The "Difficulty" Question

[BRC]

So, I've heard this from a few different DMs, where they ask their players some variant on "How hard do you want Combat to be?"

And Players often come back with some variant of "Give us a challenge!" or "Oh, we want high lethality"

As has often been noted, what players usually want is some variant of "We want to always Win, but feel challenged". Few players want to come out and say "Give us easy fights that pose no real challenge".


So, the question is, what is the best way to approach the Combat Difficulty conversation such that you get a good idea of what player's want without them feel like they're asking for you to softball them?

Off the top of my head, I think a good starting point is to break down what sort of things a player might enjoy out of a game and ask them to compare those things.

Rather than say "How hard should combat be, on a scale of 1 to 10", ask them to rate various statements alongside each other, like

"Combat should be a test of our ability to make tactically correct decisions" vs "I care more about being creative in combat than making optimal moves". Frame combat not as a sliding scale of difficulty, but as a test vs an opportunity for expression.

Or, ask how stressful combat should feel. Frame choosing "easy" not as admitting you CAN'T win hard combats, and more about choosing a more relaxing, less stressful experience.

Anybody else have thoughts on how to approach this?

[Zuras]

Normally players aren’t interested in generic “difficulty”, they’re interested in verisimilitude. They want the opponents that should be dangerous to feel dangerous. They don’t need every fight to be dangerous, they just need a few periodically where they feel threatened.

The ideal setup for player satisfaction is when they have a fight where they’re ambushed themselves, or otherwise fight without any advantages, and feel like they barely survive, then later fight the same or similar opponents on the party’s terms, and curb-stomp them.

As long as the players feel like their choices and tactics matter, they will usually be happy.

[PhoenixPhyre]

As a side note, players (at least in my experience) are notoriously lousy at determining actual difficulty. They'll be like "phew, barely scraped through that one" when, from my side, it was an easy victory. Basically, if they had to use any significant resource, it was a "hard fight". And if anyone went to zero, it was a "really really hard fight".

[BRC]

As a side note, players (at least in my experience) are notoriously lousy at determining actual difficulty. They'll be like "phew, barely scraped through that one" when, from my side, it was an easy victory. Basically, if they had to use any significant resource, it was a "hard fight". And if anyone went to zero, it was a "really really hard fight".

This can be a side-effect of D&D's resource-based game-play, where individual encounters are expected to form a gauntlet.


For example, an encounter against a big group of weak enemies may feel trivial if the Spellcaster starts throwing fireballs, or soul grindingly difficult if the spellcaster either decides to save their slots, or has already used them up. Spending resources doesn't usually FEEL like triumphing over difficult odds unless you are backed into a corner and have no choice but to use something you were saving. You rarely know how hard the fight would have been had you NOT used those resources.

[Kurt Kurageous]

I think this is really part of the session zero, "How do you feel about character death?" It's not specific to combat, but implies combat because that's the source of most fatality in the game.

I try to create combats that are significant in purpose to the story. Combat is a form of conflict. Conflict is essential to a story.

Sure the party can win, and will win, but what is the price of victory? Is this fight unavoidable? Is there a better way? Is there another purpose at work?

Don't forget, there are two sides to a fight. If you are dealing with intelligent foes, remember they get a vote, too. If they attacked first, they most have thought they could win. Why did they think so? Or were they on orders to do a suicide mission? What happened to motivate them so highly?

If you are dealing with unintelligent foes, you may often have short and pointless fights, or highly dangerous ones with implacable senseless foes. Oozes that live in a cave that attracts unwitting creatures will become a cave with a lot more oozes.

[Talakeal]

I would assume that one of my posts likely inspired this thread, but I will throw my 2cp in nonetheless.

I find that my players grossly overestimate the difficulty of any encounter.

If they have a "close call" where loss was actually a possibility, it must have been a horribly over-tuned impossibly deadly scenario that they only survived due to luck or brilliance.

What they consider a "balanced" fight is one which is easy enough for them to lose no matter how bad their dice rolls or tactics.


I have always looked for an objective measure of balance, and I generally shoot for "each adventuring day uses up most of the party's resources" leaving them with between 1 and 25% of spell slots, HP, and charged items / abilities remaining, assuming average dice rolls and tactics.

Of course, exactly where this lies depends on party synergy, player experience, and how well their specific builds do against the obstacles in question (for example, undead will be easier for a cleric and a paladin than a mind controller and a rogue in 3.5).

[Zuras]

This can be a side-effect of D&D's resource-based game-play, where individual encounters are expected to form a gauntlet.


For example, an encounter against a big group of weak enemies may feel trivial if the Spellcaster starts throwing fireballs, or soul grindingly difficult if the spellcaster either decides to save their slots, or has already used them up. Spending resources doesn't usually FEEL like triumphing over difficult odds unless you are backed into a corner and have no choice but to use something you were saving. You rarely know how hard the fight would have been had you NOT used those resources.

It’s also just the perspective of controlling a single character versus a small army. GMs get desensitized to how threatening it feels to be one good hit away from being knocked out when it’s your only unit.

[Ionathus]

Echoing some of the other sentiments in here: my players respond positively to having more creative choices and feeling like they're up against a smart opponent.

Actual challenge is secondary to the fight "feeling" good, of it having moments of high drama where a single action changes the course of the battle, and having moments where everyone goes "oh ****, the monster's breath can do 74 damage" and having to think on the fly.

This is just speculation on my part, based on what my players react positively to, but I think when players say they want "challenge," they're often talking about unpredictability. If the outcome of the fight is obvious to everyone, why are we still rolling dice? Nobody wants to slog through fighting low-level bandits if the stakes are 10-15 HP on your level 12 paladin.

At the end of the day, D&D is not a perfectly-tuned and balanced tactical wargame. "Difficulty" in D&D isn't just about HP and modifiers and probability of dying. It's also about the difficulty of analyzing the situation and thinking creatively, and finding a way to tell the best story for yourself and your friends. Finding a way to reward that mental game without veering into Calvinball is a huge balancing act of every combat I run.

[icefractal]

I have always looked for an objective measure of balance, and I generally shoot for "each adventuring day uses up most of the party's resources" leaving them with between 1 and 25% of spell slots, HP, and charged items / abilities remaining, assuming average dice rolls and tactics.

YMMV, but to me it would feel like a desperate situation if we were regularly ending the day with our resources down to 1% or even 10%.

Like, IC, there's not a concept of "encounters per day". The fact that we've already faced five encounters doesn't mean nothing else will attack us that day (nor does the fact that we've only faced one encounter guarantee that anything will). So being unable to take on another encounter means "we're in great danger, we need to retreat ASAP and hope nothing attacks us - and really we should have retreated at least one encounter earlier"

That happens once, and it's just "wow, we had a tough time that day". It happens consistently and it's more like "we're in over our heads, barely managing to keep up with the onslaught, but so far we've survived". Which might be fine if that's the intended tone, but don't be expecting my character to act confident in such a situation unless they're a daredevil.


On the more general question of combat difficulty, my personal feeling would be -
If the average fight is deadly enough that making non-optimal moves* spells defeat, then the difficulty is too high.
Some fights can be that challenging. But to me, if I just wanted pure sweaty** combat I'd play a competitive video game - TTRPG fights should (usually) have room for the RP.

*I'm not talking about ridiculous moves, just stuff like "I'll give them one chance to surrender instead of going straight to CdG", or "I'm going to keep attacking the guy who personally kicked my dog rather than go after this random minion who would be more tactically ideal"
**Sweaty in the sense of intense focus and effort

[Talakeal]

YMMV, but to me it would feel like a desperate situation if we were regularly ending the day with our resources down to 1% or even 10%.

Like, IC, there's not a concept of "encounters per day". The fact that we've already faced five encounters doesn't mean nothing else will attack us that day (nor does the fact that we've only faced one encounter guarantee that anything will). So being unable to take on another encounter means "we're in great danger, we need to retreat ASAP and hope nothing attacks us - and really we should have retreated at least one encounter earlier"

That happens once, and it's just "wow, we had a tough time that day". It happens consistently and it's more like "we're in over our heads, barely managing to keep up with the onslaught, but so far we've survived". Which might be fine if that's the intended tone, but don't be expecting my character to act confident in such a situation unless they're a daredevil.

I posit that anyone who isn't a daredevil wouldn't be an adventurer. Dungeons are full of monsters and treasure! It is the very definition of a high stakes high reward occupation!

That being said, its more to do with the nature of the game as being about resource depletion. If you don't grind the PCs down, there is literally zero challenge or danger.

To use a real life analogy, its more like setting a budget for a business than playing Russian roulette. And, much like a business, if you do happen to go over on occasion, it isn't the end of the world, you just dip into your slush fund or take out a loan (in an RPG this usually means potions, scrolls, and favors from NPCs).

Although I do agree 1-10% is cutting it pretty close, 15-20% is a much more common number.

[Xervous]

Difficulty isn’t as clear cut as raising the number slider on a Bethesda game.

To start this conversation I’d ask the players what their expectations are for: a small group of orcs, one lone town guard, a whole orc encampment, a dragon, an orc assassin ambush. I’d ask what other games the players have sampled and if their expectations are easily summarized with those games.

“So there’s a dragon living in Mt Certain Death. You could go there if you’d like. What do you expect to happen if you do?”

“So there’s a large fortress full of angry orcs, again it’s your choice to go here. What do you expect to happen if you go here?”

There are no wrong answers from the players here. It gives me an opportunity to say “well that would probably lead to X because of how I run things...” which in turn could prompt the players to reply.

Player: “Oh a dragon we can go to? Let’s fight it!”

GM: “Okay everyone dies.”

Player: “What? You said we could go there. Why let us waste time just to die?”

Or

Player: “Let’s financially take over the nearby towns and get some trade going with the nearby dwarf city for some equipment and...”

GM: “Look, I’m not here to play a city and politics simulator. There’s a big world to explore, this system is suited to combat, I’d have offered to run XYZ if we were shooting for more political stuff.”

Player: “Could we build influence with the rulers by solving problems at our pay grade and use the good favor to assemble an expedition to take down the dragon?”

The topic of difficulty isn’t so much a number, it’s a pattern. It’s about knowing the answer to “knock knock”, the meaning of life and everything, or there are 10 kinds of people in this world. It’s about presenting stuff that meets expectations, preferably by setting up your players’ expectations.

[Telok]

Funny. I just say "Opposition will be based on the logical consequences of the setting, the npc's situation & abilities, and the parties actions. Combat is up to you".

The primitive tribals with flint tipped spears don't get more powerful because the party is kitted out with power armor and jet packs and laser cannons, the supermax prison of the chaotic evil trickser god does not have lesser guards or defenses because the party is a bunch of ugly stupid sword weilding face stabbers with no magic or technology. Deal with it.

Held true even in D&D. No problems.

[Easy e]

If you ask a person what they want a few things happen:

1. They don't know how to express what they want
2. They don't know what they want
3. They tell you what they think you want to hear
4. They actually tell you what they want

The problem is, you don't know which of those things has happened until you do something and see how the people react to it. If it is a positive reaction, you can do more of it. If it is a negative reaction, you need to stop doing it immediately.

Of course, don't be a jerk to other people, use your safety tools, and constantly get feedback.

TLDR: Play your game and see what happens.

[icefractal]

To use a real life analogy, its more like setting a budget for a business than playing Russian roulette. And, much like a business, if you do happen to go over on occasion, it isn't the end of the world, you just dip into your slush fund or take out a loan (in an RPG this usually means potions, scrolls, and favors from NPCs).

Maybe it's different in your system, but IME I've never found any of those things to be much of a backup plan.

Like, potions and scrolls can help stretch your resources out, make it take longer before you're in the danger zone. But once you are at "we can't take on another fight", the danger is just as much. Because scrolls and potions mostly suck for in-combat use. And NPC favors? Sure, those are nice, but it's not like you can generally access them from inside a dungeon.

And (IMO obviously) but if the reason we're adventuring is just profit then that's going to:
A) Make me less of a daredevil. Fight until I can barely stand in order to save a bunch of people? Sure. Fight until I can barely stand to get a bit further into a promising ruin? Nah, it'll still be there the next day, or if not then something else will - in fact, lets go buy a herd of sheep first, I'm getting tired of these traps.
B) Make spending resources like potions and scrolls less appealing. If the whole point is to get rich, spending permanent resources is the opposite of that.

[Talakeal]

Maybe it's different in your system, but IME I've never found any of those things to be much of a backup plan.

Like, potions and scrolls can help stretch your resources out, make it take longer before you're in the danger zone. But once you are at "we can't take on another fight", the danger is just as much. Because scrolls and potions mostly suck for in-combat use. And NPC favors? Sure, those are nice, but it's not like you can generally access them from inside a dungeon.

And (IMO obviously) but if the reason we're adventuring is just profit then that's going to:
A) Make me less of a daredevil. Fight until I can barely stand in order to save a bunch of people? Sure. Fight until I can barely stand to get a bit further into a promising ruin? Nah, it'll still be there the next day, or if not then something else will - in fact, lets go buy a herd of sheep first, I'm getting tired of these traps.
B) Make spending resources like potions and scrolls less appealing. If the whole point is to get rich, spending permanent resources is the opposite of that.

I agree, resources are better spent proactively; although in my experience players tend to be too miserly to use them as such.

Typically, you can either summon something or cut a deal with something in the dungeon.

Ideally, you get more out of the dungeon than you put in. If that's not the case, you write this one off and fall back. But as they say, sometimes you have to spend money to make money.


As I have mentioned a lot in previous threads, RPGs are kind of bad because they are based on resource management, but those resources are often gated to time, which is often an unlimited resource. Mechanically, a party that pulls out when they get low will be flat out weaker than one who sticks to it because they will have less gold and XP (and goodwill from the people whom they help). From a narrative perspective, in a world where the first arrow is as likely to be a kill-shot as the 30th, its downright suicidal to go on a whole other expedition that might or might not pay off when you are already here and 1 or 2 rooms away from the big score.

[Mastikator]

Vary it. Some easy, some medium, some hard, some very hard. Sometimes even present challenges where a combat approach is likely to lead to character death but other approaches are available. When I GM I most often let the players choose if there is going to be a fight, few NPCs will attack on sight, and most will accept a surrender if offered. Not that players will ever not murderhobo, but that's their choice

[Reversefigure4]

As a side note, players (at least in my experience) are notoriously lousy at determining actual difficulty. They'll be like "phew, barely scraped through that one" when, from my side, it was an easy victory. Basically, if they had to use any significant resource, it was a "hard fight". And if anyone went to zero, it was a "really really hard fight".

Of course, the GM side has perfect information, which makes it a lot easier to read the difficulty. The GM knows that the players started in the hardest room of the dungeon, that the kobolds rolled a lot of 15+ on their rolls, and that the only thing remaining in the dungeon is an old blind kobold who poses no threat. From your point of view, they've basically won the dungeon at the cost of several HP and a few spells.

From the players point of view, the first room badly injured them, and they only have enough resources left to take on one more fight like that. And who knows what's beyond the red door? All they can see is a little old kobold who might have 10 levels in Monk!

It's just a part of the tradeoff. Everything looks more threatening from the player side of the table, because it's an unknown.

[Quertus]

I mean, I prefer Simulationist CaW "let the dice fall where they may", "If your 1st level characters fight the dragon, you're dead (unless you're my brother, because blessed by Arangee, goddess of luck)". Where the GM has the skills to train the players to be on the same page wrt information gathering, level of detail, amount of telegraphing, etc. Where the players can break to OOC (and the GM will break to OOC) over Gamist concerns like, "So, I could spend the next 4 hours micro-detailing exactly how I verify the reliability of this merchant, but... is that necessary? Can we just say 'I purchase a meal here without being poisoned' without that 4 hour minigame? Or is it worth our time to go through those details?"

I want the level of challenge of the combat to feel appropriate to the information we have and the choices we have made. I don't want Narrative contrivance of "cut-scene awesome" or "to ratchet up the tension", I want the Agency of a world that makes sense.

That said, among the 8 Kinds of Fun, Challenge isn't one I particularly care about. Give me a well-deserved cake walk, or a well deserved loss, over an unrealistic challenge any day.

[False God]

I let my players pick their stats, 3-23, and pick their level, 1-20.

They'll usually show me just by the character they made what kind of game they want.

[Pauly]

I like stupidly dangerous combat. Games like Cyberpunk and Call of Cthulhu do it well.

What I mean by stupidly dangerous is that if you do something stupid you will die.

The difficulty doesn’t come from how dangerous the opponent is, because all opponents are potentially lethal. The difficulty comes from using fire and movement, cover, terrain, knowing when to disengage and knowing when to charge in.

[stoutstien]

So, I've heard this from a few different DMs, where they ask their players some variant on "How hard do you want Combat to be?"

And Players often come back with some variant of "Give us a challenge!" or "Oh, we want high lethality"

As has often been noted, what players usually want is some variant of "We want to always Win, but feel challenged". Few players want to come out and say "Give us easy fights that pose no real challenge".


So, the question is, what is the best way to approach the Combat Difficulty conversation such that you get a good idea of what player's want without them feel like they're asking for you to softball them?

Off the top of my head, I think a good starting point is to break down what sort of things a player might enjoy out of a game and ask them to compare those things.

Rather than say "How hard should combat be, on a scale of 1 to 10", ask them to rate various statements alongside each other, like

"Combat should be a test of our ability to make tactically correct decisions" vs "I care more about being creative in combat than making optimal moves". Frame combat not as a sliding scale of difficulty, but as a test vs an opportunity for expression.

Or, ask how stressful combat should feel. Frame choosing "easy" not as admitting you CAN'T win hard combats, and more about choosing a more relaxing, less stressful experience.

Anybody else have thoughts on how to approach this?

Something I realized over the years is that it's rarely consistent from session the session or even throughout a single meeting. It's not something you can ask at the beginning of a game and forget about. You need to keep a finger on the pulse and periodically check back in and readjust.

I tend to set a theme and play the rest by ear. I build my world's fair and play to 'win' in regards to npc motivations but fate is always slightly in the hero's favor.

[Quertus]

I like stupidly dangerous combat. Games like Cyberpunk and Call of Cthulhu do it well.

What I mean by stupidly dangerous is that if you do something stupid you will die.

The difficulty doesn’t come from how dangerous the opponent is, because all opponents are potentially lethal. The difficulty comes from using fire and movement, cover, terrain, knowing when to disengage and knowing when to charge in.

This is interesting. Would a marine be desperately evaluating cover and terrain when they kick a puppy? Would a pimp be risking his life every time he slaps his blind, crippled hooker? I think not, I think that the system and setting and expected content are pulling a fast one, resulting in scenario design that focuses on a certain style of challenge. I expect that I could write content for those systems that completely lack challenging engagements, or that I could write content for other systems where every encounter risks death to fools.

For instance, long ago, someone posted a thread about “what would an intelligent, immortal undead caster do” when an armored knight, religious nut job, slinking coward, and guy in a bathrobe invaded his lair? Also, the undead has just invented a new multi-target fire Spell they’re just dying to try out (so, Scorching Ray).

Different people gave different responses wrt what counted as intelligent vs what counted as metagaming levels of genre savvy, or even what the “right” answer was.

Me? I said that an intelligent Undead wouldn’t target any of the adventurers. Instead, they would have prepared multiple targets for their spell. For example, a paper machete patch over a weakened ceiling ready to collapse, to damage and trap intruders. A vat of explosive, poisonous liquid, to damage and poison intruders. A flammable material to make smoke to remove vision and eat the oxygen. Etc.

And any idiots who charged in expecting only a stand-up fight will die to the preparations of the immortal undead.

[kyoryu]

How hard should a combat be is an irrelevant question, because people don't agree what a "5" would mean. I mean, I'd suspect most people would say an average combat should be 5 difficult, right? They'll just define that differently.

I'd probably ask a few questions:

"Take an average encounter. Describe to me what that's like, in terms of difficulty. Does anyone die? Go unconscious? How many of your resources do you use up?"

"Okay, so now a hard encounter - like, the hardest one you might find in a typical session. What does that look like?"

"How bad do things have to go for a character to do? How much bad luck, and how many bad choices do they have to make? What about a TPK?"

In my experience most players want "deadly" games that boil down to "if we screw up, WE'LL TOTALLY DIE, but we're so good that we never screw up". I could get more into this, but it seems to be a reaction to almost always winning combats to ramp up the perceived danger of loss to try to maintain some tension.

[Vahnavoi]

You can't say or do anything intelligent about game difficulty without first examining what your players do in a game and what skills they use to do it.

The proxy values people use for measuring difficulty, such as win or death rate or portion of resource consumption, are at best useless without that additional context, and at worst actively misleading.

To understand why, let's talk about randomness for a moment:

It's been shown that, in games of chance where player skill makes no difference, even when players know they are playing such games of chance, they still attribute success and failure to skill in a predictably biased way. Namely, person A will attribute their own random success to their own positive personal qualities, while dismissing their own random failure as result of externalities. However, at the same time, they will attribute random success by others to externalities, while attributing random failure by others to negative personal qualities in them.

Simply: If I win, it's because I'm good; if I lose it's just bad luck; If you win, it's just a fluke; if you lose, it's because you suck.

What does this have to do with roleplaying games? Well, many tabletop roleplaying game are at their core just glorified dice games. Skill has negligible effect, the players will just roll dice, then roll dice some more. Yet the players will have a subjective feeling of difficulty depending on sequence of rolls they made, and how often they won.

Adjusting numbers will do nothing to actual game difficulty in such a game. At best, it creates an illusion of difficulty, a magic trick pulled on the players, hoping they won't notice.

In order for there to be real difficulty to affect, there must be more to a game than rolling dice, and you have to recognize and filter any player feedback on difficulty that is just them reacting to random noise.

The next step up is a game of shifting probabilities, such as a betting game. Here, how hard the game is depends on how hard it is to calculate the best betting strategy or strategies. This kind of mathematical problem scales well, from trivial to gradeschoolers to making a professional mathematician cry. But this has nothing to do with win rates or anything like that! It's possible to have two very different betting games where the hard ceiling for win rates is 95% in both, but in one it takes a minute to figure out in your head, and in another you have to program a computer to simulate the game a few million times to get an approximation because the explicit solution goes way over your head.

And counting probabilities is one class of math problems out of of a huge design space you could use for core of a game. For every different type of problem a game includes, there's an aspect of difficulty that might not influence how often players win or lose at all, but will increase the time and effort it takes them to progress.

[King of Nowhere]

In my experience most players want "deadly" games that boil down to "if we screw up, WE'LL TOTALLY DIE, but we're so good that we never screw up". .

This sums it up nicely.
It's also roughly equivalent to the idea that the players want to have choices and be rewarded for smart playing. This includes the possibility of death or generally other unpleasant things in case of serious mistakes, because if you can play dumb and still win there's no incentive to be smart and no sense of accomplishment.

[Jay R]

First, players aren’t all the same. They don’t all like the same kind of encounter any more than they all like the same ice cream flavor or pizza toppings.

Also, you can’t get the answer by asking the players directly, because what they think changes – and they may not know it themselves.. My rule of thumb, based on my players, is expressed in my Rules for DMs:

3. What the players want today is a quick, easy victory. But what they will want tomorrow is to have brilliantly and valiantly turned the tables to barely survive a deadly encounter where it looked as if they were all about to die.

In practice, this requires an encounter to *look* more difficult than it is. [It also requires players who will think outside the box. To believe that they were brilliant, they have to do something different from what they think other players would do.]

But the crucial fact is this: players are not all the same.

My current group all played AD&D, and more than half of them played original D&D. They grew up in this game with the assumption that characters could and would die, depending on their choices and die rolls. Many newer players assume that we are just telling the story of heroes who are going to win, and even the possibility of not succeeding seems unfair.

So know your current players, and tailor the game to them. But don’t focus only on what they say while playing, or even at the end of the session. As my rule above indicates, your best indicator is how they’re talking about last week’s session.

[kyoryu]

This sums it up nicely.
It's also roughly equivalent to the idea that the players want to have choices and be rewarded for smart playing. This includes the possibility of death or generally other unpleasant things in case of serious mistakes, because if you can play dumb and still win there's no incentive to be smart and no sense of accomplishment.

Yeah that's what's said.

But when you try to get people to put numbers on it, they get awfully cagey. "Yeah, you should die if you play poorly and/or have bad luck". "Well, okay, how much?". "Well, I don't think there's a number."

If you're playing a recreational sports league, I'll give you a number - a given team/participant should probably be winning 33%-66% of the time. I don't think that's necessarily the right number for roleplaying games, to be clear. If a team is winning more than 80% of the time? They probably need to be in a higher league. That starts becoming Not Fun for everyone else.

So, okay, then you ask "how bad do you have to play? How bad does your luck have to be?" And you can put numbers on that, too. "Okay, so if you break up your play into 10% increments, from your worst off night to your best most genius one, and you do the same with your luck, which groups do you have to be in, or below, for death to occur?" And they'll still be cagey. Because I think that many RPG players are used to the "always win" model that linear games encourage, and because player death frankly sucks.

This is one of hte major reasons I really really like to focus on non-death stakes, and get irritated that the default assumption is that loss has to equal death in RPGs. It just makes it harder to really talk about this and make real tension.

(Interestingly I think, when you had a stable of characters like in really old-school D&D, I think PC death worked a lot better. It was losing a soldier in XCOM, not deleting your Skyrim save).

[PhoenixPhyre]

If you're playing a recreational sports league, I'll give you a number - a given team/participant should probably be winning 33%-66% of the time. I don't think that's necessarily the right number for roleplaying games, to be clear. If a team is winning more than 80% of the time? They probably need to be in a higher league. That starts becoming Not Fun for everyone else.

Although that presumes a competitive scenario where "winning" and "losing" is really a big thing.

I've had tables that were very happy with constant cakewalks, where all that matters is what they decided to do, not whether they succeed at doing so. And tables who didn't care much about choosing what they were doing (ie were happy with a very linear campaign) but cared a lot about success/failure chances.

[icefractal]

I think the problem with any reasonable-for-sports win/loss rate is that not only does a sports team not die when they lose a match, they don't even stop playing for that season!

Like, non-death consequences often come up as a panacea, but IME, most of the "you didn't die, but ..." results are still pretty harsh, and spell the loss of whatever over-arching goal the battle was part of.

Say for example -
You tried to sabotage Baron Evil's siege engine and got defeated. But you weren't killed, he just throws you in the dungeon, and then a couple weeks later you get the chance to escape when one of his guard-golems goes berserk and causes chaos in the tower. Well it's still way too late to save the town that was the reason you came here. That siege already happened, they're all dead, fled, or enslaved. That's not any kind of "mixed result", you just straight-up lost.

And sure, nothing wrong with failing sometimes. But if you have, say, a 60% chance to win each fight, and a quest/goal requires five fights on average, then you've got about a 7% chance to succeed at that goal. That's not good odds. You're not going to seem like a competent group.

So you'd need consequences that don't individually stop you from reaching your goals, although they could cumulatively stop you (like say, score during a season). And while there are some contexts where that makes sense, it's tricky to justify in most.

And that's assuming non-death consequences that affect only yourselves and have no lasting effects! Lots of the suggestions I've seen are worse than that, sometimes to the extent that my feeling is "screw that, I'd rather just die if I lose", because they result in a death-spiral / despair-spiral in the longer term.

So (IMO) at the point where fate is in the PCs' favor enough to say things usually work out ok when they lose, that prisons will have a chance to escape, foes won't do anything to take them out of the action for too long, gear will be recovered if needed - then why not just let them win most battles in the first place? That seems less contrived TBH.

[Ionathus]

This is one of hte major reasons I really really like to focus on non-death stakes, and get irritated that the default assumption is that loss has to equal death in RPGs. It just makes it harder to really talk about this and make real tension..

Agreed. Failed combat where there are narrative stakes is a lot more tolerable.

The idea that there should always be "a chance" of death is very risky when you're DMing. Let's imagine, for the sake of argument, that you and your players were all capable of looking at a combat setup and saying "this combat poses an X% chance of the party dying." What % would your players find tolerable?

If they want combat to feel "dangerous" they might pick a double-digits number, say 10%. But then, do they plan to have 10 dangerous combats across the campaign? I'm sure they expect to have double or triple that. And most TPKs are seen as a pretty significant "fail state" for most players (not that TPKs are all that common, especially in 5e). So to mitigate that, the DM might reduce monster strength until objective risk is down to 5%, maybe 3%, maybe even 1%. But what's the point of that, even? Now sure, there's a "chance" of death, but is it meaningful? Did quantifying it make everyone feel better about the "risks" involved in encounter design? I'd be inclined to doubt it.

That's a totally made-up example. But I hope it communicates my point that many RPG players aren't actually looking for real danger. I think many players are more like the audience in a theatrical swordfight or a fire-breathing show or a circus acrobatics act. They are there to do something dangerous and thrilling, but the prospect of something actually going worst-case-scenario wrong does not appeal to them at all. Many players want the illusion of danger. They want to be thrilled by the dangerous monster, they want to be scared like in a thriller or a horror movie or a rollercoaster. They want that little gut-punch of fear when the villain tips their hand and shows that they're a credible threat.

But at the end of the day they know that this is all made up, and so they're not likely to keep watching and clapping if the trick actually gets messed up. It turns out, yeah, I don't actually like watching somebody get horribly burned, even if five minutes ago I was saying "oh my gosh, he could get horribly burned if this went wrong, this is so exciting!" The illusion of danger is a powerful entertainer, but the safeguards have to be there.

That's for your average group, IMO. There are absolutely hardcore players who really want to push the limits and get into actual deadly encounters in RPGs constantly. I haven't really met them at my tables. I have met players who say "oh yes, absolutely, I would be okay with my character dying" and then got to two failed death saves in 5e and very quickly realized they were wrong, they don't actually want to see the firebreather get horribly burned, please turn the safety rails back on.

I think D&D is one of the games that thrives on those safety rails, and introducing narrative stakes for defeat (rather than perma-death for the whole party) is a great way to do that.