all.
Challenge Defined
Challenge is the extent to which a game’s outcome relies on the player’s skill.
That’s it. That’s literally all there is to it. This is why it’s so frigging baffling to me that people don’t get this. It ain’t hard. A Challenging game tests your skills. A Nonchallenging game doesn’t.
Craps and Candy Land are totally Nonchallenging. They’re barely games at all. The outcomes depend wholly on die rolls in the former and the random arrangement of cards in the latter. You can’t be good at either one.
Gambling on Craps is a different matter entirely. And a crap-ton of fun even if you can’t win in the long run.
Chess and Go are wholly Challenging. Winning’s entirely about making the right moves at the right time. There’s nothing random about it. And the only external factor — who goes first — is a minor factor.
Imagine you take a hundred players of various skill levels and make them all play the same game with the same setup. If the outcomes — final scores, victories, rankings, whatever — correlate strongly with the players’ skill levels, the game’s Challenging. If you have a few crappy winners and a few skilled losers mixed in, the game’s mostly Challenging. If lots of unskilled players win and lots of skilled players lose, the game’s on the Nonchallenging side.
Note the vague nonquantifiable imprecision. Because Challenge ain’t an all-or-nothing thing. There’s lots of space between perfect correlation and nonexistent correlation. It’s a spectrum. A phase space.
The Synergy of Victory
Now I, personally think that definition is super easy to grok. But experience has shown that some of you need some juxtaposition and context to really get what I’m driving at. So let me expand.
First, going forward, I’m going to drop Winning and Losing from the discussion and instead refer to Positive and Negative Outcomes. The distinction’s going to be really important a few headings down.
Lots of games these days aren’t about winning and losing. Or they aren’t just about winning and losing. Many games grade and reward performance and rank players. So there’s a huge spectrum of outcomes. And that’s especially true in tabletop roleplaying games.
And because this whole defining outcomes and grading performances thing is going to be a huge-ass sticking point below — and in the Fall of this year — I’m keeping shit really abstract for now.
In the broad spectrum of every game of every type ever defined, there are lots of factors that might figure into a game’s outcome. To simplify things — and to stay focused on tabletop roleplaying games — I’m going to discuss just three.
First, as I mentioned above, there’s Player Skill. That’s a broad term that describes any mix of player traits, qualities, talents, and skills a game can test. Games can test coordination, balance, strength, reaction time, fine motor control, strategy, perception, tenacity, memory, resource management, risk tolerance, interpersonal abilities, intrapersonal abilities, self-awareness, patience, conscientiousness, and so on, et cetera, ad infinity and beyond.
Second, there’s Random Chance. That’s not such a broad term and I hope I don’t have to explain it. Because I’m not going to.
Third, there’s Avatar Strength. This one’s pretty specific to tabletop roleplaying games and games with so-called arpeegee elements. Avatar Strength refers to all the numbers and mechanics that define a player’s in-game character. Including the character’s skills which are just mechanical statistics and have nothing to do with the capital S Skill I keep mentioning above.
And yeah, there’s a bunch of rabbit holes I’m steadfastly refusing to jump down. First, I ain’t going to talk a lot about which skills a given game should test. Second, I ain’t going to discuss the issue of Avatar Strength being a function of character optimization — a Player Skill — and random die rolls — Random Chance — depending on the game. Third, I ain’t discussing whether character optimization is a Player Skill tabletop roleplaying games should test. Fourth, I ain’t going to discuss the pros and cons of decoupling Avatar Strength from the game’s statistical balance to downplay its impact on the outcome.
Fifth, I apparently completely cut out the rabbit hole about flow-state and the problem of Challenge — not Difficulty — falling as Avatar Strength rises and I only just now discovered I cut it out as I was trying to figure out where I meant to put this graph I drew which is now homeless but which I will include nonetheless so you can all wonder about the thousand words that could have been and demand I write another article explaining it all.
It’s been a long time since I invited y’all to dance for an article, huh?
Rabbit holes aside, the point is just that, a Challenging game is one in which Player Skill has more to say about the outcome than Random Chance, Avatar Strength, or anything else such that, all else being equal, more skilled and more experienced players usually achieve more positive outcomes.
A Nonchallenge is a Nongame
Challenge is essential to game design. To the extent that games that offer no Challenge at all — games in which the outcome is wholly independent of Player Skill — aren’t games at all. They’re hobbies, activities, distractions, and diversions. Psychologically, they don’t affect players’ brains like games do.
Why? Because Challenge is the extent to which the players’ choices impact the outcome and it’s also the dimension of gameplay that allows players to play better with time and practice. In other words, it’s where both Agency and Progress live.
But this ain’t an all-or-nothing thing. It never is. Challenge makes games games, but the degree to which games challenge players can vary widely. Some players want their games to really test them, others need to feel tested but don’t want to be tested too hard. This is why Challenge is a play aesthetic and varies from player to player.
That said, players who don’t want any Challenge are wrong. Because they don’t want to play games. They want to do something else. And, hot take, there’s a huge number of neophyte Dungeons & Dragoneers that don’t want to play a game. They’re just not savvy enough to say it that way. And while that’s fine, Wizards of the Coast needs to know better and decide, once and for all, whether D&D is a roleplaying game or a collaborative storytelling activity. It can’t do both; trying to do both is making everyone miserable.
But I digress…
The point is that a game’s not a game — and doesn’t provide the psychological payoff that games, specifically, provide — if it lacks sufficient Challenge. That’s why, in D&D, the players must make the choices, devise the strategies, draw the conclusions, and choose the approaches. That’s why push-button-to-use-skill-on-skill-shaped-lock is so unsatisfying. And it’s why you can’t play a character that’s smarter than you. And why Charisma isn’t about your character knowing what to say, it’s just how they say what you want them to say.
But I digress… again.
Difficulty Defined
So that’s Challenge. Challenge is the extent to which Player Skill — not Random Chance and not Avatar Strength — correlates with a game’s outcome. So what’s Difficulty? Isn’t it just a way of saying how much Challenge there is?
No! It just ain’t that simple.
You can have Challenging games that aren’t Difficult. You can have Difficult games that aren’t very Challenging. But you can’t have Difficult Nonchallenges.
Challenge is how Player Skill is balanced against the other factors to determine a game’s outcome. Difficulty is the level of Player Skill required to get a good outcome. To win or get a high score or have a good run or whatever. Whatever the game considers Good Performance, Difficulty is how much Player Skill it takes to achieve it.
A game can be highly dependent on Player Skill. Roguelikes — not Roguelites — for example have no advancement to speak of. There’s no Avatar Strength to consider. Every run’s an independent trial. Randomness is a big factor, but skilled players can perform well even when the RNG is against them by mastering the game’s systems. In fact, that’s the point of a Roguelike. To master the game’s systems so you can handle any run the game throws at you.
Roguelikes are very Challenging by their nature.
But Roguelikes can also vary tremendously in Difficulty. The Player Skill required to have a good run can vary wildly. That’s the difference between Challenge and Difficulty.
And while players — and some game designers — conflate Challenge and Difficulty, the smartest designers know they aren’t the same and that Challenge-seeking players aren’t all looking for punishing Difficulty.
When Game Mechanics Disguise Themselves as Challenge and Difficulty
Challenge and Difficulty are game design things. And they’re specifically to do with how players interact with a game’s challenges. But tabletop roleplaying games like to fling around the words Challenge and Difficulty. They’re even used as game mechanical statistics.
And that confuses the hell out of everyone.
In Dungeons & Dragons for example, Character Level and Challenge Rating measure the balance between the characters’ Avatar Strengths and the statistical, mathematical game obstacles and enemies. And they take into account Random Chance.
Thus, Challenge Rating measures everything except Challenge. In theory, it’s still useful. After all, if there’s a gap between an obstacle’s stats and Avatar Strength weighted for Random Chance, Player Skill can make up the difference. In other words, if a 3rd Level Character is 40% likely — by the game’s math alone — to overcome a 3rd Level Obstacle, a Skilled Player should be able to bring those odds up to something acceptable.
In theory.
Thus, Encounter Difficulty — remember, in the Dungeon Master’s Guide, encounters are defined as Easy, Medium, Hard, or Deadly — Encounter Difficulty measures the gap that Player Skill has to fill. An Easy Encounter is one most players can win regardless of their Player Skill and a Hard Encounter is one that most players will lose unless they lean hard on their Player Skill.
In theory.
But that ain’t quite how it works in practice. And there are a few reasons for that. Some derive from the fact that most Adventure Designers and Homebrewers — and more than a few professional Game Designers — don’t really understand this crap. And that’s why so many Game Masters keep trying to fix perceived problems with the game’s Challenge by futzing with the numbers. Or blaming Game Balance.
But the big problem’s that Dungeons & Dragons doesn’t really do Challenge and Difficulty on the Encounter Level. That’d be stupid. And thus Challenge Rating and Encounter Difficulty are actually about Rates of Attrition.
This is actually a really brilliant approach to a pretty big problem in tabletop roleplaying game system design.
And I will explain precisely why I think that dirty, dirty word is actually brilliant design and why the idiots at WotC really need to stop listening to dumbass players and stripping it away. But first, I’ve got to explain something else.
All Challenges Great and Small
We’re going deep into the game design weeds here. And really, we shouldn’t have to. If D&D designers did their jobs right — and arrogant, prideful Game Masters would just read the books and follow the instructions instead of breaking everything and then complaining the designers should have explained why the game doesn’t work if you break all the rules — you wouldn’t need to read this and I’d still be an accountant laboring away in the debit mines.
Man, I miss the smell of ledger ink in the morning. But anyway…
The whole Attrition thing — and the reason why Challenge and Difficulty don’t work the way you think they do — is down to Macrochallenge. And Macrochallenge is an aspect of Scenario Design.
A Higher Level of Design
Scenario Design is the part of game design wherein you string all the game’s pieces together into something playable: a level, a dungeon, an adventure, a campaign, or whatever. In tabletop roleplaying games, Scenario Designers build actual, playable adventures. And Scenario Designers include professional and independent publishers and individual Game Masters who write their own adventures.
Scenario Design’s the purview of Adventure Designers and Homebrewers.
Scenario Design is the reason, by the way, that you don’t improvise your whole frigging game no matter how good an improviser you think you are. Sorry.
Scenario Designers turn game components and game engines into actual, playable, fun games. Remember, the Dungeons & Dragons core rulebooks don’t contain a game. They’re a mishmash of rulebook, game engine, and developer toolkit. Adventure Designers and Homebrewers use those tools to make actual games. They do the Scenario Design. And then Game Masters make the game happen.
From Microchallenge to Macrochallenge
It is easy to design a Scenario. Just build a gauntlet of individual Encounters and dare the players to get through them all. That’s the simplest Scenario there is. A slightly more complex Scenario involves scattering those Encounters around an explorable space. Sound familiar?
But that setup’s got a problem. If the players fail any Encounter, they lose the entire Scenario. It’s an All-or-Nothing Affair. And given every individual Encounter must include some chance of failure, the more Encounters you put between the players and the finish line, the more likely they are to eventually fail one of them.
In other words, you’ve got a game the players must play perfectly every time until they’re done. Fair enough if you’re designing a video game because the players can just keep respawning until they manage to not fail through to the end. But tabletop roleplaying games are built on persistent, permanent consequences and they’re built to go on pretty much forever.
The solution’s to design a Scenario such that the individual Encounters — hereafter called Microchallenges — don’t matter so much and that the Scenario itself provides some kind of big, overarching Macrochallenge.
Macrochallenge Exemplified
This Macrochallenge versus Microchallenge thing can be tricky to grok. Let me offer a few examples to make it clear.
I’ve been replaying Middle Earth: Shadow of Mordor lately. I love the combat but the challenges don’t scale well with your Avatar Strength so the game gets easier the longer you play it. That ain’t great. The point is, I got to the point where it was trivially easy to slay groups of orcs. But then I hit this side quest wherein I had to chase down and kill three groups of orcs within a strict time limit. And that’s a Macrochallenge. Killing the orcs — the Microchallenges — was easy, but victory required me to do it quickly and efficiently. And to navigate quickly between the groups.
Mysteries are great Macrochallenges too. Investigating a mystery might involve extracting information from witnesses, hunting for clues, and fighting off thugs — Microchallenges all — but however well you handle all of that shit, victory is about actually deducing the answer to the question posed by the Scenario.
In my module, Fall of Silverpine Watch, the players must either destroy a ghost or convince it to stop being a problem. Destroying it’s really hard unless the players thoroughly explore the dungeon and use the magical items they find. And interacting with the ghost is almost impossible without solving the mystery surrounding the ghost’s final moments.
The outcome of a Macrochallenge is dependent on something more than just the outcomes of the individual Microchallenges. And Macrochallenges are often great tests of Player Skill even when the individual Microchallenges aren’t.
Does This Sound Familiar?
Let me provide one more Macrochallenge example just to make sure y’all really get this shit.
Imagine you’ve got this heavily Encounter-based game wherein the players explore of their own volition and face lots of independent Microchallenges. How do you avoid the Play Perfect Forever trap and build in a Macrochallenge?
First, make it really hard to fail the Microchallenges. Really hard. Like, only a complete idiot would fail a given Microchallenge. It ain’t impossible to fail one, but it’s pretty close to it.
Second, build a bunch of resources into the game such that, as the players face individual Microchallenges, they either expend or lose those resources.
Thus, the Scenario’s outcome doesn’t hinge on how the Microchallenges come out, but rather on the efficiency with which the players deal with each one. The players can win every Microchallenge but still end up with a bad outcome if they waste too many resources on the way.
That, right there, is the Attrition Macrochallenge.
Attrition is Not a Four-Letter Word
People slag on the Attrition Macrochallenge in tabletop roleplaying games like Dungeons & Dragons a lot. And that ain’t fair. Because it’s really a brilliant way to solve a very difficult problem in roleplaying game system design.
When you design a roleplaying game system, you know — if you’re smart — that you’re designing an incomplete game. You’re building an engine. Someone else will have to do the Scenario Design. And you can’t count on that person being a Scenario Design virtuoso. Because you’re selling your engine to the masses. To armchair designers and interested laypeople and rank amateurs.
Good roleplaying game designers know they’ve got to build systems that empower amateurs to design good Scenarios. And they know those amateurs ain’t gonna read hundreds of pages of Scenario Design theory. And even if they would, those hundreds of pages wouldn’t turn most of them into good Scenario Designers.
The trick, then, is to hardcode as much good Scenario Design as possible right into the game system. Which ain’t easy. Scenario Design is a really complicated art. Hell, I’d argue that it’s way more difficult — and important — than Engine Design.
This whole Macrochallenge thing is a vital Scenario Design component. Without it, you’ve got a string of Play Perfectly Until You Fail Microchallenges that feels crappy and it isn’t obvious why it feels crappy because Scenario Design is really subtle. You can play through countless examples of great Scenario Design — like Hollow Knight and Dark Souls and Link’s Awakening and Megaman — without ever noticing what they’re doing that feels so good. And examples of good Scenario Design are thin on the ground these days because it’s way easier to shit out Yet Another Open World Action-Adventure Sandbox.
I won’t claim Dungeons & Dragons has had an Attrition Macrochallenge baked in from the beginning. It really hasn’t. But the seeds of one were there thanks to the intuitive design savvy of folks like Gygax, Arneson, and others. It wasn’t until the 3rd Edition Dungeon Master’s Guide that the Attrition Microchallenge was fully built out and presented as the Core Scenario Design.
As I said, you can’t turn every Homebrewer into a Game Designer. It’d be stupid to try. But what you can do is codify a good example of basic Scenario Design in your developers’ toolkit and ensure it’s built around a solid Macrochallenge. That way, even if the Scenario Designer brings nothing else to the table in terms of good design, they’ve got a minimally fun gameplay experience.
The Attrition Macrochallenge was a great choice. It lent itself perfectly to the play experience Dungeons & Dragons already provided — it was basically an evolution of the game’s design — and it was also fairly innocuous. Hell, absent any other mechanics, the Attrition Macrochallenge isn’t even about winning or losing, but rather about how far you can push and how much you can accomplish.
The Attrition Macrochallenge is easy to layer other Macrochallenges over. A timed adventure is basically just an Attrition Macrochallenge with a hard cap on how inefficient the players can play before they lose. Random encounters and restocking dungeons are about dropping crap in the way of players’ progress due to their inefficiency. Event-based encounters take control over the gauntlet away from the players, punishing them hard for inefficient play by not letting them recover their resources. And optional side areas provide rewards for players who are willing to risk inefficiency.
It pisses me off when people shit all over the Attrition Macrochallenge. It’s a brilliant, elegant design solution. And it drives me nuts when Game Masters reject it out of hand and then piss and moan that the game doesn’t work when it’s ripped out.
If you don’t like the Attrition Macrochallenge, that’s fine. It’s a pretty rudimentary Macrochallenge, after all. But know that it’s doing something important. It can’t be removed; it must be replaced. And you’d better be ready to do some serious heavy lifting. And it ain’t fair to demand the system designers accommodate you. It’s unreasonable and impractical. And if you don’t understand why that demand is unreasonable and impractical, you’re probably not savvy enough to do the heavy lifting you need to do. No offense.
It pisses me off more when people simultaneously complain that Dungeons & Dragons is too easy — that it doesn’t challenge the players — and also that everything in Dungeons & Dragons boils down to life-or-death-victory-or-defeat.
No, Dungeons & Dragons isn’t too easy. The Microchallenges are supposed to be borderline unfailable. And yes, there is a state between victory and defeat. It’s called efficiency. You’re just too damned fixated on the Microchallenges to see why that shit matters.
But what pisses me off the most is when the stewards of a once-great game full of brilliant tools that let total amateurs build totally fun games without skill or experience say shit like, “Well, we’ve found most players don’t play more than one encounter per in-game day so we’re just stripping out the resource management aspects no one’s using.”
See, I can accept clueless amateurs and armchair designers don’t get this shit. And I know such people always think they’re smarter than they are. So I can accept they’ll blithely ignore the clear instructions and then blame their broken-ass game on the instructions not explaining enough why not following the instructions breaks things. But I absolutely cannot accept so-called game designers taking their cues from those dumbasses.
The end.
Man, you’re right. I did indeed not get attrition.
That graph, by the way, seems pretty clear to me. A designer can increase the challenge a little bit each adventure, presumably following increased player skill as they learn to play the game “better”, or vary challenge up and down, as long as the moving average goes up.
Thinking back to the non-cr system with its tiers, I suspect you prefer the right-hand graph but I look forward to reading all about it.
Don’t forget that not all PC level advancements are the same, and that some levels (i.e. 5) are significantly more powerful than others, creating a jagged increase in difficulty. (h/t megadungeon)
Good article, thanks for spelling this all out in this manner. Especially the part about the microchallenges in D&D being nigh-unfailable. This is something I *should* know, but it keeps slipping my mind. I hope this article will help me internalize it.
If I’m understanding this correctly, here is an example for whomever in the Q&A was wondering. These is not the “type of macrochallenge” you were after, just a specific example:
You need to get the baby through the Stormlands to the Safe Place. If she arrives happy and healthy, your reward is likely to be greater (Macrochallenge). It’s 80 miles away (Whatever stat). “Mood” is another whatever stat from 0-10.
Likely overthinking this but mentally wrestling with the “use the dice and combat when needed” and the “micro challenge is nigh unfailable”. Nigh unfailable sounds suspiciously like the Tao of Dice and if you are certain of the outcome. Probably comes with practice and of course there are other macro challenges.
What might be uncertain in the night unfailable combat could just be the amount of resources required to reach a favorable outcome, or their could be variable outcomes to any given encounter whether it’s combat or not.
The cost of missing an attack in combat is simply a wasted action, yet we roll for that fighter’s attack despite the fact that he can attack all day and it didn’t cost anything but time and he can just try again till he succeeds because the time cost is important.
We roll the entire combat despite the fact that the only cost of screwing up a single combat is a few HP or an extra spell or three wasted, because the HP and spell costs are important.
The loss in even a deadly combat in D&D 5th edition is not “You die”, it’s you spend more resources and three encounters later you either need to retreat without accomplishing the objective or you die. It’s nearly impossible to die in a single encounter made by the book, IIRC deadly is defined in the DMG as “might put down one PC till someone else heals him, and we made healing him real easy”, But the attrition matters and that’s why you’re rolling dice. I’ve told my player’s “You win” when a fight was a forgone conclusion and there was not going to be any important attrition, but you need that attrition.
This is why I always hesitate at calling a battle early. Sure, they’re going to win, but they were always going to win – the question is what the win will cost them.
The micro-challenge isn’t in doubt, but the macro-challenge is. So you use dice to solve the micro-challenge because how well they succeed on the micro-challenge determines the success of the macro-challenge.
Yes, attrition has always been great. I also don’t like how WOTC has been catering too much to people’s complaints.
If you complain about it, they might listen.. 😀
Oh WOW! This article crystalized so many things for me! Challenge as player skill, challenge not just meaning difficult, the challenge being dependent on the scenario design, and it not being a game without the challenge, these all resonated with bits of other Angry advice bouncing around in my skull.
That’s why I can’t just have a game based on random tables. That’s why I need to put in decision points. That’s why just rolling to overcome the obstacle feels so meh.
My favorite line though, something I think I’ll paste to my DM screen: “That’s why, in D&D, the players must make the choices, devise the strategies, draw the conclusions, and choose the approaches. That’s why push-button-to-use-skill-on-skill-shaped-lock is so unsatisfying.” This is how quicksand can be a satisfying encounter!
And then the whole attrition macrocycle. That explained so much! No wonder it wasn’t fun to just increase the difficulty of each encounter to deadly!
Thank you Angry, this was so illuminating! (Although I was caught off guard when you called D&D brilliant… but I see now you were talking about 3.5)
It’s not even that 5e isn’t brilliant, as that it was designed with pretty much no respect. The real lesson wizards designers brought into 5e was that ever being honest with the people buying their game about how the game worked ever again was their biggest mistake, and everything needed to be slight of hand to keep the babies from throwing their toys out the pram.
FWIW I think the TSR-era D&D games actually do the attrition macrochallenge quite well. It may not have been as codified as it was in 3e+ in terms of designing a scenario to have a specific amount of attrition scaled to the PCs’ levels (instead favoring open adventure locations without a necessarily pre-ordained path for the players to follow), but that was fine because the *players* provided that by deciding on their own how deeply to push with the HP/spells/etc. they have remaining, and knowing that if they leave and come back they’ll likely have to deal with reinforced defenses, reset traps, new ambushes, etc. from the remaining inhabitants. The whole “dungeon level danger and rewards roughly correspond to character level” thing TSR-D&D had going on is a fantastic way to have players do the work of balancing the attrition on their own. They decide how deeply and how long to delve, and turning back always has a cost.
Yes. Old-school is also far more geared toward player skill. The “push-your-luck” element is a tremendous, tremendous game mechanic.
WOTC has gotten to where they need to sell Oreos. Oreos are really consistent and nobody really hates them, but they’re also mediocre cookies. You’re even supposed to eat them with milk…they’re not even complete out of the package.
So to make Oreos “fresher” or “better” is a waste of time. All you can do really is put a different flavor-chemical into the white glop in the middle, or maybe add more of the glop. For 5e, moar books! Moar rules! And if the Wal-Marters whine that the glop isn’t sweet enough that glop will get sweetened because Carl Icahn does not give a shining damn about the integrity of role-playing systems.
I’d say 3e actually started the trend of pulling attrition out of the game, sadly. Maybe it was deliberate, maybe it wasn’t, maybe it was just the dozens and dozens of splatbooks that WOTC put out. Either way, I don’t know whether 2e D&D had Everlasting Rations, Everfull Mugs, or Decanters of Endless Water, but 3e sure did. The very existence of those magic items meant that unless you’re the sort of DM who also likes to regularly hit the party with Disjunction, Rust Monsters or long stints inside dead magic zones, you basically can’t use attrition of survival resources as a macrochallenge, and travel becomes just a matter of how long you futz around lost than how long your supplies last. I know you could use Wilderness Lore/Survival to then get along in the wild more or less, but at least it cost you more full days of time to do it.
Some might then say ‘but travel should be easier and more convenient once the party levels up, that’s why the spell Teleport lets you go about 900 miles in a single casting, even at minimum caster level.’ Sure. But it doesn’t make the DM’s job any easier, and unfortunately WOTC doesn’t feed DMs macrochallenge mechanics to replace the lost ones when time and resource attrition are no longer immediately significant. But then that’s more about the sheer optionality and power of casters, arcane and divine, once you hit (and indeed before you hit) about level 9 or so, and which is basically baked into 3e if not all editions bar 4e’s Procrustean Bed, which WOTC definitely had a diarrheaic episode in.
It’s more that in the earlier editions the attrition macrochallenge was something you pushed through to get into domain level play, which didn’t really have attrition decision points at the same scale. While in third edition the attrition just went all the way up to the end of the game, meaning that while you could transcend attrition, you couldn’t transcend to anything anymore.
I think what impairs the attritional value of the older editions is 1. Their love of reducing challenge in favor of risk-taking instead, with potions of random effects and similar nonsense and 2. The design intention that D&D was meant to be played as a competitive game between parties, meaning that you could often coat-tail off another party’s near failures to slingshot yourself to ‘unearned’ success.
Excellent article. I really like this insight into the role of the scenario designer.
This is another of those articles where I read something I in particular want to call out as exceptionally incisive or perceptive or illuminating, but then I read another and another and another, until it winds up being the whole thing!
I never really gave attrition any thought beyond “Of course you’re tracking encumbrance and ammunition!”, never designed with attrition in mind, I just designed and ran my adventures, and they mostly ran pretty well. (Which is a testament to the designers up to 3rd edition more than my adventure design skills.) I always did it intuitively rather than intentionally, saving my “intentionality” for the…. conceptual(?) challenges of the adventure rather than the mechanical.
You said in your last BS article that when people complain, there’s often a problem, they’re just always wrong about what the problem is. Could it be the case for attrition? I do think attrition is an essential macrochallenge in D&D, and it shouldn’t be removed, but its implementation does rub me the wrong way regarding combat ressources (offensive spells and limited use abilities). I find it turns combat into a game of “should I use my abilities now?” rather than “how can I use my abilities to their best effect?”, which I find less interesting.
The same idea is forming in my head. I suspect the issue isn’t that I don’t get attrition. This article didn’t tell me much new except a useful definition of challenge. Instead I suspect I don’t get how to make an attrition-based adventuring day exciting.
Because there isn’t much tension in a combat where it’s not about live or death, but about whether you lose ten or thirty hit points out of your hundred. And then you do that five times and if you didn’t do well enough, well, there’s now finally a moment of tension where you choose whether to push on or just… leave. Go home and rest. No boss fight, no boss loot, just an anticlimax, try again tomorrow.
Though I’m likely wrong (and it’s also obviously simplified.) Like I said, I suspect I don’t get it.
>Because there isn’t much tension in a combat where it’s not about live or death, but about whether you lose ten or thirty hit points out of your hundred.
This is the basic principle on which Dark Souls functions. In Dark Souls, the tension comes from how challenging going into a boss fight with “too little” health or healing is.
In DnD and the like, (since the “prepare to die” mindset of Dark Souls doesn’t translate well) the tension has to come from externalities. From narrative – the Roleplaying half of the Game.
Retreating should have consequences – the enemies will be on guard next time, so it’ll be harder to get in; the villain completes their dark ritual and/or escapes; the prisoners get executed. Of course, it’s equally possible that the consequences is simply “you can’t retreat to rest, so now maybe you die in the boss fight”. But it’s not how most people design their games, for the “play forever” reason mentioned by Angry above.
So what resources could you add to this in order to make it interesting?
There is an optional rule in the 5E rules that makes short rests the “once per day” rests, and Long Rests takes a week. How would you as a player play 5E if your GM introduced this concept?
Now say I as the GM also put the dungeon several day’s travel from a spot where you can have an actual long rest? And I make you track rations and water every day.
It also means you need to set up camp somewhere. Which means camp supplies. How about we also introduce the variant rule on encumbrance, which is quite logical when it comes to how much someone can logically carry.
What if none of the player characters get dark vision? Now you need torches and light sources.
Now suddenly you have a lot of attrition to care about, and not every encounter is a combat encounter.
The delayed feedback is the worst part for me. When every battle gives the same positive binary “you won” signal, it’s hard to see that efficiency. Worse, when you do overextend and fail, there usually isn’t a single obvious mistake to learn from. If you’re in a standard inefficient battle group, the mistakes that depleted resources too quickly could have been split across multiple sessions.
Also, it’s fun to use spells and abilities.
Agreed.
Attrition is a useful tool, but it is definitely widely overused.
I feel like the players created the characters they did for a reason. And usually, the reason is for the character to be able to do one or more things (abilities, spells, special feats, etc.). Keeping them from doing those things they created the character to do seems like counterproductive tedium. If a player really likes to have his skulker sneak around, let him sneak around; if another player really like her druid to plant trees, let her plant trees; and if the other players really likes his guy casting fireballs, let him cast fireballs. Putting a counter on it just for the sake of telling them they can’t do their cool things x% of the time seems unnecessarily tedious. If I’m going to take away my players’ ability to use their cool things it’s going to be for a specific purpose with an interesting (hopefully) story element behind it, not just because the game-clock or inventory checklist says so. I’m a Dungeon Master, not an auditor.
You’re looking at it backwards. Attrition isn’t so you can eventually take away the player’s tools, it’s so violence doesn’t immediately take away the player’s tools.
Without attrition when a monster wants to resolve a conflict by killing a character you have two outcomes: the character is alive and the player can play, or the character is dead and the player can’t play.
If you take away attrition you either take away the risk of violence or you make it so violence will eventually kill you more or less at random. Attrition gives players opportunities to respond to violence, it makes it possible for some responses to be better than others, and
it gives player’s agency over the life and death of their characters.
This doesn’t make sense:
“If you take away attrition you either take away the risk of violence or you make it so violence will eventually kill you more or less at random.”
How do you figure? It doesn’t take attrition to make violence dangerous if you act foolishly and make poor choices, but not terribly threatening if you act wisely and make good choices. THAT is player agency over the life and death of their characters.
I think it’s definitely true that something rubs people wrong about the combat/adventure design in 5e. I don’t know if that’s the edition you play, but I’ve certainly seen it in my games and in discussions online.
I suspect there are many factors at play, but one of them is how players visualize consequences of fights. Healing out of combat is trivial in 5e (short rests, back to full at long rests, etc.), so that taking an extra turn to kill a guy with cantrips rather than levelled spells/features doesn’t feel worth it. In contrast, in 5e there are *so many* limited use features. And you can watch those slowly drain.
So, I think you end up with a situation where players are biased towards tedium (inefficiency isn’t a function of HP/time, but rather limiting use of other resources) and there aren’t punishing consequences for doing so. So, you build up a negative experience (watching resources drain, in tedious combat after tedious combat) without any long-term reward metric. Or a DM ditches this approach, runs one combat a day, the players go Nova turn one and there’s a lot of short-term gratification without any long term emotion.
I find it quite amusing when players and DMs of D&D 5e express dissatisfaction, claiming that the game is too easy. As this article astutely points out, the system was designed with a focus on resource management and attrition-based Macrochallenges in the past. Yet, paradoxically, those very individuals who complain about the game’s ease also tend to grumble about mechanics that add difficulty to resource management, such as Exhaustion, Encumbrance, and other debilitating mechanics. These mechanics are present precisely because D&D 5e is intended to be a resource management system.
While I won’t argue that Exhaustion, Encumbrance, and other mechanics are flawlessly designed, it’s important to acknowledge their purpose within the game. Ironically, those who remove or disregard these elements often find themselves complaining about the game’s lack of challenge. If you genuinely desire a more challenging experience while playing D&D 5e, my suggestion is to embrace these rules and witness how swiftly the game becomes difficult, perhaps as originally intended.
I think there is something to be said about a preference for the challenge to be somewhere different: Players have a feeling that attritional combat doesn’t really matter, that the difference between beating an encounter with 90% versus 85% of your resources shouldn’t matter (as in, in a good game designed differently it would not be something that matters), that failure inside an encounter should be caused by events inside that same encounter, and not events in an encounter that actually happened maybe 3 real life weeks ago.
Quick appreciation post for how much work goes into even a bullshit article like this.
Don’t remove a fence until you know why it was there in the first place
If you don’t have resource attrition, you lose most of the drama of pitting yourself against an uncaring universe. Resource limitation is key.
I remember that Chesterton wrote about the dramatic power of ‘Robinson Crusoe’ coming from its themes of resource scarcity mitigated by ingenuity.
Expanding on this: so many effective narratives derive their power from economy.
Not just resource economy such as “how can this shipwrecked mariner survive with just this list of items?”
I’m also thinking of *dramatic* economy. By which I mean “doing a lot with a little, or doing a lot under severe constraints”.
Drama comes from severe limits. For instance:
“How can this girly cheerleader defeat the Slasher?”
“How could Roger Acroyd have been murdered in a locked room?”
“How will this legally blonde bimbo leverage her knowledge of cosmetics into winning a murder trial?”
“How will this penniless Fugitive with the full force of the US Marshals service after him both escape and clear his name?”
“How will this Grade II Docking Technician deal with a Xenomorph with just a spacesuit and a tether-gun?”
And – returning to RPG – “how will this rag-tag party of misfits with only half their spells left defeat the Lich?”
You definitely hit on it here. Incidentally, this may be why I am pretty tolerant to story spoilers. Someone might tell me “So-and-So dies at the end” but that still leaves me wondering how. The how is actually the interesting part. It’s another one of those “The joy is in the journey, not the destination” things.
THANK YOU, Angry, for writing this. Now I finally understand why I love Dark Souls’ combat (it’s challenging) yet keep bouncing off the games themselves (too difficult).
Very interesting article Angry. I found it very instructive to put into words things I had a vague intuitive awareness of.
One of the reasons I think people hate the attrition system though is because of the way it is mechanically implemented in D&D (particularly 5e). The “6 to 8 encounters a day” leans very heavily on combat. The attrition design is easy to implement on combat, but much harder on other types of challenge. Spells or abilities start to completely bypass most environmental challenges pretty early (food/water scarcity, bypassing obstacles, warmth, rest in dangerous location – talking to you, Leomund’s tiny hut… Exhaustion is also too punitive to really be seen as an attrition-type mechanic.) All of this mean that if you want to have some diversity in your encounters and still rely on an attrition design, you’ve got a lot more heavy lifting to do than if you were just to string a couple of combat one after the other. All of this because combat take a lot of ressources while other challenges will take a comparatively small amount of ressources.
Thank you for explaining the fundamental concepts and the terms for them. If the Game Mastery series had a glossary, they would belong in it.
Moreover, it helps me understand why I hate D&D combat so much. Low difficulty encounters to challenge my resource management feels like a push-button-to-use-hp-reducer-on-hp-filled-obstacle-lock. It’s not the system’s fault that I am looking for combat as a punishment mechanic for having failed to overcome the obstacle without risking injury or death in violent struggle.
Hey Angry, can you list Macrochallenges other than Attrition?
So one example might be time (can you reach/kill/do X before 5 rounds/hours/days have passed?)? But that might be another spin on attrition…?
Agree, would be good to have a shotgun of different macro challenges. Not as fleshed out systems, but just a sentence or so for each, and perhaps an example of another game which implements it?
I’ve struggled to work any other macrochallenges into my game, which means that outside of combat-heavy situations (e.g. social, investigative, etc; most urban scenarios) where attrition doesn’t work as well, I’ve found it difficult to build challenge.
Quote from the article “The Attrition Macrochallenge is easy to layer other Macrochallenges over. A timed adventure is basically just an Attrition Macrochallenge with a hard cap on how inefficient the players can play before they lose. Random encounters and restocking dungeons are about dropping crap in the way of players’ progress due to their inefficiency. Event-based encounters take control over the gauntlet away from the players, punishing them hard for inefficient play by not letting them recover their resources. And optional side areas provide rewards for players who are willing to risk inefficiency.”
Ooof. The number of times lately that I’ve seen people complain about something in D&D and I point out it’s addressed in the rulebooks is way too high. Added to it the response is always something about how whatever rule that is is stupid or that they didn’t notice it because the rulebooks are poorly written.
I’m starting to realize people pay way more attention to the race and class parts of the rule book and not the actual chapters that have to do with how the game works. Or that introduction.
This rule of attrition is one of those concepts that just started making sense to me on my who knows what readthrough of the rulebooks and is really changing my scenario design. Thanks for the clarity.
I think an interesting complaint we might have then behind the player experience that there’s something extremely unsatisfying about playing non-spellcasters in 3.5 is simply that, as attritional resource depletion is the macrochallenge of the game, spellcasters are doing more playing the game per playing the game than comparable martial characters who have fewer, more coarse, or no resources to manage.
Some martial classes e.g. battle master have some resources, but you can also increase challenge for martial classes by giving them limited use (or limited use/day) items.
5e maybe did a slightly better job of addressing this: The third edition battle master didn’t come out until one of the final supplements. Finding a way to make sure the item ends up in the martial’s hand, and not the spell caster’s, is the bigger challenge there.
I wonder if 4e was an attempt to extend additional resource depletion to all classes, and the backlash against it was seen by the designers as evidence that resource attrition was not what people wanted…
Well that’s not true.
The martial characters main resource to manage is their Hit Points, they are in the front line supposed to soak up the damage.
If they are the bow shooting type they need to manage their arrows.
Then there’s Rations and Torches. Something that most people seems to have stopped playing with. Once you throw that into the mix you get even more resources to manage. The Fighters are strong, they can carry more of it. Heck encumbrance is a resource too.
I mean no offense by this, but in a way your comment sort of exemplifies the issue. The problem is that people stopped tracking the “boring things,” and 5E has made some of it very easy to ignore. HP for instance isn’t an issue. Encumbrance is hardly ever tracked (There’s great optional rules for it in the rules actually). Not to mention everyone and their mother has dark vision, so why bother bringing a torch along?
And for some reason no one needs to eat and drink.
Not to mention money: With all this taken out of the game, and living costs not being tracked either, suddenly money becomes worthless at some point. Where as if it’s scarce it to becomes a resource to track.
I absolutely agree. It doesn’t help that D&D never had a working economy (see Angry’s article on that…) and that many players can’t be bothered tracking stuff, so DMs handwave it. but actually, its part of the fun and an easy part of video games to learn from (see Angry’s article on that…) if you build in ‘re-stock’ opportunities. In two decades of playing and DMing 3.5, my group and I never felt there was a lack of attrition for non-mages.
Martials resources are ammunition, any potions they carry and, most importantly… HIT POINTS. I find martials tend to gage party robustness on HP and healing left, while casters look to how many spells are left to them.
Fantastic article Scott!!! As pretty much always.
I convinced my group to switch back to 3.5 from 5e for our next campaign, over basically this point.
The attrition mechanic still exists in 5e, to some degree, but designing adventures that have meaningful attrition (enough the that the short rest heals and arcane recovery and endless cantrips and near-infinite carry capacity the players have access to doesn’t just nullify it) has been a recurring frustration from my side of the screen.
Mostly on the design side. I’ve thrown out dozens of simple fun straightforward adventure ideas, because they don’t feel rewarding to get through unless I also add some extra challenge or complication that the party can’t just make trivial with a nap. So now every scenario and ‘simple’ dungeon crawl ends up withsome more complex element so there is still something that feels like the party ‘overcame’ the scenario (or failed to overcome if they took to many naps or made bad choices etc.)
attriction is bullshit wining ones heart is not a game of twenty sided dice play and gualent contradictory insights how is one supposed to even understand the game play if logic has but little to do with it. Personally i think the game designer is a little nuts the psychic department is a little airy if you ask me honestly say you got four choices all 80 miles away and you are trhe traveler how are you supoosed to know which direction to choose its not like theirs four of the same person in each given area. A tiny tiny bit of logic may need to be processed into the script adventurer after all this script doesnt come completely out of one flew over the cuocuo nest did it id hope not. A touch of quantum mechanics would be nice after all the woman at the end is not one of four quadruplets is she and if she is then woops sie daisy excuse my unknowingly thought pressumtion. either way life and challenges on a board game and in reality are two different areas of expertise and while one might be able to construct it does that one know anything about her challenge or does she even have one.Challenges are all good and fun and honestly i dont mind a good challenge but blasting ramdom thoughts telepathically at ones head and then expecting him to know which telepathic thought is a lie or a different person or even a legitamet heard thought and then to convey a rude protreyal of the persons character is a little to non compasationate which may have that player wondering what he was doing in the first place taking the time try and understand and align with a dungeon master of such crazy and apparent unloving nature how a dungeon master can be so blantently different for one side to the next would probably make me think he or she was at risk of having some bipolar issues. Just expressing my thoughts its the sane thing to do when one feel crapped on maybe nobody else processes their feeling around here shit maybe no body has feeling around hear maybe they just have judgement from a distorted mind set that comes from not wanting to acknowedge the truth about things because it may put them a risk of exposure after all they are supposedly in charge of it all and yet they still have a usb drive that holds realy truth in it and they kept it from the world to continue with their ultimate bullshit denial system that serves then only because the constantly wright or at least they think they are.
… what?
Just want to say thank you for not deleting that comment. It gave me the biggest laugh of the day. I think the Ultimate Warrior reads your blog.
I can’t wait for you to publish a statblock for the “cuocuo” bird.
I think, as the title suggests, I _don’t_ get attrition as a game/scenario goal.
Summing up what I’m taking away from this — in practical terms, good game/scenario design is to get through a series of combat encounters without losing too many hit points, spells, and arrows, so that you retain enough for the BBEG.
While I might find this kind of scenario occasionally interesting (a survival/gauntlet type of scenario), if it were the norm for all scenarios, I would start preferring to spend the evening doing laundry — I might find that missing sock — a hell of a lot more exciting Is there a whole bunch of relatively trivial combats that offer no real risk, no real excitement, no real reward.
I’m baffled that your ultimate point is this is some sort of wonderful positive thing that game designers are screwing up by deemphasizing.
Isn’t it just as valid for microchallenges to be not about _preserving_ resources, but rather _gathering_ resources — clues, items, allies, information, that aid and assist you towards achieving the macrochallenge?
I think you might be getting a little too caught up in the definitions of resource and attrition.
“Narrative” resources (i.e. clues, items, allies, information, etc.) are still resources that can be used/gained/lost/found.
The Macro challenge is dealing with the ghost. The micro challenge is finding the right information to calm down the angry ghost, or the right items to enable you to put down the angry ghost. The micro-er challenge is having
Good game/scenario design is blending, balancing, and linking the micro-challenge with the macro-challenge. The “scenario difficulty” with the individual “encounter difficulty/ies”. The various rewards/penalties of success/failure at any given point. In having appropriate/satisfying consequences (both narrative and “mechanical”) to the player’s actions/decisions.
Is having to use your top level spellslot to escape death a consequence? only if you need that top level spell slot later. Is finding (or failing to find) additional information a consequence? only if it increases your options later. Is calling in that favour an NPC owes you from 3 sessions ago? only if those favours are limited. Is having to fall back and come back the next day a consequence? only if it subsequently impacts anything. Is *gaining* a new ally a successful outcome? not if you had to burn the bridge with a more important NPC to get there.
I don’t think (and nor do i think that angry thinks) that attrition is the single pillar upon which good game/scenario/encounter design rests upon. But every time you handwave away resource management, you remove consequences… and by extension you start removing the stakes.
This is very well summarized. Thanks Germanwench.
“But every time you handwave away resource management, you remove consequences”
Yes, but consequences of what? Poor bookkeeping? Lack of proper inventory control?
I tend to agree with Kent in some respects. If the outcome of a game I’m playing is going to be significantly impacted by whether I kept good enough track of how many torches I bought, or how many pints of water I stocked up, or whether I remembered to mend my armor every day, I really don’t see myself playing that game for long. I find it exhausting enough that I have to manage inventory for my family in real life; I’m sure not going to be excited about devoting my precious recreation time to it.
And narrative resources are not susceptible of game mechanic attrition. Limited favors, later options, gaining allies and burning bridges are all decided by game interaction, not rule mechanics (at least in the games I play). And THOSE are the kinds of attrition that generally can add to the game experience. At least, in my game experience.
Consequences of your choices.
Did you prioritize saving your friend from the dragon and let it eat your horse and now you can’t carry enough water to cross the desert? Did you pack light so you could carry more treasure and now you don’t have enough torches to explore the whole cave? Did you waste your time in combat trying to charm the dragon and all your companions are on death’s door by the time you defeat it and now you have a pile of treasure but don’t have the strength to get it home safely?
That’s not attrition unless you are defining “attrition” as every single consequence of every single event. Losing your horse is like interrogating an enemy; it creates a narrative plot point to move from (“now you can’t ride across the desert so you’re going to have to find another route”, “now you know there is an ambush waiting for you so you’re going to have to find another route”). Reducing all narrative consequence to “attrition” is an unnecessary reduction. Not all choices are resource-driven.
And being in the middle of a thrilling, suspenseful, dramatic dungeon delving, only to have it cut short — not because you were turned back by fearsome monsters, not because your path was blocked by an impenetrable maze, not because you were bested by a dark champion; but rather because you didn’t check off enough torches on your inventory sheet? That does not sound like a game I’m likely to return for. And once again, if that sounds like fun for you and your table, then you’re doing it right. Congrats and have a great game. But attrition doesn’t add anything except tedium to the game experience for some people.
Yes, that’s attrition. So is the loss of time. And TIME does matter in Dungeons & Dragons as written even without ticking time bombs because random encounters arise based on time spent.
What you are doing now is raising a strawman to defend your own preferences. That an attrition-based game boils down to rigorous and careful tracking of elements on a character sheet, which you personally don’t like, is absurdly reductionist and incorrect. It’s okay not to like it. You’ve made your dislike clear. Everyone now knows you personally don’t like it.
Moving on…
As I understand what you are saying about CR: CR does not measure how challenging the creature is to the players, but rather how challenging it is to the player characters, and more importantly of how much of a party’s resources are apt to be used up.
I put it that way to encourage gamers to distinguish between players and player characters.
Brilliant. Cleared up a lot for me and offers a clear way to distinguish between different rules engines – those that do or don’t use attrition, or different ways in which it is achieved. And also levels of attrition preference, which offers a way to discuss play style preferences with players. I personally refer a steepish attrition curve, where likely success/failure of the macrochallenge becomes clear sooner than later, but I have a player who i now suspect prefers a much shallower curve. And now i know how to discuss it. Thanks.