What makes a character competent?

[Xervous]

Taken generally, what does it mean to you for a TTRPG character to be competent?

To me competence carries with it implications of reliability in the field being so described. It covers the trials the character might face and brushes aside the trivial without a second thought. The character, in their fields of competency, is not unwillingly endlessly in over their head unless the genre or the GM happens to be a horror story.

[LibraryOgre]

While what it looks like, mechanically, can vary a lot between systems, I would say "Can usually succeed in relatively average situations in a single broad area of endeavor", which is so vague it might as well be my brother-in-law.

Competent as a warrior-type means "can give and receive damage for several rounds of sustained combat."
Competent as a scout-type means "can sneak about better than they can be found by most people."

Basically, an average person should have only a slim chance of beating them in their chosen field of endeavor, and that field should be relatively broadly applicable. Your face SHOULD usually be able to wheedle information out of someone. But if their chosen specialty is very niche (i.e. "No one weaves baskets underwater better than Diego de la Mer!"), then they may be competent in that, but not PC-competent.

[kyoryu]

1. In general, characters should be able to impact the problems that they are dealing with. This may require planning or work (antagonists may be more "powerful"), but it should be feasible that the characters are at least in the running.
2. Characters are generally capable of doing the things that they do. When attempting reasonable tasks, they fail because of circumstances, not because of incompetence. They may fail because a task is just too tough, but they won't just randomly fall down the stairs.

The first point is actually really wide... the kids in Stranger Things are competent. They're outgunned, and they're fighting above their weight class, but they can and do impact things and do a lot to help stave off the threats.

There's a common belief that "competent" means some kind of hyper-competence - that you're the best, that you never fail, etc. This isn't required.

While what it looks like, mechanically, can vary a lot between systems, I would say "Can usually succeed in relatively average situations in a single broad area of endeavor", which is so vague it might as well be my brother-in-law.

I'd add "based on the overall genre/situation" to that. In a Kids On Bikes game, competence is defined very differently than it would be in a Special Forces game or a Supers game.

[Tvtyrant]

Taken generally, what does it mean to you for a TTRPG character to be competent?

To me competence carries with it implications of reliability in the field being so described. It covers the trials the character might face and brushes aside the trivial without a second thought. The character, in their fields of competency, is not unwillingly endlessly in over their head unless the genre or the GM happens to be a horror story.

Competence to me is a combination of a strong amount of pragmatism, common sense, creativity and no panicking when running into something they are not immediately equipped to handle. They usually focus on solutions instead of getting caught up in the scale of the problem, and are present in the moment. If the world were ending and someone was bleeding they would bandage the wound while looking for immediate safety, if they were facing aliens without weapons they would be focused on hiding, then shelter, then resources.

Basically competence is clarity, the ability to see which issues can actually be resolved and resolve those.

Edit: An everyday example would be the difference between someone who when their car breaks down freaks out and spends 10 minutes lamenting their fate, and the person who calls AAA. Neither knows how to fix a car, one person acknowledged it before there was a crisis and took steps to prevent it from being a crises. Or if you have a broken pipe and you try to fix it but fail vs the person who has a plumber in their phone; you don't have to fix a problem personally, you just have to know a way a problem can be fixed and be ready for it.

[Kyutaru]

Competent means they can find options in most situations. A character who is frequently in a scenario where there is nothing he can do and he's completely useless and the outcome of the encounter is out of his hands lacks the necessary skillset to be deemed competent. Even in a hopelessly outgunned battle, competent characters still find ways to be valuable assets to the team by utilizing their skills.

There are stories where even powerful characters are rendered incompetent by their limited experience and inability to apply themselves to the given conflict. They end up making decisions that in hindsight were not the most optimal because they didn't know what else to do at the time. Competent characters make no excuses and strive to find a solution even when it seems like there is nothing that can be done. Superheroes are frequently put into a one-sided battle and have to demonstrate their competence by coming up with ways to use whatever is available to them in the environment, in their pockets, or in their cunning.

[Man_Over_Game]

Taken generally, what does it mean to you for a TTRPG character to be competent?

To me competence carries with it implications of reliability in the field being so described. It covers the trials the character might face and brushes aside the trivial without a second thought. The character, in their fields of competency, is not unwillingly endlessly in over their head unless the genre or the GM happens to be a horror story.

Good enough in their niche that the rest of the party can rely on them to take charge.
Good enough out of their niche that allies can count on their support.

[Telok]

To define competence for a game system you ideally want a baseline snd a scale for a task or action, a consistent level of specificity in the skill/tasks, and a baseline for what both incompetence and competence look like.

Driving a car or riding a bicycle would probably involve getting around locally on a regular basis and occasional long(er) distances, without significant or frequent damage to anything or anyone. But do you expect a competent driver to be able to avoid sudden obstacles or hazards? How often? Shouls a competent bicyclist be able to do tricks, or just get around? Does the system distinguish between the ability to drive cars, bicycles, canoes, and remote operated submarines?

For a competent generalist computer programmer I'd expect passing familarity with a wide array of things and the ability to become proficient in a specific set of tools and activities in a couple weeks. For a competent database or network administrator I'd expect a narrower focus but a deeper understanding of their specialty.

Finally, is there any chance of incompetent people succeeding? Does success look different between competent and incompetent successes or failures? I've seen incompetently build and administered databases. They are things of horror and pain. But they mostly worked, most of the time, for a sort of minimum functionality. Competently done databases can be handed to competent database administrators and you'll be getting useful information in 20 minutes, and complex searches & deep anaylsis in a day or two*.

So how specific are your skills, tasks, and task resolution system? Is the difference between incompetent & competent a simple "may not try vs. may try" or is it "worse at vs. better at"? Is your resolution success/fail or is it degrees of success and failure? These will all change how competency is defined in your game.

Lets say you want to define competency at driving a car as bring able to avoid an acciden you can see coming about half the time. If you have a generic "drive' skill and use a d20+mod vs. dc system then competent is defined as making your mod equal to the dc-10. But that's competence in a vacuum, meaningless without context. Is your scale dc 5 to back out of a parking spot, 15 to avoid a crash, 25 to jump a mile wide canyon? Is it no roll to drive normally, 20 to avoid the crash, 40 to drive on snow and ice? Do incompetent or unskilled people get to roll? Do you calculate the mod just from skill, or from something like skill+stat+1/4th level+car max speed? Do you want compent drivers to also be competent nuclear submarine pilots? Are you doing yes/no success or degrees of success?

Just asking what a competent skill check looks like depends a lot on how the entire system works. D&D these days decided on a rolling system, yes/no successes, everyone rolls for anything, and dc scale first, then tried to fit skills into that. Other systems have chosen dice systems, then defined competency, then set dc scales and the rest of it. Some systems (Paranoia) are intentionally partly to mostly lol-random because they're supposed to produce silly or randon results. Some systems can be deterministic based on stacking mods, effects, whatever because they're about choosing what you use your resources to succeed at.

*Don't use this to judge your local DB admin. If the database contains information from before 2000 there's a decent chance it's one of the nightmare inducing messes and the poor admin inherited it with little or no info on how it is set up. In addition to possible bad data that might just randomly crash it.

[Kaptin Keen]

Taken generally, what does it mean to you for a TTRPG character to be competent?

To me competence carries with it implications of reliability in the field being so described. It covers the trials the character might face and brushes aside the trivial without a second thought. The character, in their fields of competency, is not unwillingly endlessly in over their head unless the genre or the GM happens to be a horror story.

Yea ... being able to succeed at the things the character was designed to succeed at. That's almost always my only source of dissatisfaction in RPG's: When I feel the game promises me that a style of character is good at something - and then it isn't. Or another style of character is actually better for no apparent reason (eg. wizards are better rogues than rogues are).

[evilmastermind]

Taken generally, what does it mean to you for a TTRPG character to be competent?

To me competence carries with it implications of reliability in the field being so described. It covers the trials the character might face and brushes aside the trivial without a second thought. The character, in their fields of competency, is not unwillingly endlessly in over their head unless the genre or the GM happens to be a horror story.

Reilability is the one of the two most important things, because it's all dependent on how likeable he or she is to the players. I've seen "competent" characters shamed and abandoned by groups of players who thought that their skills weren't worth bringing them along. So you can be skillful, reliable but you also need that little bit of charisma to be considered competent at your job... it's subjective and in the players/group's minds.

[Quertus]

This is tricky. Much like "fun", "competent" can be difficult to define. So, instead, let's look at the negative space, to get a feel for the shape of competence.

-----

What can make a character (feel) incompetent?

#1 - the Player.

Lots of what people are taking about in this thread is making good choices, using what you have, evaluating the scenario. Simply put, player skills.

If the player is incompetent, the character is going to feel incompetent,
and there's not much you can do about that. Put me in Warhammer 40k, and it's a safe bet that my character will be incompetent, because that universe just… I haven't seen the elephant.

#2 - the GM

Player: "I put on my pants".

GM: "OK, you put your pants on your head, because you didn't say *where* you put your pants on."

A bad GM can definitely have PCs come across as "pants on head" incompetent through failing to serve as a reasonable interface between the player, the character, and the world.

#3 - the system

Bounded Accuracy has a dozen hobos have a better chance to solve problems than Reed Richards. 5e really pulls a "Captain Hobo" on supposedly competent characters.

Or look at the Warhammer line - most stats are around 30-40, and that's your % chance of success (or half that if untrained), with 96-100 being a botch. I really want to "play it honest", and run through the party trying to order a meal in a restaurant with such a system.

#4 - the character's mechanics

So, I've got this character who *you'd think* could be totally intimidating. And the GM is totally into social things. But, even when I intimidate with perfect timing at great advantage, the character never successfully intimates anyone. Why? Because I "forgot" to put any resources into intimidation. (Never letting anyone be intimidated by that character is one of the best things that that particular GM has ever done).

#5 - the character's personality

Quertus, my signature academia mage for whom this account is named, is intentionally *not* the Determinator. Or look at (at least for most of the strip).

The character could have every mechanical advantage, and a player who *could* make better choices, yet still come off as incompetent (at least, in certain fields), because that's the way that they're roleplayed.

-----

So, now that we've drawn the negative space, what does a competent character look like?

The GM does a good job of describing the scenario. The player has the ability to evaluate the scenario, and make good choices. The character has the tools and wherewithal to act on the player's wisdom. The GM interprets the player's action declaration benevolently, and has the skills to catch and clarify any miscommunications. And the system doesn't get in the way and say "lol, you fail anyway".

[NichG]

In my mind, competence is less about what a character can do, and more about the character being able to correctly determine what they can do and act or plan accordingly. A incompetent character is one who takes actions that rely on them being able to do things which they can't deliver. I don't expect a competent blacksmith to be able to dismantle the army of angry iron golems, but I do expect them to say 'having me go out there and hammer on an iron golem is stupid, it won't do anything other than get me killed, you need a warrior not a blacksmith'. Similarly, they should be able to speak confidently as to the things they can provide and not mistakenly underestimate that. A warrior who is afraid to get into fights they can fairly reliably win isn't displaying competence to me.

Similarly, if the character could reliably succeed by doing things a certain way, but fails or believes they can't succeed because they can't think of that method or are stuck on a preferred method, it looks incompetent. E.g. if a Fighter says 'I can't take on this demon because it can teleport at will, regenerates, and we don't have anything I could use which would break it's DR' then fine, that's within the sphere of competence - they know their limits. Even if, say, there was a caster who could let them deal with some of those problems, I don't think it's a big mark against their competence as a Fighter to not think 'oh, lets get the Wizard to drop a dimensional anchor and have the Cleric bless my weapon' (though it would be a bit of a mark against their competence if they were presenting themselves as a party leader or strategist/tactician).

If they say 'I can't take on this griffon because it flies and my signature weapon is a sword' then it starts to look incompetent - using a bow is within the range of things that a Fighter is supposed to be proficient in, and using a ranged weapon (even if you're bad at it) against a flier that only has melee attacks is a viable plan. So if they say 'I can't do it' when I can think of ways that they could, that looks incompetent.

[Raijinken]

Competency? Like what our GM usually, if not, ALWAYS slaps in our faces: The Player behind it!

[aglondier]

I think that if they can function just fine in their perceived weight class, and regularly punch well above it, then they are competent PCs.
I played a Shadowrun campaign where complementary characters and well-oiled teamwork had just beyond starting characters surviving runs that should have been tpks...

Hmmm...probably helped that the players had been gaming together for years, and two had a near precognitive knowledge of what the other would do in any situation...

[Rynjin]

I can put a pretty simple numerical value on it: when a character can always succeed at something more than 50% of the time, they are competent with that task. Basically, in a d20 system if a character can Take 10 to consistently complete level appropriate skill checks, or hit an enemy on a 10 or lower die roll, etc. they are competent at skills, or attacking, or what have you. Skills is a particular important one as that level of competence means you can ALWAYS successfully complete that task when not under some kind of undue stress...as someone should be. A trained, competent carpenter shouldn't fail at making a chair even 5% of the time unless there's something distracting them.

Hyper-competence would be about a 70% success rate or higher on level appropriate challenges.

[icefractal]

I think that depends a lot on what the task is.

50% chance to safely land a heavily damaged plane on a way-too-short improvised runway? Yeah, that's a competent pilot.

50% chance to safely land a plane on a normal runway in normal conditions? Not a competent pilot, would probably lose their job immediately.

Honestly for things that are part of someone's normal job, even a 5% chance to fail is usually too much. Like, how often should a pilot fail to land a plane? I would say in normal conditions, never. Their skill should be enough that they only fail when things go very wrong.


Combat is usually an example of "things going at least somewhat wrong", and so lower chances are acceptable than in most areas. Even then, I feel like a 50% whiff rate isn't going to feel that competent, even if it compares favorably to most opposition. IIRC, WotC did testing on this and found that an average 70% hit rate was the most generally enjoyable, at least in a D&D paradigm where a single hit is seldom decisive.

[Telok]

.Like, how often should a pilot fail to land a plane

Oh, nobody ever fails to land a plane. It's just a question of how many chunks end up scattered across the landscape and how close to the intended landing spot they are.

In that vein, what do we mean when we say someone is competent? That they can do something well and often that they can do it well with regularity. By that definition any character is competent at anything they can succeed at more than half the time. I’d ignore player & DM contributions to competency as that’s something which is generally outside the scope and control of an RPG system.

If competency is a simple 51% success rate then in the WotC D&Ds and d20 systems pretty much every human in existence is competent at DC 10 tasks, every spellcaster is competent at casting spells (in fact they’re awesome at it because they almost never fail to cast a spell when they try to), and even regular horses are competent to highly competent at DC 5 intelligence tasks. This has the unfortunate side effect of implying things like the act of figuring out how to lift a bar in order to open a gate or barn door being higher than a DC 5 intelligence task*. Perhaps a more accurate conclusion would be that the concept of competency is something that cannot, or maybe should not be expressed with a single mechanic that comes down to what is effectively a random coin flip.

I would argue that you need to consider at least three things in evaluating a character’s competency within a system of game mechanics. The scope of the tasks being considered, the chance of success at the tasks, and the magnitude of success at the tasks.

You want to consider the scope of the tasks that you’re trying to evaluate. Something like ‘hit a low threat rating goblin in an empty room during the day with an attack ability’ is going to be so specific and basic that nearly all possible characters within an RPG system would be considered competent. At the other extreme ‘meaningfully participate in a generic combat scenario’ is so vague and general that, again, nearly all possible characters will be considered competent. The scope of the tasks will also depend on the mechanics of the RPG. A game like Riddle of Steel will care about individual parries and ripostes within a single exchange of blows in combat, games like Fate may resolve a battle involving three armies in a single roll. Even within the current editions of D&D you can consider competency in individual aspects like attacking in melee, defending from physical attacks, and taking damage, versus an overall competency in combat in general. To move the D&D example away from combat you could have a situation where two characters engage in opposed skill checks until one has four more successes than the other or until 10 checks have been made, compared to a situation where one character makes one skill check to determine total success or absolute failure. Any definition of character competence will have to be specific to the mechanics of the system and the scope of the task.

The chance of success is what almost everyone will be focusing on. I’d consider the sustained rate of success to also be important, but it depends on the RPG again. Most games use some sort of resource mechanic, eventually the character runs out of resources and has to stop or else they risk serious harm or setbacks. Compare a system where you have a character resource that improves a character’s chances in a stealth situation with one that doesn’t have skill affecting resources. In the system with resources the character can try to fine tune their rate of success to the situation and may, over the course of several encounters, run out of the resources to maintain a particular level of success or even have any chance of success at all. By comparison the system without stealth skill related resources simply has the same chances of success at all times for an unlimited number of encounters (presuming equality in the difficulty of the encounters there). You can even have systems that mix the two options, D&D has casters who can nearly perfectly auto-succeed stealth encounters a few times a day and non-casters who have a chance (sometimes high, sometimes low) of succeeding in stealth encounters an effectively infinite number of times in a row. A definition of competence will depend on how many times a character is expected to need to succeed between some sort of resource reset and if they need to succeed every time or just most of the time.

Different RPG systems have different ways of measuring success. D&D measures combat success by a sustained rate of attack success with variable amounts of effect over the course of multiple attempts, giving you a range from a massively successful combat where nobody on your side took damage or used resources to an unrecoverable TPK. On the other hand, D&D traditionally measures non-combat actions like skills with a binary success/failure mode based on a single roll. Again, something like Exalted with its social combat and a range of outcomes will have a different concept of a competent “face” character than D&D with it’s usual “roll once to persuade/intimidate/deceive and completely succeed or totally fail” model. If a system allows for degrees of success in a task then competency will look different than a system that only checks for absolute success or failure.

TLDR: consider at least three things in evaluating a character’s competency within a system of game mechanics. The scope of the tasks being considered, the chance of success at the tasks, and the magnitude of success at the tasks. Any definition of character competence will have to be specific to the mechanics of the system and the scope of the task. A definition of competence will depend on how many times a character is expected to need to succeed between some sort of resource reset and if they need to succeed every time or just most of the time. If a system allows for degrees of success in a task then competency will look different than a system that only checks for absolute success or failure.

Taking just D&D 5e: Is a competent combatant someone who has a minimum 16 attack stat, 1d8 or larger basic attack, 14+ AC, and a 12+ Con with at least average HP rolls? Is it a character that can always help the team win in any level equivalent encounter? Or is it a character that can encompasses the player’s concept and not be a drag on the party? For example, if they want to play an unlettered and naïve primitive warrior, can that be done without the party being worse off in combat when the character participates? Are any or all of those “combat competent”? And remember, combat is generally the easy area in which to define this.

*

Spoiler

Show

I will be accused of ‘not using the skill system correctly’ because you aren’t supposed to roll if there’s not penalty for failure. Fine. The barn is on fire, make your intelligence check to figure out how to lift a bar in order to open the door before you burn to death. The point of the example isn’t about how to use a skill system, but how to define character competency within the mechanics of a skill system.

[Vahnavoi]

In a roleplaying game, there are two kinds of competence that matter:

1) a priori competence - everybody acts as if your character is competent, because that's what reads on your character sheet. F. ex. your character is stated to be smart, so everyone acts if their decisions are smart.

2) the ability of the player to pursue their goals in the game by making mechanical decisions.

Before you go awkwardly tinkering with mechanics of any game, you first have to decide which type of competency you're aiming for. First of these can be close to independent from player skill, the second cannot be.

[Seto]

In the context of a rpg character, I tend to understand it as a metagame term rather than in-universe. To me, a competent character (as opposed to a competent person irl) is a character whose mechanics allow them to represent their concept. If I build my character as a cook, they should be able to consistently make good dishes, but their ability in combat has little bearing on whether or not they're competent. If I build a spy, they should be smart, sneaky and have special training.
If, however, I try to build a monster hunter and my character ends up struggling against anything bigger than a horse, I'll be disappointed in my character and consider them incompetent. As in, they cannot reliably perform in situations they're expected to be able to handle.

If we're talking about in-universe, then you're just asking to define "competency". What is a competent person? It's kind of a broad term. I'd say, someone who's good at what they do. They might also exhibit other talents, and be more broadly competent than what might be expected of them. If it's given without context, I'd say it's a person who's giften in several areas.

[KineticDiplomat]

Assuming this is about mechanics, I think competence can generally be described as a sweet spot where "for the typical challenge the PC might face in this system/setting/character specialty, he has a reasonably solid chance to make it through with reasonable certainty, but may still suffer negative consequences and has an outside chance of failure - the chance of failure and consequence increasing in the case of poor decisions. Major challenges will be dangerous, and possibly lethal (insert system equivalent), but the odds of failure go down with good decisions." If the dice give you a solid foundation to address challenges and the world, but cannot guarantee victory without player decisions, competent.

I would define under-competent as "cannot reasonably expect to survive or contribute to typical challenges." With a caveat that this does not mean every Ph.D needs to be a gunslinger. If the dice cannot reasonably give you a chance at mechanical success no matter the player decision, under-competent.

I would define over-competent as "is sufficiently mechanically powerful that typical situations pose no threat, and major challenges require errors to fail...or anything higher than this." Basically, if the dice and dice alone can win for you, you're probably "over-competent." There is a bit of an issue where people believe that because they can make a hyper-capable character, that is competence, when it is in fact the simple fact that with time and pressure you can find the gaps in a setting and rules and produce ludicrous results. This is because the (usually numerous) authors of the system have to be perfectly iron clad and in unison, often across multiple books years and teams apart, whereas the supposedly "just using RAW competent" character only has to find a few disconnects in a massive system. It's not particularly hard to do.

[Alcore]

PC, competent and table top...

If competency levels are equal between real life people a competent PC should be successful in area of expertise 50-60% of the time for an average difficultly. Difficulty should be scaled up or down accordingly (which is easier to do in some systems than others) bepending on how niched the PC is.

More broad PCs will face weaker encounters yet the DM doesn't have to tailor a location to make it survivable.



This is just competency that we are talking about; you just need to clear the bar. Or be the bar.

[icefractal]

I'm kind of confused where people are getting 50% from. I think of competent as roughly "professional" level, like people would generally trust you to do the thing in question.

And I can't think of many jobs where a 50% success rate at "average" tasks would be considered acceptable.

On a more purely OOC perspective, I'd say that what makes a PC seem competent at something is:
* Succeeding noticeably more often than failing at it. On screen.
* When everyone is rolling, almost always being better at it than untrained people are. As in, not failing when they succeed.

And a 50% rate doesn't seem to achieve that.

[Alcore]

I'm kind of confused where people are getting 50% from. I think of competent as roughly "professional" level, like people would generally trust you to do the thing in question.

And I can't think of many jobs where a 50% success rate at "average" tasks would be considered acceptable.

It's quite simple really... (and i didn't say tasks, i said difficulty)

A level 2 fighter with str14 and a masterwork weapon has a +5 to hit. On an average difficulty encounter he should be facing something with an AC between 16-14 (roughly 50-60%). This enemy should also being hitting the same (unless damage is far greater) and, ideally, the monster should run out of HP first.

This monster would probably kill most npcs of lower level and some PCs without cunning maneuvers on the PCs part.


Of course the example above involves dungeons and dragons. It is rather clunky in the balance department. Mutants and Masterminds does a better job if a lot more work for the DM.

[Rynjin]

I'm kind of confused where people are getting 50% from. I think of competent as roughly "professional" level, like people would generally trust you to do the thing in question.

And I can't think of many jobs where a 50% success rate at "average" tasks would be considered acceptable.

On a more purely OOC perspective, I'd say that what makes a PC seem competent at something is:
* Succeeding noticeably more often than failing at it. On screen.
* When everyone is rolling, almost always being better at it than untrained people are. As in, not failing when they succeed.

And a 50% rate doesn't seem to achieve that.

50% is specifically the metric for d20 systems, because typically Taking 10, or "Routine Checks", or some such mechanic exist.

A bare minimum 50% success rate is EXACTLY THE SAME as a 100% success rate when you're not in immediate danger or something, in other words, since you can put in a routine amount of effort and always succeed.

And when you are in danger...well, even competent people can choke in some sort of time crunch when bullets are flying and things are exploding.

[Vahnavoi]

Probability and percentages for single die rolls or single die rolls are a crappy way of gauging competency. You should be measuring success over real play time - f. ex., what would it take for a character to be successful over a 4 hour session of gameplay? You cannot determine this in a scenario-independent manner - not without turning the characters into spherical cows in a vacuum. In plain terms, to gauge competency, you need to know what the character needs to be competent at.

[Florian]

This is tricky

Mr. Senility already gave the answer without realizing it.

Warhammer, Traveller and A Time of War (Battletech) use the best approach: Base Skill, Task Difficulty and Time (Coupled with degrees of success and failure).

First, keep in mind the golden rule: At or above 50%, you succeed without rolling.

Take a look at WH40K/D100: Base Chance is based on Attribute, starting out in the aforementioned 20-30% range. Attribute can be raised by further in 4x 5% steps, skills associated with the attribute can be raised by further 2x 5% steps. Beyond that, you can either add or remove one degree of success or failure by further investing in your skill.

Now the interesting part is Task Difficulty. This ranges from +60% to -60% percent, from trivial to hellish.
Time is the second axis, also adding a further +/- 60% to the range.

This gives you a range to work with. For example, having to fix a nuclear reactor right now (-120%) is impossible, having al the time in the world to fix it (0%) will depend on whether the base chance is above or below 50%.

[OldTrees1]

And when you are in danger...well, even competent people can choke in some sort of time crunch when bullets are flying and things are exploding.

1) In RPGs we tend to model a 99.9% success rate as 100% and even model 97.5% as 100% in d20 systems. The error rate is just too small for the resolution of our RNG of choice.
2) In many professions routine tasks are expected to have very high success rates. 99% is considered negligent. Sometimes even 99.99% is negligent.
3) Yes a higher than normal stress* can decrease the success rate. However dropping from 99.9% to 99% (10x the failures) would still resolve as 100% in a d20 system.
* RPG characters are also competent in things where combat is normal (rather than "higher than normal") stress.


I see skill as growing from "unable to attempt" thru "possible but with a shrinking chance of failure" to "automatic". Running across a chain is something I could not do, much less do in combat. If a character is improving in that direction I expect them to eventually be able to roll, and then later not need to roll. So skill is this ever shifting window that makes new abilities possible and old abilities guaranteed. This lets you build off of those old abilities.


I see competence as checking 3 things:
1) Are the baseline abilities (old abilities that are now guaranteed successes) "level appropriate" for a professional of that field? Does a professional of that rank have a better baseline than the mechanics create?
2) Are the risky abilities "level appropriate" for a professional of that field? Would we expect a professional of that rank to be more reliable in those abilities?
3) Are the impossible abilities "level appropriate" for a professional of that field? Would we expect a professional of that rank to be able to do those abilities?


Now, what does "level appropriate" mean? That depends on the RPG, but pick a reasonable metric, and then be consistent.

[Vahnavoi]

1) In RPGs we tend to model a 99.9% success rate as 100% and even model 97.5% as 100% in d20 systems. The error rate is just too small for the resolution of our RNG of choice.
2) In many professions routine tasks are expected to have very high success rates. 99% is considered negligent. Sometimes even 99.99% is negligent.
3) Yes a higher than normal stress* can decrease the success rate. However dropping from 99.9% to 99% (10x the failures) would still resolve as 100% in a d20 system.

Every. Single. Thing. About probabilities and resolution space of d20 roll only holds if you're talking about a single die roll, and ceases to be true when talking about tasks involving multiple die rolls, such as combat. It isn't hard to create scenarios where you might have 5% chance to fail a single roll, but below 1% chance to fail the entire combat. Or, put differently, a success rate above 95% but below 100%.

The simple way around "limitations" of your RNG is to not use it in a deliberately crappy way.

[Kyutaru]

I see skill as growing from "unable to attempt" thru "possible but with a shrinking chance of failure" to "automatic". Running across a chain is something I could not do, much less do in combat. If a character is improving in that direction I expect them to eventually be able to roll, and then later not need to roll. So skill is this ever shifting window that makes new abilities possible and old abilities guaranteed. This lets you build off of those old abilities.

It's something a good DM handles appropriately and another reason that having predefined skill tables is horrible for roleplaying games. Running across a chain is an automatic thing for a high level, highly skilled acrobat. It doesn't require a roll unless there's some reason he might fail impeding his usual mastery. It's the kind of thing DMs only force rolls for when conditions are awful, similar to the difficulty task adjustments mentioned above that reduce the chance of failure till it becomes guaranteed or impossible. Some things simply can't be done and some things can always be done and definitions of competency must have a control value to measure against. We understand that running across a chain is hard. But what about walking across a bridge without falling off? It's pretty easy for anyone to do but I'm sure a drunk might find it a challenging DC.

[icefractal]

It's something a good DM handles appropriately and another reason that having predefined skill tables is horrible for roleplaying games.
...

I'm confused here, because the rest of your post seems to be an argument in favor of predefined skill tables. At least, they do all the things you mentioned:

Acrobatics:
Drunk w/ no skill: -2
Novice: +4
Expert: +10 (3rd level, +2 dex, +2 item or feat)
Grandmaster: +25 (10th level, +7 dex, +5 item or feats)

Narrow bridge: DC 5
Drunk has a 30% chance to fall off, the others automatically do it.

Stationary chain: DC 20
Drunk can't do it, novice has a 25% chance, expert has a 55% chance, grandmaster automatic.

Swinging chain: DC 25
Novice can't do it, expert has a 30% chance, grandmaster still automatic

Vigorously swinging chain: DC 30
Expert has a 5% chance, grandmaster has an 80% chance

Vigorously swinging icy chain at full speed: DC 40
Even the grandmaster finds this tricky, 30% chance
But a 20th level "goddamn demigod of acrobatics" could probably do it no sweat, as appropriate.

IMO, fixed DCs work great. The only benefit to no fixed numbers would seem to be the ability to fudge it - to have the swinging chain be easy or automatic when you want to let the PCs show off, and then turn around and make that same chain a tough roll when it fits the story. And personally, that's not a thing I find desirable as either a player or GM.

[OldTrees1]

Every. Single. Thing. About probabilities and resolution space of d20 roll only holds if you're talking about a single die roll, and ceases to be true when talking about tasks involving multiple die rolls, such as combat. It isn't hard to create scenarios where you might have 5% chance to fail a single roll, but below 1% chance to fail the entire combat. Or, put differently, a success rate above 95% but below 100%.

The simple way around "limitations" of your RNG is to not use it in a deliberately crappy way.

Was that tone (a period after every word?) intentional? I am going to ignore it since it does not fit the conversation.

I was still talking about single checks. Some of those checks might happen during combat (like running across a chain as part of your movement), but I was still taking about single checks.

Yes, it is possible to increase the resolution of the RNG by increasing the number of checks per check. The 5th level rogue rolls 1d20 for climbing the wall. The 10th level rogue rolls 2d20 at advantage for that wall. The 15th level rogue rolls 3d20 at advantage for that wall. Etc.

However that seems to be excessive resolution rather than just allowing rounding to eventually hit 100%. Taking a single check and tacking on extra checks just to give a chance of failure seems like a "deliberately crappy way" of handling competency. So, using your own advice, I don't see a reason to prevent a single check from eventually rounding to 100%.

[This paragraph is talking about multiple checks unlike the previous post or rest of this post]
I especially don't want to have to roll 27d20 to resolve a 20th level character's movement when a 1st level character only rolled 1d20 for their movement. Eventual rounding to 100% on single checks allows high level characters to have quantitative and qualitative growth in their abilities with only a small increase in the number of dice rolled for the same action. It is much nicer to roll 1d20 each for 3 components of a 20th level character's movement and have automatic passes on the other, now mastered, components to the movement.

It's something a good DM handles appropriately and another reason that having predefined skill tables is horrible for roleplaying games. Running across a chain is an automatic thing for a high level, highly skilled acrobat. It doesn't require a roll unless there's some reason he might fail impeding his usual mastery. It's the kind of thing DMs only force rolls for when conditions are awful, similar to the difficulty task adjustments mentioned above that reduce the chance of failure till it becomes guaranteed or impossible. Some things simply can't be done and some things can always be done and definitions of competency must have a control value to measure against. We understand that running across a chain is hard. But what about walking across a bridge without falling off? It's pretty easy for anyone to do but I'm sure a drunk might find it a challenging DC.

I hear you. I reach a different conclusion but I do hear you.
The player knowing what their character is capable of, and what is automatic for their character is useful information. You don't like using skill tables for that. I see skill tables as a very useful communication aide. However we both agree that the player being informed means the player could know when their character leveled up qualitatively (rather than just quantitatively).

To demonstrate the communication aide, if icefractal was the DM, I would know when my naturally talented specialized acrobat (see icefractal's acrobat) leveled up qualitatively and I would know when my clumsy drunk generalist acrobat (-2 Dex, -2 Drunk, half ranks generalist, no items) leveled up qualitatively. At 13th level or so my drunk could safely be drunk on narrow bridges.

But to reiterate, I hear you. You don't like skill tables, and your gaming is better off without them (as demonstrated in another thread). Different tools for different DMs.