Character Weakness

[Steerpike]

What's this? sparkletwist and Steerpike have a (friendly) disagreement about game systems? Preposterous!

Video I posted.

My philosophy on character creation in a nutshell. Thanks.

I don't know why people keep repeating the myth that you need to have bad stats to lean into roleplaying your character flaws. All having no good numbers on your character sheet does is make you suck at the mechanical portion of the game.

It's not just about roleplaying character flaws as personality traits. It's about "tension and its release." It's that having deficiencies at the mechanical part of the game can, as the video puts it, "turn a slog into a white-water rapid chase," and as they put it in the description of "hardcore mode," deficiencies create a feeling that you are "[surviving] one nail-biting episode to the next."

In other words, if your characters are not always uber-competant at combat, magic, etc - if they are mechanically flawed or under-powered - this deficiency creates a particular kind of scenario where you have to play differently for your character to survive and thrive. You can't rely on your stats to bear you through every combat: sometimes you have to run, hide, employ clever or creative tactics to defeat or circumvent enemies that you wouldn't need to resort to if your character were a superpowered badass who could confidently slay most monsters.

It also creates a very different sort of atmosphere for parts of a game. If the PCs are all high-powered and uber-competant and confident they can take all comers, a dungeon may still hold a kind of challenge and danger, sure, but if they're riddled with flaws and weaknesses and are fully aware that the creatures in the dungeon might very well be too much for them to handle in a straight-up fight - if they suck at the mechanical parts of the game - suddenly the dungeon becomes super terrifying, becauser the PCs are vulnerable.

Example from my campaign:

[ic=Character Weakness Yields Suspenseful Fun]Water and sewage rushed through the tunnel, but as the characters prepared to head south, something bubbled and seethed in the water, and the group retreated. Caulis called on its psuedodragon familiar to scout; the creature returned with a report of something large and many-headed wallowing in the sludge. Alabastor suggested a way forwards: he would conjure an illusion of the brickwork to mask their movements, so that the creature would perceive nothing but a blank wall. He hastily wove the illusion and the party hurried along the walkway as stealthily as possible. When Alabastor himself attempted to follow, however, a brick crumbled beneath his tread, plopping into the water.

The thing in the water stirred. At first all that could be seen was a monstrous claw, but gradually the hybrid abomination emerged from the ooze: a huge, scaled horror with the body of an albino crocodile and three heads, crocodilian, eel, and gigantic rat, with a lamprey for a tail. A pair of monstrous pincers protruded form its flanks. Sister recognized the monster as a putrecampus, a "Chimera of the Sewers." Sensing movement, the tunnel-monster breathed forth a plume of miasmatic gas from its crocodile-head, catching both Caulis and Alabastor. The homunculus ignored the fume, but Alabastor collapsed, spasming, his face turning black as poison wracked his body. Desperately, Caulis conjured a phantasmal force, creating for the chimera the delusion that the roof had collapsed on it. The other party-members watched as the thing writhed and splashed as if in pain, blood spurting from its several mouths. Hurriedly, Sister, Armand, and Garvin dragged Alabastor to safety down a side-passage, Sister restoring him with a cure spell. Meanwhile the putrecampus shook off the illusion and charged, trying to follow the party down the passage; only its great size prevented it from reaching them, and they ducked into the adjoining grate control room.

Penned now in the second grate control room, the party caught their breath, Alabastor still shaking and wheezing, vomiting blood as his body struggled to expel the toxic gas. [/ic]

This was a huge high point in this session and gets remembered fondly at my table.

In this situation, the party suspected (quite rightly) they were outmatched, and Alabastor's terrible Stealth and Constitution saving throw (his sucking mechanically, in other words) led him to be attacked and subdued. This was insanely tense, first as everyone's trying to sneak past this horrible thing, then the blind panic of having to retrieve Alabastor, with wizards desperately combing through spell cards to find anything to distract the creature, then finding ways of escaping it using the terrain.

Now, if they were all super competant and confident they could destroy the putrecampus, this would have been a 45 minute fight. That could have been cool, but this 15 minute mad scramble chase left my PCs shaking with giddy relief. Alabastor's player told me he was sure Alabastor was going to die, and there was a real feeling of having barely escaped, a feeling which the mechanical asymmetry of monsters vs PCs  and specific character weakness created.

To use a non-tabletop example that's a bit hyperbolic, this is the difference between something like Doom and Alien: Isolation. Superficially these have a lot of similarities - they're both under the science fiction-horror umbrella to some extent, and in both you move through a series of high-tech spaces encountering horrible monsters and trying to accomplish a series of objectives.

Doom clip. Hyper-competant character races through the level mowing down every demon he finds with a vast array of weapons. Flashy and frenetic, sure - this is still a fun experience. Power-fantasy has its place, I'm not saying otherwise, but it's a particular type of experience. The game is about visceral joy and has its challenges, but I wouldn't describe it as "tense" or "suspenseful."

Alien: Isolation clip. Clip 2. Clip 3. Now you've got weapons in Alien: Isolation, but they're only really useful against minor enemies and as a means of temporarily fighting off the xenomorph. The game is a very tense cat-and-mouse where you are constantly on edge creeping through areas, using your tools and ingenuity to keep yourself alive. If the xenomorph catches you, you're screwed.

In both games, there's a kind of asymmetry of power/mechanical-competance. In Doom it's on the side of the player, clearly: the monsters all individually suck compared to the player, and the fun of the game depends on this, as the player tears them apart with bullets, chainsaws, electricity, etc. If the power balance dramatically shifted, the game would cease to be fun in the same way. In Alien: Isolation, the asymmetry is on the side of the monsters, and a very different type of fun is created, which would be totally lost if the player could reliably kill the aliens (in other words, if they sucked less mechanically).

EDIT: One could try to play Doom with the intense paranoia of Alien: Isolation, but it would be silly and wouldn't really feel all that scary or tense, more absurd. Likewise one could try to play Alien: Isolation as a gun-slinging badass, but you'll just die immediately and repeatedly, which just isn't fun. So the level of mechanical suckitude - character weakness, or power asymmetry - crafts a particular type of player behaviour, and creates a specific atmosphere and affect.

[sparkletwist]

It's not just about roleplaying character flaws as personality traits. It's about "tension and its release." It's that having deficiencies at the mechanical part of the game can, as the video puts it, "turn a slog into a white-water rapid chase," and as they put it in the description of "hardcore mode," deficiencies create a feeling that you are "[surviving] one nail-biting episode to the next."

I totally agree with the idea that sometimes characters not being competent in some areas can add tension and excitement, or just make for interesting roleplay hooks. However, that's not really what I was talking about, and I think a couple of important points have gotten lost.

The entire second half of the video was about, and I quote, "when you roll up a stat array that's absolute horse shit." It wasn't talking about characters that had deficiencies, but rather about characters who were just deficient, that is, a character that just plain isn't (mechanically) good at much of anything. This is an important distinction to make. It's also talking about one player rolling up one terrible stat array, not a whole group of players who are in over their heads and a game (or a game session, at least) themed around that feeling. This is another important distinction to make.

If a character is good at some things and isn't good at other things, that makes a well-rounded character, and I'm totally on board that playing that character's flaws can be just as much fun as excelling. To use an example from our games, Dagny is very smart and capable and will have plenty of chances to show off when it comes to casting spells, knowing strange lore, or (oddly enough) wrestling people, and playing her having a chance to excel is always good fun. On the other hand, her Charisma is lousy and I've not bothered to invest any skill points into Diplomacy, so she's rude and caustic and awkward-- and playing out her general social ineptitude is also quite a bit of fun.

This isn't that. At least, I don't think so. If your stat array is "absolute horse shit," then you're almost never going to have a chance to shine, aside from the DM throwing you a bone, and then that's just obvious DM fiat. And if you don't get that and instead you just suck most of the time, yes, you could call that "hardcore mode" or whatever, but it could just as easily (perhaps even more easily) get extremely boring and frustrating, because of the way D&D works. You make a skill check, and you fail. In combat, you roll the dice and you miss, and then you have to sit and wait for your turn, again. You can roleplay, sure, but you can do that anyway... and a lot of the time other players are probably going to be busy planning strategy with their competent characters to be able to play along too much. So it's back to being bored and frustrated.

I've played bumbling or unlikely hero types before, but they were always at least somewhat mechanically competent, so when the dice came down, they were at least able to contribute mechanically. Instead, I specialized in somewhat offbeat skills (who knows when a few ranks in Profession(Maid) will come in handy?!) and I tended to roleplay what they were doing as more based on fear or luck rather than acting decisively, but the character didn't (and didn't have to) mechanically suck in order to do any of this... and ultimately was able to add more to the game because they didn't.

You can't rely on your stats to bear you through every combat: sometimes you have to run, hide, employ clever or creative tactics to defeat or circumvent enemies that you wouldn't need to resort to if your character were a superpowered badass who could confidently slay most monsters.

I don't think this is really relevant because if you can just rely on your stats to get you through every combat without any thought, i.e., walk up and roll an attack and hit and win, or something similarly monotonous and pointless... then that's more the DM's fault for just not making a game that is very interesting. Even competent characters should have to run, hide, or employ clever or creative tactics, because the game really isn't very challenging or interesting if these situations don't come up. So I'm not really sure that this is an argument in favor of any value to be had in playing mechanically suboptimal characters.

If the PCs are all high-powered and uber-competant and confident they can take all comers, a dungeon may still hold a kind of challenge and danger, sure, but if they're riddled with flaws and weaknesses and are fully aware that the creatures in the dungeon might very well be too much for them to handle in a straight-up fight - if they suck at the mechanical parts of the game - suddenly the dungeon becomes super terrifying, becauser the PCs are vulnerable.

This connects directly to my second point above. While I understand where you were going with your Doom vs. Alien: Isolation comparision, a single weak character isn't the same thing, and that's more what the video and my point were about. You're talking about a game that's more survival horror and less heroic action, and that's a valid choice, but in a case where you've got one character who is just mechanically terrible, the rest of the party may well be perfectly competent in battle, and be happy to take on at least some of the threats head on. In that situation, there's one character that is more or less just dead weight. It's not about atmosphere at that point-- everyone is playing Doom, but the rest of the party is toting rocket launchers, double barrel shotguns, and the BFG 9000, and you're trying to make do with the pistol and 3 bullets. That doesn't really seem like it enriches roleplay in any way.

[Steerpike]

I'll definitely agree that if everyone else in the party is very competent and you're the one person with a character who is totally useless, then depending on the group, that can definitely be less fun. But I don't think most of the video was really about that, even the second half - I think the "absolute horseshit" thing is mostly a bit of hyperbole. For me the the video is essentially just saying, don't get so hung up on a character being good or the best or having the right numbers, because that's not where a lot of the fun in the game lies: screwing up and getting in over your head and barely escaping dire situations and dying gloriously are all also fun, and often more fun than competantly wiping out the enemy in combat, and characters with major mechanical weaknesses are more likely to get into those kind of sticky situations than characters who are uniformly mechanically good. Characters sucking and screwing up creates types of fun that are distinct from characters succeeding.

I don't think this is really relevant because if you can just rely on your stats to get you through every combat without any thought, i.e., walk up and roll an attack and hit and win, or something similarly monotonous and pointless... then that's more the DM's fault for just not making a game that is very interesting. Even competent characters should have to run, hide, or employ clever or creative tactics, because the game really isn't very challenging or interesting if these situations don't come up. So I'm not really sure that this is an argument in favor of any value to be had in playing mechanically suboptimal characters.

I guess my point here is that playing characters who are in some sense mechanically suboptimal, i.e. not always confidently fit for the task at hand, encourages creative play and often the most creative play happens outside of typical combat situations.

EDIT: In other words - the worse characters' stats are, the more they'll have to think and act outside-the-box to get things done, because doing things the expected or obvious way becomes less likely to yield results. This can go too far and can push into un-fun territory, but it seems to me that games that deliberately create characters who aren't especially hyper-competant often end up fostering a lot of creativity and problem-solving and crazy plans - this is one of the reasons I like survival horror, as you note. On the flipside, games with very competant, very balanced PCs, like say 4th edition D&D, are probably inclined to have less of this. Would you agree with any of this? I think this is closely linked with how much emphasis a game puts on combat being the core activity. The more it's assumed that mostly what characters will be doing is directly fighting things, generally the more wargame-like, balanced, and rules-heavy the game becomes, and the less fun being mechanically suboptimal becomes (because it just feels like having a worse set of numbers in a game about winning with numbers).

[sparkletwist]

For me the the video is essentially just saying, don't get so hung up on a character being good or the best or having the right numbers, because that's not where a lot of the fun in the game lies: screwing up and getting in over your head and barely escaping dire situations and dying gloriously are all also fun, and often more fun than competantly wiping out the enemy in combat, and characters with major mechanical weaknesses are more likely to get into those kind of sticky situations than characters who are uniformly mechanically good. Characters sucking and screwing up creates types of fun that are distinct from characters succeeding.

I took the video more literally. It talked about awful stats, and characters with a death wish, and blundering your way through, so I figured that it was talking about playing characters who were just terrible. But anyway, I agree that a lot of the fun in a game can come from all that stuff you mentioned (except maybe for dying gloriously, but that's a whole different debate) but I don't agree that having suboptimal characters is really necessary or even all that desirable to create that sort of atmosphere.

I mean, an optimized point buy character is almost always going to have at least one dump stat, and in that low stat there's a prime opportunity for a mechanically good character to nonetheless have some great hooks to be drawn into troublesome situations. On the other hand, I've played in and run plenty of Fate (and Fate-like) games where min-maxing isn't quite as much of a thing and the characters were all pretty mechanically competent in terms of their skills, but they nonetheless got into all kinds of amusing hijinks due to compels on their aspects; that's basically the point of Fate, to some extent. In both of these cases, the characters would generally be classified as "mechanically good" but it in no way hindered their ability to get into entertaining trouble when the situation warranted it.

In other words - the worse characters' stats are, the more they'll have to think and act outside-the-box to get things done, because doing things the expected or obvious way becomes less likely to yield results. This can go too far and can push into un-fun territory, but it seems to me that games that deliberately create characters who aren't especially hyper-competant often end up fostering a lot of creativity and problem-solving and crazy plans - this is one of the reasons I like survival horror, as you note. On the flipside, games with very competant, very balanced PCs, like say 4th edition D&D, are probably inclined to have less of this. Would you agree with any of this?

If your character is not good, then you have, as you astutely put it, "a worse set of numbers in a game about winning with numbers," and, yes, then you have to think of creative and unconventional solutions in order to scrounge up a bonus, since just rushing into the situation and rolling the dice probably won't be enough. However, characters that do have good stats are just as capable of employing creativity and problem-solving and crazy plans-- and they'll probably be able to put them to better use because they have more to work with!

So, maybe it's just me, but I've usually found playing a competent character to complement that play style, not hinder it. Getting into trouble and mishaps is fun, and thinking up a wacky scheme and carrying it out is great, but, at the end of the day, after going through that trouble, and getting narratively invested from a player standpoint-- well, at that point, succeeding is probably going to be quite a bit more fun than failing, because ultimately players want to achieve their narrative goals. And that's where having better numbers is just plain better, because at some point it's probably going to come down to bigger numbers winning and smaller numbers losing.

[Steerpike]

I think we basically agree on most of this, actually. I think you're mostly worried about imbalances between players, and generally wanting players to succeed. While I do agree that power imbalance between players can occasionally be a problem, I don't see it as a big one most of the time, and I think the fun gleaned from rolling for stats and occasionally getting few unexpected terrible results outweighs the downsides, for me. That said, for my players these days, I generally give them the option to roll or point buy according to their own taste. But the last 3 PCs I made myself in friends' games, I've always rolled. I'm currently super-pumped to be playing a pox-ridden longbowman with arms like tree-trunks but a Constitution absolutely ravaged by alcoholism and multiple rounds of disease.

Where I think we might differ more fundamentally is that I really, really enjoy games where all of the characters are, in a sense, sub-optimal: I like games (both running and playing) where the deck is really stacked against characters, not just in the sense that they have low stats, but that the power balance itself is more asymmetric. This is why I like very lethal games, unbalanced monster encounters, negadungeons, horrible poisons/diseases/spells wreaking havoc with PCs without prior negotiation, etc, and I know you genuinely and quite understandably don't like a lot of those things most of the time, and I don't think you're wrong, I think this is a preference thing.

For me, it's just that if the base-line assumption is that PCs are basically competent and can expect to be more-or-less up to the challenges posed to them without a ton of outside-the-box creative thinking, I get less interested as a player and to a certain extent as a DM. I don't think it's impossible to get into fun scrapes or create tension with competent characters, it's just that the more mechanically optimized players become, the easier it feels to coast by on "standard dungeoneering," and the less suspense I feel. So in a sense I think I disagree to a certain extent with this part of your argument:

However, characters that do have good stats are just as capable of employing creativity and problem-solving and crazy plans-- and they'll probably be able to put them to better use because they have more to work with!

It's not that players with good stats are incapable of coming up with creative, off-the-wall plans, it's that they won't need to rely on them as much. If you're a very competent fighter with a ton of equipment and HP facing down a large group of goblins, you might need a few tactics to win the day beyond "I attack," but you can probably take them in a fairly conventional fight unless they're truly a totally unmanageable horde (in which case you're effectively "sub-optimal" in this situation). But if you're a mechanically sub-optimal character with fewer HP and fewer magical goodies to begin with, that same group of goblins can suddenly spell death, so now you need to break out the caltrops and flaming oil and lure them into a chokehold and use the dungeon's traps against them or form a temporary alliance with the kobold neighbors or lure the ankheg in the nearby tunnels to the goblins or what-have-you.

Like, sure, the hyper-competent fighter could theoretically do that stuff, but why bother when he's an action surge and a few combat maneuvers and maybe a healing potion or two away from victory in a fair fight? In other words, the more the deck is stacked against characters, the more they're forced to come up with clever solutions to have a hope in hell of surviving. The less it's stacked against them, the more they can rely on their natural competence - you need to be just hoping that players feel like coming up with crazy plans for their own sake, without actually really having to in order to succeed.

You note that competent characters are going to be better at executing the crazy plans, but at least when I'm DMing, a lot of the crazy plans don't require that many rolls at all. Maybe "outside-the-box thinking" is the wrong word: let's call it "off-the-sheet thinking" (apart from a few inventory items or utility spells, at least). A lot of the stuff I'm talking about doesn't depend on a lot of checks at all. I get the feeling that kind of play probably isn't your cup of tea, and that's fine... but it is mine!

EDIT: This is in some ways similar to what is sometimes called an "OSR-style challenge." As Goblin Punch puts it:

These are obstacles that meet the following requirements:

    - No obvious solution.  (Straight combat is always obvious.)
    - Many possible solutions.
    - Solvable via common sense (as opposed to system mastery).
    - No special tools required (no unique spells, no plot McGuffins at the bottom of a dungeon).
    - Not solvable by a specific class or ability.

Essentially, mechanically sub-optimal characters will find more challenges become closer to this type of challenge, because more competent characters can rely on combat, their special tools, or specific class abilities to carry the day.

[sparkletwist]

For me, it's just that if the base-line assumption is that PCs are basically competent and can expect to be more-or-less up to the challenges posed to them without a ton of outside-the-box creative thinking, I get less interested as a player and to a certain extent as a DM.

What about PCs that are basically competent but are facing superior challenges and as such can't expect to get by without unconventional and creative solutions? As I mentioned above, I think that if the players can always just coast by without any sort of creative thinking or inventive tactics, then it's more a failure of the DM to create any interesting challenges.

On a numeric level, you could argue that there isn't a whole lot of difference between the situation where you have competent characters who are in over their heads and facing superior challenges, and the situation where you have suboptimal characters facing level-appropriate challenges that are too difficult for them because they're suboptimal. And when you're running the numbers for that one specific encounter, that's not really wrong. However, I don't think it plays out the same way at all, because in the first case the characters are still level-appropriate for their lower level, which means that they're still competent adventurers at that level, and they're able to interact with the rest of the setting in a level-appropriate way. This one challenge you can't handle doesn't make you incapable at everything else. That can help immensely when it comes to solving problems, which is what I meant by "more to work with." For example!

But if you're a mechanically sub-optimal character with fewer HP and fewer magical goodies to begin with, that same group of goblins can suddenly spell death, so now you need to break out the caltrops and flaming oil and lure them into a chokehold and use the dungeon's traps against them or form a temporary alliance with the kobold neighbors or lure the ankheg in the nearby tunnels to the goblins or what-have-you.

If you're a character who is just plain not that good at things, you can't fight the goblins, but you also probably won't have the attack bonus needed to throw the flaming oil far enough or the Diplomacy check needed to form an alliance with the kobolds or the Intimidate check needed to lure the ankheg or the Disable Device check needed to use the dungeon's traps against them or... whatever. You're just not up to the basic adventuring tasks. On the other hand, the competent characters who are facing the truly frightening horde of goblins can't take them on in a straight fight, but they still have all of their skills and abilities to fall back on and will probably be able to pull off at least some that other stuff.

Essentially, mechanically sub-optimal characters will find more challenges become closer to this type of challenge, because more competent characters can rely on combat, their special tools, or specific class abilities to carry the day.

Any solution that doesn't require any real input from the system is essentially available to everyone, no matter what their stats are, because the mechanical aspects of the system don't even matter. I agree with your conclusion that characters without any other good options will make more use of these unconventional solutions, but that's only because they cannot win within the framework of the rules. The whole thing essentially circumvents the entire task resolution system and makes it impossible to analyze or quantify because it's all based on whatever the DM feels like letting players get away with-- and if the players and the DM have vastly different opinions what a "common sense" solution is, then there's probably going to be trouble. It's not my cup of tea, no, but more than that, it seems to subvert the entire point of having a game with rules.

[Ghostman]

So in a hypothetical RPG about baseball players, a team of mechanically suboptimal PCs resorting to beating up the rival team's star pitcher before an upcoming match would count as off-the-sheet thinking because the game system doesn't have any rules for personal combat?

I am of the opinion that if you want PCs to primarily engage in non-combative solutions, then you ought to use a rules system that provides mechanical support for such solutions. If combat is intended to be so dangerous as to be avoided when possible, then you probably shouldn't to bother with a very in depth combat subsystem -  a simple lightweight way to handle it should be enough as long as the probabilities are biased against PCs. You'd instead go for more rules that deal with avoiding the monsters, disengaging and fleeing and hiding. Also rules for things like diversionary tactics, gadgetry and demolitions. Because a RPG system should focus on things the game is about, not on things it isn't about.

[Steerpike]

What about PCs that are basically competent but are facing superior challenges and as such can't expect to get by without unconventional and creative solutions? As I mentioned above, I think that if the players can always just coast by without any sort of creative thinking or inventive tactics, then it's more a failure of the DM to create any interesting challenges.

I was thinking that this could be a source of our disagremeent here, which is that I think you're seeing competence as basically objective, and I'm seeing it as very relative. Like for me if the DM is consistently making encounters that demand creative thinking or else the PCs will get smushed, the PCs are in effect not "mechanically competent" enough to tackle those challenges.

On a numeric level, you could argue that there isn't a whole lot of difference between the situation where you have competent characters who are in over their heads and facing superior challenges, and the situation where you have suboptimal characters facing level-appropriate challenges that are too difficult for them because they're suboptimal.

That's pretty much exactly my position. Like competence for me isn't a quality inherent in an individual, it's the ability to meet a certain challenge. A 1st level character might be a competent goblin-slayer but a totally incompetant dragon-slayer. A bad ability roll is just another layer of incompetence, a handicap to the character rather than an added challenge from the DM, but they're sort of the same thing, really, just putting weights on different ends of the scale.

However, I don't think it plays out the same way at all, because in the first case the characters are still level-appropriate for their lower level, which means that they're still competent adventurers at that level, and they're able to interact with the rest of the setting in a level-appropriate way.

Increasingly, I don't want level-appropriate to ever really be in the vocabulary at all. When I play I generally want the feeling of being at least a bit in over my head. Incidentally I am not always great at this as a DM. it works sometimes, but often I really underestimate my PCs. When I get it right though, it's usually been really fun (as in the example I posted).

If you're a character who is just plain not that good at things, you can't fight the goblins, but you also probably won't have the attack bonus needed to throw the flaming oil far enough or the Diplomacy check needed to form an alliance with the kobolds or the Intimidate check needed to lure the ankheg or the Disable Device check needed to use the dungeon's traps against them or... whatever. You're just not up to the basic adventuring tasks.

I wouldn't necessarily call for a roll for most of those things. Lighting pre-spilled oil on fire with a torch wouldn't take a roll. I wouldn't call for an intimidate for the Ankheg, either - it sees you, it attacks, you run, no roll necessary. I don't see why the trap would require a disable device - I mean, for example, luring the goblins into a crushing room or a poison gas corridor or whatnot, not literally jury-rigging traps like computer programs to specifically attack the goblins. I might call for Persuasion for the kobolds, granted; on the other hand, if PCs offer the kobolds something valuable, I'd at the very least give them advantage on the roll or make the DC lower, possibly even forgo it altogether (in systems without social skills this is a moot point, I'd just roleplay the kobolds).

Any solution that doesn't require any real input from the system is essentially available to everyone, no matter what their stats are, because the mechanical aspects of the system don't even matter.

Agreed.

I agree with your conclusion that characters without any other good options will make more use of these unconventional solutions

Ok good!

...but that's only because they cannot win within the framework of the rules.

Or it would be risky and difficult for them to do so. Like, a mechanically sub-optimal character can still attempt things mechanically. It's just they're taking on a greater risk by doing so, they're hoping the dice will help them through it, but the dice aren't friendly to them if their stats aren't great. I do think this is a sliding scale. I don't want PCs who are limbless torsos with 0 skills who are absolutely and totally incompetent or something. I do actually want some competance and ability, and for characters to have a degree of specialization and skill, abstracted mechanically. I am not saying have 0 rolls and have totally incompetant characters all the time in every way. But I think the game becomes more interesting when relying on the character sheet is frequently downplayed over non-crunchy problem-solving, and where the feeling of being in over your head creates tension and spurs creativity. I like that more than mathematically mapping out a good character build and then putting it into play in a series of statistical contests, i.e. 4th edition combat for example.

The whole thing essentially circumvents the entire task resolution system and makes it impossible to analyze or quantify...

Correct, yeah. I am not in favour of like 0 skill checks or never rolling dice or something, and I think it's important that there at least be theoretical mechanical options (fight the goblins straight-up or whatnot). But if you ask me which is more fun - doing, say, a 4th edition skill challenge minigame where you're keeping track of multiple successes and failures and rolling lots of dice where everything needs to be quantified as having such-and-such complexity and all that stuff, or, on the other hand, tackling one of the OSR challenges on that list, say, "The door only opens when sunlight shines on it, even a tiny amount. The door is on the second floor of the dungeon. Maybe try mirrors?" - I want the second.

it's all based on whatever the DM feels like letting players get away with-- and if the players and the DM have vastly different opinions what a "common sense" solution is, then there's probably going to be trouble.

You're right that the style of play I'm describing involves less interaction with the rules, and that it requires a skilled DM who is good at making judgments and communicating carefully with players to make the shared mental reality they're imagining as consistent as possible. It is not a DM-proof way of playing at all.

it seems to subvert the entire point of having a game with rules.

Well no, I do still want some rolls - not everything needs to be off-the-sheet - and anyway, you need ways of adjudicating what happens when the monster shows up and wants to snack on players and they haven't planned well. But you're right that I am increasingly disenchanted with baroque Pathfinder-level rules granularity; this is a notable shift from how I used to feel say 3-4 years ago.

So in a hypothetical RPG about baseball players, a team of mechanically suboptimal PCs resorting to beating up the rival team's star pitcher before an upcoming match would count as off-the-sheet thinking because the game system doesn't have any rules for personal combat?

Good analogy in some ways. The thing is, I don't think D&D is inherently an RPG about killing monsters necessarily, at least not in all forms. Killing monsters is just a part of the game, maybe even a kind of last resort in some ways. There's a reason early D&D gave XP for treasure rather than monster-killing.

I am of the opinion that if you want PCs to primarily engage in non-combative solutions, then you ought to use a rules system that provides mechanical support for such solutions. If combat is intended to be so dangerous as to be avoided when possible, then you probably shouldn't to bother with a very in depth combat subsystem -  a simple lightweight way to handle it should be enough as long as the probabilities are biased against PCs.

This is pretty much where I'm at in terms of combat. These days I much prefer rules-light-ish D&D - 5th edition at most, but even more like Lamentations of the Flame Princess - as opposed to rules-heavy D&D like Pathfinder, 4th edition, etc.

I don't necessarily agree that all non-combat solutions always need mechanical support, though. In some instances I could see that being the case, including having rules for stealth and fleeing and stuff, so I'm in agreement there. But there's no way to mechanize critical thinking, right? That's a kind of non-system-mastery player skill, essentially.

[sparkletwist]

Like for me if the DM is consistently making encounters that demand creative thinking or else the PCs will get smushed, the PCs are in effect not "mechanically competent" enough to tackle those challenges.

Ok, well, within reason, I like that style of play, too. I didn't (and I still don't) think that's what the video was really about, but it seems like this has conversation has kind of meandered, anyway.

Increasingly, I don't want level-appropriate to ever really be in the vocabulary at all. When I play I generally want the feeling of being at least a bit in over my head.

That might have been the wrong word to use. I meant that the character isn't going to feel out of their league when it comes to 'routine' tasks-- I like the feeling of a competent local badass taking on the demon lord much more than I like the feeling of a complete loser blundering his way through a battle with an orc, even if the "number gap" might be the same in both cases.

I do actually want some competance and ability, and for characters to have a degree of specialization and skill, abstracted mechanically. I am not saying have 0 rolls and have totally incompetant characters all the time in every way. But I think the game becomes more interesting when relying on the character sheet is frequently downplayed over non-crunchy problem-solving, and where the feeling of being in over your head creates tension and spurs creativity.

That makes sense, and, along the same lines, I hope you realize I'm not saying that the game should play like a video game (or a game of 4th edition with a crappy DM) where it's nothing but pre-defined mechanics leading to die rolls.

On the other hand, I do want the game's mechanics to support the sort of tasks that characters are going to be expected to do, because that's what the rules are for. Not relying on numbers written on the character sheet in favor of roleplayed problem solving can be fun, but it also elides the need for a system at all with all of the reductions in objectivity, consistency, and balance that doing that introduces. As such, I think the game works better if it is something that is done sparingly. If a lot of situations that aren't covered by the rules are coming up, then you probably should have better rules. Unless you're just going extremely rules light, I guess, but we're talking about games like D&D.

The sort of approach I prefer is that even if the game is rules light, there are still rules that express a set of conventions for how the problem solving should play out and interact with the system, even if there aren't any articulated rules for doing any specific task like a system like Pathfinder would have. For example, if there's a challenge in a Fate game with a very high difficulty (like, 8 or more) then that challenge can realistically only be overcome by coming up with interesting ways to gather up a few advantageous aspects and then stacking the bonuses. To me that's a fun merging of creative problem solving and mechanics, as players come up with the different sorts of advantages they might want to create, or declarations they could make, or whatever, which is largely non-mechanical... but it ultimately does interact with the system and adds a definite +2 bonus for each aspect you're able to invoke. There are actual rules, and I think it wouldn't be nearly as enjoyable (or fair, or coherent) without the presence of those actual rules.

But if you ask me which is more fun - doing, say, a 4th edition skill challenge minigame where you're keeping track of multiple successes and failures and rolling lots of dice where everything needs to be quantified as having such-and-such complexity and all that stuff, or, on the other hand, tackling one of the OSR challenges on that list, say, "The door only opens when sunlight shines on it, even a tiny amount. The door is on the second floor of the dungeon. Maybe try mirrors?" - I want the second.

I mean, me too. That example is so lopsided it's practically a strawman, given how broken and awful the 4e skill challenge system was, combined with how completely dry and awful the examples of the skill challenges in the book ended up being.

I am increasingly disenchanted with baroque Pathfinder-level rules granularity

Me too, of course. Fate and other various systems based around the same general framework as Fate have been my preference for a while now.

On that note, we were going to play Fate at some point... whatever happened with that? I still want to!

Do I have to bribe you by running a LotFP game or something? :grin:      

[Steerpike]

Ok, well, within reason, I like that style of play, too. I didn't (and I still don't) think that's what the video was really about, but it seems like this has conversation has kind of meandered, anyway.

Yeah, fair, I'm definitely expanding on why I happen to like less-than-optimized characters, as opposed to just the video's perspective.

I meant that the character isn't going to feel out of their league when it comes to 'routine' tasks-- I like the feeling of a competent local badass taking on the demon lord much more than I like the feeling of a complete loser blundering his way through a battle with an orc, even if the "number gap" might be the same in both cases.

Right, I see what you're saying. I do think it can be fun to play a character who even sucks at the routine stuff - as you said with Dagny, she's a terrible negotiator but a good inventor/wizard, for instance. This feels very much like a matter of personal taste.

On the other hand, I do want the game's mechanics to support the sort of tasks that characters are going to be expected to do, because that's what the rules are for. Not relying on numbers written on the character sheet in favor of roleplayed problem solving can be fun, but it also elides the need for a system at all with all of the reductions in objectivity, consistency, and balance that doing that introduces. As such, I think the game works better if it is something that is done sparingly. If a lot of situations that aren't covered by the rules are coming up, then you probably should have better rules. Unless you're just going extremely rules light, I guess, but we're talking about games like D&D.

I think this comes down to a difference in philosophy as to what the rules are for. I see the rules as being there as a guideline for the DM to make rulings to adjuciate the world, and those guidelines become necessary when someone's performing some sort of action with a reasonable chance of failure. So, it's important that the rules are there for the DM to help make those sorts of calls, because theoretically the PCs could probably attempt to muscle their way through encounters. But because I want to generally discourage that sort of approach, the rules are important as a kind of bulwark against that hack and slash strategy- they have to be there, and be detailed enough, in order for the PCs to actually feel discouraged from taking the most obvious approach.

For example, if there's a challenge in a Fate game with a very high difficulty (like, 8 or more) then that challenge can realistically only be overcome by coming up with interesting ways to gather up a few advantageous aspects and then stacking the bonuses. To me that's a fun merging of creative problem solving and mechanics, as players come up with the different sorts of advantages they might want to create, or declarations they could make, or whatever, which is largely non-mechanical... but it ultimately does interact with the system and adds a definite +2 bonus for each aspect you're able to invoke. There are actual rules, and I think it wouldn't be nearly as enjoyable (or fair, or coherent) without the presence of those actual rules.

This isn't that far off how a lot of my games run. It just depends on the types of challenges being created. Some of those I make do involve rolls or could involve plenty of rolling, but there might also be non-rolling solutions.

Like, for example... in an early dungeon in my current face to face campaign, there is a room full of this horrible infectious, fast-growing lichen that colonizes characters' bodies, kills them, and turns them into monstrous undead. It's incredibly dangerous (very high DC to avoid contamination, and almost inevitably fatal without extremely fast magical aid), and the characters know this going in (so it's not a "gotcha" effect). There are shambolic lichen-encrusted undead throughout the room which are extremely slow but really, really scary if they touch you, because they also spread the lichen. The characters need to get a tablet on the far side of the room.

There are lots of ways to do this. My characters ended up being very inventive with spells, freezing the lichen in patches (no rolling) and slowly making their way through, while fending off the slow-but-very-dangerous lichen-shamblers (so, there was some limited combat).

But there was also an almost entirely roll-free option, so to speak, which was to use a couple of weird machines in another part of the dungeon to create clonal duplicates of themselves (bodies only, no consciousnesses), and another machine to swap their minds with those clones, and then they could run in, run out, and swap minds back (or just use their original bodies and swap into the clones afterwards), and then just kill or leave the clone bodies - it's not a solution my players went with (though they did play around with both machines to fun effect) but it could have been managed almost roll-free, or in a way in which the rolls mattered less. And I had other weird things in that dungeon that could probably have been used to similar effect.

So, like, the rules have to be there in case the players rush in and just decide to try and kill the shamblers and wade into the fungus and hope they don't die. There need to be Constitution saves and grappling rules and ACs and rules for spells and all that stuff. But there were also solutions that really didn't need a lot of rolls.

I mean, me too. That example is so lopsided it's practically a strawman, given how broken and awful the 4e skill challenge system was, combined with how completely dry and awful the examples of the skill challenges in the book ended up being.

Right, I mean I'm being hyperbolic to try and prove a point, I'm just saying, thinking is basically more fun than rolling - you need some rolling to simulate risky situations and the like, but the core of the game's fun, for me, is creative problem-solving while roleplaying a character, as opposed to mathematical craftsmanship.

On that note, we were going to play Fate at some point... whatever happened with that? I still want to!

Yeah I know, I'm so sorry about this! And I really want to too! Seriously I have a ton of notes for the adventure and am pumped to play it and have lots of interesting Fate-y situations mapped out for it. Here's what happened: I finished my PhD just as I was writing up Alptraum. Afterwards, I immediately got hired on teaching contracts, to the point where I was working at 3 different institutions for awhile. This is obviously good - I weas really pleased to get the work - but it's meant that for the last year, I've been teaching 4 courses at once each term, often new courses I hadn't run before, and often spread out throughout my city, with lots of commuting time between - some days I'd have to travel back and forth across the city several times. This term I had 146 students, and I just finished all their term papers, and still have about 120 exams to mark. Throw on top of that conference papers, a monograph proposal, and other post-PhD career development stuff, as well as my other social commitments, and my free time shrank to almost nothing. I have paused and unpaused my regular real-life game a whole bunch over that time too (we haven't played since January). I have tinkered with Alptraum over time but in typical Steerpike fashion I have probably made it over-ambitious. I do hope to actually polish it up and finish it, though, so despite the fact it's taken me FOREVER I plan on honouring our deal! It's also made me appreciate Fate's flexibility, even if it's still not my favourite system ever. I'd play it over a lot of other systems, anyway.

[sparkletwist]

I do think it can be fun to play a character who even sucks at the routine stuff - as you said with Dagny, she's a terrible negotiator but a good inventor/wizard, for instance.

Well, if you're going to use Dagny as an example, I'm not sure if I really agree. She is a rather min-maxed character, after all. Her Cha is bad so she could have a high Int... and because I'm not a filthy munchkin, I roleplay the consequences of that. I actually saw her (and other optimized characters) as counterexamples to the video's seeming premise (at least in the second half) that getting into roleplaying a character's flaws and weaknesses is somehow less valid or viable with a mechanically effective character.

This feels very much like a matter of personal taste.

Yes and no. Like many things in RPGs, how well it works for you and your group ultimately comes down to personal taste, but I think there are certain more-or-less objective consequences for the game if you play a certain way.

I see the rules as being there as a guideline for the DM to make rulings to adjuciate the world, and those guidelines become necessary when someone's performing some sort of action with a reasonable chance of failure. So, it's important that the rules are there for the DM to help make those sorts of calls, because theoretically the PCs could probably attempt to muscle their way through encounters. But because I want to generally discourage that sort of approach, the rules are important as a kind of bulwark against that hack and slash strategy- they have to be there, and be detailed enough, in order for the PCs to actually feel discouraged from taking the most obvious approach.

This makes sense, but I guess I don't really see how that actually refutes anything that I said. Maybe you weren't trying to, but it does seem like there's a disagreement here, and I honestly don't see what point you're trying to make. I have no problem with saying that the rules are guidelines for the DM to adjudicate the world-- but then I'd say that means there should be actual guidelines for the tasks that the players are most commonly expected to do, so that the DM can actually have help from the system in the task of adjudicating the world. Going "off-the-sheet" too often means the DM isn't getting any help from the system, because things are happening that the system doesn't support. This will always happen from time to time, and it's why tabletop RPGs allow such creativity, but if it happens too often, I feel like it's a failure in the system. After all, if there are no rules then the DM has no guidelines for adjudicating anything. How is this not a bad thing?

Yeah I know, I'm so sorry about this! And I really want to too!

Don't worry about it. No rush or pressure of course. I just wanted to let you know that I was still interested!

It's also made me appreciate Fate's flexibility, even if it's still not my favourite system ever. I'd play it over a lot of other systems, anyway.

I'm happy to hear this, if only because I'm kind of a shameless Fate fangirl. Admittedly it always did (and still somewhat does) confuse me that a lot of what you seem to like in RPGs lately kind of meshes with the Fate mentality, especially when it comes to abandoning baroque rules for creativity and abstraction, but you were never quite as on board as Fate as I am.

[Steerpike]

I actually saw her (and other optimized characters) as counterexamples to the video's seeming premise (at least in the second half) that getting into roleplaying a character's flaws and weaknesses is somehow less valid or viable with a mechanically effective character.

That's interesting. I think the video was presuming random rolling, not point buy. But I think the video would be more in favour of a min-maxed Dagny type character than, say, a "balanced" character with more evenly good stats, because then the character has some serious weaknesses (this seems to be the situation being described at the beginning of the video, with the fighter who can't speak persuasively).

I have no problem with saying that the rules are guidelines for the DM to adjudicate the world-- but then I'd say that means there should be actual guidelines for the tasks that the players are most commonly expected to do, so that the DM can actually have help from the system in the task of adjudicating the world. Going "off-the-sheet" too often means the DM isn't getting any help from the system, because things are happening that the system doesn't support.

This is the core difference that I was trying to get at, especially the part I bolded. If I'm understanding you correctly, I think you see the rules as being a fairly central part of any roleplaying game in an ideal situation, and a big part of the fun of the game, and that if you're not regularly using the rules, the rules are probably poorly designed or a bad fit for the game being run. Like, you want the rules to be in place to model the stuff the characters do regularly, and if the rules aren't being used a lot in the game, that's probably a sign the system is deficient in some sense, or at least the wrong system for whatever is being done with it.

Conversely, I want the rules to be more like a kind of fallback option for the DM when simple conversation with the players and descriptions of actions and places won't cut it. For me the rules need to be there in case the DM needs them, but a lot of the time they're just not that necessary if the game is run in a particular style - and that style, I think, is one aided by asymmetric power balances between players and monsters, traps, etc.

Like, for example: imagine there's a huge lava pit in one hallway of the dungeon, beyond which lies a treasure, and in another part of the dungeon there's a fountain whose water causes people to float (think Wonka-esque fizzy lifting drink), but it's guarded by an extremely dangerous naga who will let characters drink if they convince it they have a noble intent and are pure of heart. In yet another chamber there is a cursed treasure - an amulet which physically harms a character if they lie, potentially killing them if a grievous enough lie is told (btw, I just made this up, going to use it now in a game...).

So, the characters could try to jump the pit. Or they could try to kill the naga, or sneak past it. Those approaches are all going to need some sort of mechanic, I think - jumping, combat, stealth. Even just straight-up persuading the naga might or might not need a roll - if they have some sort of obvious proof of their moral purity, they might be able to just talk their way past without rolling, but if they're lying or just persuading with little proof, I'd probably call for a roll. As a DM, I need rules and rolls to adjudicate those sorts of encounters. But on the other hand, if the characters have explored thoroughly and put these pieces together, they might realize that if a character wears the amulet and carefully vouches for the party in such a way that no lies are told, they can all get the drink and float across (the naga knows about the amulet). If they went with that solution - or something similarly clever - I don't see why I need to shoe-horn rolls in at all. I just wouldn't need help from the system, or any hypothetical system, to adjudicate that course of action.

Now, here's precisely why hyper-competant characters and/or a "balanced" approach to encounter design can spoil this: if the pit can easily be leapt across or the naga is clearly a pushover (or if the characters are god-like as leapers/fighters/stealthy-types), there's less incentive to engage it in conversation or do something inventive like use a cursed object to convince the naga of good intentions. And also, I really do need the rules to be there if the naga conversation goes sideways or if some PC decides to risk things and jump, or if they go with some sort of other solution, like trying to distract the naga with an illusion or what-have-you. So there are rules, but PC creativity and critical thinking can make them unnecessary if the solution fits. Part of this is that as a DM I will tend to "say yes" to this kind of solution, or warn PCs if I think a solution is going to call for a roll.

Admittedly it always did (and still somewhat does) confuse me that a lot of what you seem to like in RPGs lately kind of meshes with the Fate mentality, especially when it comes to abandoning baroque rules for creativity and abstraction, but you were never quite as on board as Fate as I am.

It's basically just the meta-narrative thinking layer of Fate that sort of bugs me as a player, but that said, I've had fun designing situations for it. Honestly I think I'll have more fun DMing Fate than playing it. I have briefly experimented with it as a DM but never gave it enough time to really get comfortable with it.

[sparkletwist]

Conversely, I want the rules to be more like a kind of fallback option for the DM when simple conversation with the players and descriptions of actions and places won't cut it. For me the rules need to be there in case the DM needs them, but a lot of the time they're just not that necessary if the game is run in a particular style - and that style, I think, is one aided by asymmetric power balances between players and monsters, traps, etc.

I'm not opposed to that approach, in theory, but with the caveat that I don't think simple conversation and descriptions will cut it any time there's a significant amount of uncertainty in what the outcome of an action is going to be. In a typical RPG adventure, that can probably be quite often.

You are both right and wrong in your interpretation of how I see the rules, in that I do see them as a big part of the game and they should model the kinds of things that the players are likely to do, but I don't see not using the rules to be a bad thing inherently. If you are having one of those sessions based on storytelling and RP where everything just flowed and the dice were never picked up because it was always obvious what "should" happen... great! That kind of "not using the rules" is fine with me, because I'm not one to shoehorn mechanics in where they aren't needed.

However, the deficiency to me is when there is uncertainty and there isn't a rule to cover the uncertainty. At that point, the DM is having to make something up without any guidelines, and that runs the risk of being unfair, poorly thought out, entirely contrary to what the players thought was going to happen, lead to immersion-breaking disagreements at the table, or whatever. If that happens often, that definitely is a deficiency in the system, because it's not helping the DM make rulings in cases where that help is definitely needed.

And also, I really do need the rules to be there if the naga conversation goes sideways or if some PC decides to risk things and jump, or if they go with some sort of other solution, like trying to distract the naga with an illusion or what-have-you. So there are rules, but PC creativity and critical thinking can make them unnecessary if the solution fits. Part of this is that as a DM I will tend to "say yes" to this kind of solution, or warn PCs if I think a solution is going to call for a rol

I don't really see how this example proves your point at all. You laid out a clever example puzzle with a defined solution based on some interesting pieces that you dropped throughout the dungeon, but then you also seem to have acknowledged that you would need rules to fall back on if the players did just about anything but the very narrow set of specific solutions you thought of, and how often do players ever do exactly what you expect they will? (I mean, unless you set it up so that the players have to either solve a puzzle in one of the specific ways you think they ought to or deal with a punitively difficult die roll that will probably kill their characters if they fail, but I'm assuming that's not the approach you're taking, because that's just bad DMing...)

So anyway, the rules don't exist to cover situations where you're sure what's going to happen, but rather to cover situations where you don't know what's going to happen. If they solve the puzzle in exactly the way you were expecting, then you've already planned for all that, but what if they don't? What if the cryomancer tries to conjure enough ice to freeze the lava? What if they try to charm the naga with the magic snake charmer's flute they found a dozen sessions ago and you almost forgot they even had? What about all that other crazy stuff?

In these cases both the players and DM need some idea of what's going on, and like I said above, the rules exist to resolve uncertainty, or at least provide the DM a framework in which to make a ruling to resolve the uncertainty. The DM could simply make an ad hoc ruling out of the blue, but that's often less optimal, for the reasons I went into above-- especially if what the players are trying has a pretty good chance of not working. The other part of that saying you're quoting is "say yes or roll the dice," after all, and if you're going to roll the dice, you need to have some structure in place to know what those numbers are, and it usually helps (and greatly increases player agency) if the players know them too.

Honestly I think I'll have more fun DMing Fate than playing it.

Interestingly, that's more or less how I feel about some of the more nasty OSR tropes, and negadungeons, and other player-disempowering stuff. I wouldn't want it inflicted on me, but I guess I can deal with it if you're going to hand me all the power.

[Steerpike]

If you are having one of those sessions based on storytelling and RP where everything just flowed and the dice were never picked up because it was always obvious what "should" happen... great! That kind of "not using the rules" is fine with me, because I'm not one to shoehorn mechanics in where they aren't needed.

OK, we're closer than I thought on that front then.

However, the deficiency to me is when there is uncertainty and there isn't a rule to cover the uncertainty. At that point, the DM is having to make something up without any guidelines, and that runs the risk of being unfair, poorly thought out, entirely contrary to what the players thought was going to happen, lead to immersion-breaking disagreements at the table, or whatever. If that happens often, that definitely is a deficiency in the system, because it's not helping the DM make rulings in cases where that help is definitely needed.

I wouldn't oppose this either. I think we do have some differences about the particulars of the best sorts of rules to help cover unforseen uncertainties, but basically we're aligned here.

One way to think about the kind of thing I'm talking about is that the off-the-sheet solutions I'm imagining tend to involve considerably less uncertainty. More on this below.

I don't really see how this example proves your point at all. You laid out a clever example puzzle with a defined solution based on some interesting pieces that you dropped throughout the dungeon, but then you also seem to have acknowledged that you would need rules to fall back on if the players did just about anything but the very narrow set of specific solutions you thought of, and how often do players ever do exactly what you expect they will? (I mean, unless you set it up so that the players have to either solve a puzzle in one of the specific ways you think they ought to or deal with a punitively difficult die roll that will probably kill their characters if they fail, but I'm assuming that's not the approach you're taking, because that's just bad DMing...)

Well I was thinking up an example on the fly - if I was running this for real, there would be many more possible solutions, but I'd need to lay out a whole dungeon. Like, maybe there's a totally different possible solution around fire immunity if you can earn the blessing of the Immolated One which requires a humanoid sacrifice, and the naga has an evil twin in the dungeon who can be convinced to distract the fountain-naga if the PCs agree to befoul the fountain after drinking from it, and there's a mad druid who can polymorph the characters into bats if they bring them hallucinogenic mushrooms from the svirfneblin caves, etc - the dungeon becomes complex and the possibilities multiply, but a lot of these possibilities are still oriented around the application of various dungeon features and figuring out interactions and thinking laterally, rather than plugging in Character Ability X to fit Challenge Y, or straightforwardly fighting things, or doing other stuff that clearly is going to call for a lot of dice-rolling.

One could imagine a sliding scale of different rules here. Straight up combat with the naga is going to require more rolls and rules than talking with it, even without the amulet, for example. And the types of solutions that are most interesting to me often have less need for rules.

In these cases both the players and DM need some idea of what's going on, and like I said above, the rules exist to resolve uncertainty, or at least provide the DM a framework in which to make a ruling to resolve the uncertainty.

So yeah, I don't think we disagree on this fundamentally or in theory. This is also what I see the rules as being for.

What if the cryomancer tries to conjure enough ice to freeze the lava? What if they try to charm the naga with the magic snake charmer's flute they found a dozen sessions ago and you almost forgot they even had? What about all that other crazy stuff?

Oh totally! And some of those solutions will require tons of rules and some will require very few and many will require on-the-fly rulings because no system can be comprehensive about everything. But to bring it back to the point about character weakness, all of these sorts of crazy solutions become more and more necessary if the PCs are faced with challenges that can't be overcome in an obvious or straightforward manner. And I'm suggesting that the most rolls-intensive solutions, like combat or a series of clear-cut skill checks, are usually the more obvious ones.

To put it another way: creative thinking is often a way of reducing uncertainty, and therefore risk. If a creative solution to a problem is going to involve as much uncertainy and risk as the straightforward solution, it's essentially a redundant or pointless solution. But if it serves to mitigate risk and uncertainty, it becomes worthwhile. Sometimes that might take the form of advantages/bonuses to rolls, sure, but in other instances it can eliminate rolls entirely, or drastically reduce them; solutions that are less uncertain and less risky tend to have fewer rolls (at least as I like to DM), because rolls model uncertainty. Weak characters (or strong opponents) increase the uncertainty and risk associated with obvious solutions, and so foster solutions less likely to be based on straightforward or orthodox applications of the game's rules.

EDIT: An extremely good example from a game I'm playing in currently (this was last night, in fact). The DM is running Wizards' actually-pretty-darn-good Tomb of Annhilation, 5th edition D&D, a sort of mash-up/remix/remake of Tomb of Horrors and Dwellers of the Forbidden City which so far has felt a bit Heart of Darkness and a bit Indiana Jones:

[spoiler=Example]We're in the shrine of the Crocodile God. Our guide tells us a myth about how the Crocodile befriended humanity after they gave one another a piggyback to show one another their respective lands. Inside the dungeon, there are traps galore. A PC dies to a flame trap. My own character is at 0 hp, we have to use a cure to keep going, and another character is super weak. But we study some engravings in the dungeon and see a particular set of panels that require two people to push, standing on one anothers' shoulders, which we have to do to unlock a door. We remember the story our guide told us. We approach the final treasure, atop a column with a flight of stairs, magic in some way. We're all weak, low on cure spells. But these carvings of the piggybacking croc/human thing are all around the walls. We figure... what if we try the piggyback thing? It works, no traps go off as we climb the stairs, we get the treasure, and we piggyback out of the dungeon, unmolested by any of the traps we tripped before. We had to think about dungeon features, explore, and use some creative thinking, but only because the traps were too mean to just risk our luck with.

Now we had to roll when we were just trying to find/disable the traps, but we didn't need to roll for piggybacking, because that was a lower risk solution without uncertainty, which we only arrived at through careful observation, experimentation, engaging with the information we'd been provided, etc.

It felt quite a bit like this scene from the third Indiana Jones, where Indy is using his father's notes to navigate the trials of the Grail - the first two traps he almost screws up and has to pass a Dex save (basically), but by the third he's figured out the logic of the place and just has to walk forwards, depending on his knowledge rather than his reflexes. He could have tried to somehow use his whip to get across or to climb or something crazy, but he uses his head instead. The subsequent Grail scene is also a pretty good example of on off-the-sheet D&D-type challenge, in that there's no "roll" to pass the trial (apart maybe for a saving throw to avoid dying from the poisoned cups), just a choice based on careful reasoning and information available to everyone (Christ was a carpenter).[/spoiler]

[sparkletwist]

One way to think about the kind of thing I'm talking about is that the off-the-sheet solutions I'm imagining tend to involve considerably less uncertainty.

This makes sense, and I like your list of examples, but your examples are still features in the dungeon that you're thinking of as possible solutions to this problem-- which of course means you've already accounted for them, so they aren't in the realm of uncertainty. I know that you're making up the whole example yourself, so it's basically impossible for you to do anything but, since you can't make a list of things that you haven't thought of. The point I'm trying to make, though, is that in a real game the players are going to think of some things that you haven't accounted for, and perhaps try to use items they have or character abilities or features of the dungeon or whatever in ways that you didn't anticipate they ever would. I tried to throw out a few possibilities in my last post about a cryomancer and a magic flute, or whatever, as examples of things that the DM might not have thought of, and you responded:

And some of those solutions will require tons of rules and some will require very few and many will require on-the-fly rulings because no system can be comprehensive about everything.

This is a true statement, but it doesn't seem to me to mesh very well with your other idea that creative solutions are somehow reducing uncertainty. Unless they're the exact creative solutions that the DM planned for in advance, I guess? It really seems like there's a need for something more concrete here. To go on with that idea...

If a creative solution to a problem is going to involve as much uncertainy and risk as the straightforward solution, it's essentially a redundant or pointless solution. But if it serves to mitigate risk and uncertainty, it becomes worthwhile.

This makes sense to me, but in the absence of a system that includes mechanics that support the kinds of tasks the players are reasonably expected to do, how are the players supposed to even know that? If there is no real mechanical support for the players' actions so the DM is more or less making up everything ad hoc, how are the players supposed to have any idea what's possible beyond a lengthy and possibly frustrating (for both sides) Q&A session? I know that no rules can be truly comprehensive, and there's always going to be DM ruling and a certain degree of questioning and negotiation to get to that ruling, but I also feel like it's important to have some idea what their chances are, and this is especially true in a situation where any time the dice come out the odds are likely to be stacked against them.

I mean, we can say that a "good idea" will always just work by DM fiat, but that's very subjective and difficult to quantify, especially when the players earnestly think something will work and the DM, either by virtue of having more information about the setting or just a different opinion on things, has decided that it won't. In the absence of any sort of structure being provided by the rules, what then? How is anyone to know, and how can we avoid an argument, if not by using rules?

At one point some years ago, I played in a game where we spent the better part of an entire session hatching this complicated plan to trick the main villain involving using this psionic artifact that we'd found to conjure illusions or something like that. The whole thing failed, but not due to bad luck or plot twists or anything, but rather the GM had just decided that our plan was a stupid idea and wasn't going to work. To a certain extent, it was the GM's fault; she was a somewhat inexperienced GM and we talked this out later on and she did admit she probably handled it wrong. However, since the whole thing was ultimately based on a "crazy off-the-sheet solution" adjudicated by GM fiat, I also feel like it's relevant.

An extremely good example from a game I'm playing in currently (this was last night, in fact).

Maybe I'm not getting the full nuance of the situation, or I'm reading things very uncharitably, but this seems more like the sort of rigorous solution that doesn't really count because it's one specific thing that has been predefined by the DM to always lead to automatic success. I mean, sometimes puzzles in RPGs work like that, but I generally prefer more open-ended sorts of things, because it can be annoying otherwise-- it's no fun to just be stuck, and it eats up valuable play time. When thing are open-ended, though, that means the solution is often going to be something the DM never thought of.