This is a site for discussing roleplaying games. Have fun doing so, but there is one major rule: do not discuss political issues that aren't directly and uniquely related to the subject of the thread and about gaming. While this site is dedicated to free speech, the following will not be tolerated: devolving a thread into unrelated political discussion, sockpuppeting (using multiple and/or bogus accounts), disrupting topics without contributing to them, and posting images that could get someone fired in the workplace (an external link is OK, but clearly mark it as Not Safe For Work, or NSFW). If you receive a warning, please take it seriously and either move on to another topic or steer the discussion back to its original RPG-related theme.

Author Topic: Probability Theory and You  (Read 10256 times)

nDervish

  • Hero Member
  • *****
  • n
  • Posts: 750
Probability Theory and You
« Reply #30 on: May 11, 2020, 05:50:35 AM »
Quote from: Pat;1129485
Probability matters, but the mechanics around the probabilities often matter more. For instance, do you have to roll every time a skill comes up, or do you automatically succeed barring exceptional circumstances? Is your chance based almost entirely on your raw aptitude and training matters very little (AD&D's method), or the other way around? When you roll, under what circumstances, what contributes to the roll, and what success and failure mean can matter a lot more than the odds.


Another "system around the rolls" factor which can matter more than the dice probabilities is the range of dice results vs. the range of potential modifiers.  Which is kind of what the "fighter vs. kobold" discussion is getting at.  The reason the kobold has a significant chance of out-muscling the STR 20 fighter is because the die roll has a range of 20 points, but the difference in their STR modifiers is only 9 points.  If, instead, you added the actual STR scores, then the fighter is rolling d20+20 and the kobold is rolling d20+6, a 14-point difference in their modifiers, which makes the fighter far more likely to win.

And then you've got games like Ars Magica, which resolves everything on d10 rolls and it's not that uncommon to see characters with modifiers of +20 or +30 on some rolls, which completely outstrips the range of randomness and guarantees success at simpler tasks, or victory in opposed rolls against marginally-skilled (say, +10) opponents.  Some people dislike games which do this, usually out of ideas of "fairness" or "it should be possible for anything to happen", but it does neatly resolve things like "do you have to roll every time a skill comes up, or do you automatically succeed barring exceptional circumstances?" or "how the hell can a STR 6 kobold out-arm-wrestle a STR 20 fighter?" by allowing for situations where the modifiers are big enough to only allow one possible outcome.

mightybrain

  • Sr. Member
  • ****
  • m
  • Posts: 454
Probability Theory and You
« Reply #31 on: May 11, 2020, 07:29:13 AM »
Quote from: jhkim;1129539
The strongest possible human loses a contest of strength to an average kobold 10% of the time? That really stretches my suspension of disbelief.

That's not the situation described. The kobold has about 10% chance of reducing the strong human's speed to 0 for about 6 seconds. And even then, the human still has about a 90% chance of escaping the grapple and moving anyway.

This is grappling, not a test of raw strength. If you get tipped off balance it's not your strength preventing you from moving.

For a test of raw strength (who can lift the biggest boulder) you could use the strength × 30 rule. (Or strength × 15 for a kobold.)

mightybrain

  • Sr. Member
  • ****
  • m
  • Posts: 454
Probability Theory and You
« Reply #32 on: May 11, 2020, 07:53:14 AM »
Quote from: nDervish;1129542
"do you have to roll every time a skill comes up, or do you automatically succeed barring exceptional circumstances?"

The advice in the 5e Player's Handbook is:
Quote
The DM calls for an ability check when a character or monster attempts an action that has a chance of failure. When the outcome is uncertain, the dice determine the results.

Pat
BANNED

  • BANNED
  • Hero Member
  • *****
  • ?
  • Posts: 5252
  • Rats do 0 damage
Probability Theory and You
« Reply #33 on: May 11, 2020, 08:53:20 AM »
Quote from: jhkim;1129539
The strongest possible human loses a contest of strength to an average kobold 10% of the time? That really stretches my suspension of disbelief.

That's another important factor in ability and skill checks -- IRL, many traits or learned skills function in radically different ways, but most systems try to force them all into one or just a few mechanics. Strength checks are one of the classic examples -- there's a reason for weight categories in wrestling and power lifting, and why someone beating a guy who has 20 pounds on them in a bar room brawl is considered so impressive. A relatively small difference in size can can make the difference between an even match and winning all the times, and a tiny kobold should have zero chance against a large man in anything directly strength-related.

It's even hard to lump all skills into one category. Some skills are pretty deterministic -- in the physical sciences or engineering, where you can precisely measure, calculate, and calibrate forces, the odds of a buttress or a bridge collapsing are almost infinitesimally small. OTOH, improvising something in a jungle might be a risk, but an engineer would usually know whether it's a risk or whether the materials available are sufficient. And trying to create or discover something completely new might never succeed, no matter how many times you try, and you might not know that for sure until you've been butting your head against a potential new theory for 30 years. Those are all completely different, and that's just discussing one related set of skills. The breakdown, degree of confidence in the result, chances of success, and swinginess are going to be very different in the behaviorial sciences and the arts.

That's why I'm firmly convinced RPGs generally don't do skills well, and also why I dislike unified mechanics. They're an attempt to force symmetry on a very heterogeneous set of processes.

Steven Mitchell

  • Hero Member
  • *****
  • S
  • Posts: 3770
Probability Theory and You
« Reply #34 on: May 11, 2020, 09:15:02 AM »
I think most games build their skill probabilities under one of two assumptions:

A. Skills are like combat, and represent things that need a roll under pressure.
B. Skills are a separate thing, and are meant to represent anything you do but the most trivial thing.  With possibly separating combat, magic, or other items out under a system more like A.

In the first, the GM is supposed to let people succeed on things they should be able to do when there is no pressure. In the latter, the skills often have wide ranges of both checks and modifiers, such that auto fail and auto succeed is built into the system.

I think either can work just fine, given the tastes of different players.  If what you want is the high excitement, then there really is no need for B.  It's just extra complication getting in the way of getting to the next roll that might fail.  Whereas if you want the game to be more  about the mundane successes building up to something more slowly but inexorably by using the abilities of your character, then A can start to seem a little lacking.  It's taste.

What I don't like is a system that tries to have it both ways, but then the writer of the system doesn't put the work in.  This is yet another time to point out why the 3.* craft system sucks, by the way.  You could sum up a chunk of the 4E and then 5E improvements to WotC D&D as "realized that they had a lot of things attempting style B game poorly and then made the design decision to rip them out root and branch."

A big part of my current home design is trying to bridge that gap intelligently, and give the GM and players tools to manage the distinction.  Part of the solution is that skills have a "tier" rating, such as "Novice" or "Expert", and also have mechanical underpinnings.  The distinction between 4 success layers (fumble, failure, success, and critical) works into it as well, but making "failure" mean something different in tension versus non-tension checks.  Basically, an "Expert" has to fumble (which is very unlikely) in a sustained piece of work requiring multiple checks to fail pulling off an expert-level task.  And even if he does fumble, he's well equipped to deal with it and succeed eventually (not being under pressure).  A group of lesser beings can attempt it, but the cumulative chance of one of them fumbling and blowing the whole thing is likely, making such attempts the domain of the desperate only. If no one gets seriously hurt, they can keep trying, but it is likely to be beating their heads against the wall.  It also helps mechanically that I've built in group dynamics from the ground up, instead of tacking them on after the fact.  Or the GM can rule on it, because the math of the system backs up the likely outcomes anyway, thus saving the hassle for those rare moments when desperate outcomes and no immediate tension actually matter.

Whereas for simpler checks under pressure, the Expert still has a notable edge, but the game assumes that the tension of the moment narrows the gap.  If you are bleeding out on the icy ledge while the yetis attack, you'd rather have the expert healer, but you'll settle for the two mountain climbers with first aid if that's all you can get right now.  Whether such a system would appeal to others, I don't know, but it at least is mechanically supporting the design goals.

mightybrain

  • Sr. Member
  • ****
  • m
  • Posts: 454
Probability Theory and You
« Reply #35 on: May 11, 2020, 11:35:18 AM »
Quote from: jhkim;1129539
if you were dying -- what would be the best choice to save your life? Having a brilliant doctor come see you, or five average people with no medical training? In D&D (in the case 5E), a cleric with Wisdom 18 has a +6 in Medicine. That's an 85% chance of success of a DC10 Medicine check. But five people with +0 have a 98% chance to get a success among them.


I would consider that a case of Working Together from the Player's Handbook. That is, the one with the highest skill could roll with advantage if at least one other took the Help action. But it wouldn't increase the advantage with more than one helping. That would still tip the balance back to the solo trained healer in this example. (But it would be better, both for probability and thematically, to have a skilled healer and an assistant.) If the players insist they want to make a Group Check with all five, then there's a rule for that too, but the rule is that at least half the group have to succeed. It does slightly improve their chances over a single check but not as much as a roll with advantage from a helper. And I think any DM would be within their rights to impose disadvantage due to too many cooks.

In the context of combat, what we are talking about is roughly the amount of first response first aid you can give in about 6 seconds. What can you seriously expect to do in that time except roll someone on their side and check they haven't swallowed their own tongue? It's not surgery.

If it wasn't a combat situation (like a disease or similar) and there wasn't a specific rule for it, then I'd go with something like the festering wound rules - i.e. you can only receive 1 medicine check every 24 hours and you need some number of successes to recover. In that case, you are much better off going with a better doctor: both in overall recovery probability and recovery time.
« Last Edit: May 11, 2020, 05:43:17 PM by mightybrain »

VisionStorm

  • Hero Member
  • *****
  • Posts: 2184
Probability Theory and You
« Reply #36 on: May 11, 2020, 11:35:56 AM »
Quote from: Pat;1129548
That's why I'm firmly convinced RPGs generally don't do skills well, and also why I dislike unified mechanics. They're an attempt to force symmetry on a very heterogeneous set of processes.


I disagree about unified mechanics for several reasons:

1: I have yet to seen a single RPG with non-unified mechanics that not only also failed to adequately represent these sort of outliers in skill rolls, but that didn't additionally over complicate the entire system by providing widely disparate mechanics to handle everything.  All they accomplish is to make mechanics an inconsistent mess.

Old D&D had non-unified mechanics, not because they worked better to handle these details, but because RPGs were still new and IMO the designers where making things up as they went along, so they didn't think to handle them under a unified mechanic. Other systems did it first. Then when D&D finally did it by 3e it never went back or get revised in later editions because it simply worked better than old D&D ever did. Unified mechanics were the fix for the inconsistent mess that old D&D non-unified mechanics used to be.

2: Skill rolls in most games usually represent ability tests during risky situations of high uncertainty, like the middle of combat or working with inadequate time or materials. No system ever calls for you to make rolls to perform routine tasks. To handle more deterministic tasks, like physical sciences or engineering, you could just assign a minimum skill level to attempt certain actions and require a skill roll only during uncertain situations, or if the character is attempting to accomplish something special.

3: It's a game. That doesn't mean that therefore things don't have to make sense, ever. But it does mean that sometimes you're gonna have to make compromises to make things feasible within the context of the game rules. The reason why these points you mention seem to suck in terms of game mechanics is because, IMO, they're impossible to adequately represent in terms of game mechanics and every game is gonna suck at representing these things in a "realistic" matter. But at least with unified mechanics you don't have to over complicate the system just so that you can still utterly fail to realistically represent this wide range of variability in task resolution.

mightybrain

  • Sr. Member
  • ****
  • m
  • Posts: 454
Probability Theory and You
« Reply #37 on: May 11, 2020, 12:17:52 PM »
Quote from: Pat;1129548
a tiny kobold should have zero chance against a large man

In 5e a tiny creature cannot grapple a large creature as it would be more than one size category larger. However, in their size categories, a kobold is small not tiny, and a man is medium not large.

In general I'm in favour of systems that connect size and strength or weight and strength in some way. However, that only really makes sense within a race. As a counter example consider the chimpanzee; about the same size as a kobold and yet typically stronger than a man.

Pat
BANNED

  • BANNED
  • Hero Member
  • *****
  • ?
  • Posts: 5252
  • Rats do 0 damage
Probability Theory and You
« Reply #38 on: May 11, 2020, 01:09:55 PM »
Quote from: VisionStorm;1129559
I disagree about unified mechanics for several reasons:

1: I have yet to seen a single RPG with non-unified mechanics that not only also failed to adequately represent these sort of outliers in skill rolls, but that didn't additionally over complicate the entire system by providing widely disparate mechanics to handle everything.  All they accomplish is to make mechanics an inconsistent mess.
Skill systems with unified mechanics also fail to represent those outliers, so that's not a valid criticism. And your second critique that they necessarily make an inconsistent mess isn't true. To provide a positive example, consider B/X D&D: The d20 attack roll is different from the generic skill (or "do things in the dungeon") d6 roll, which is different from the % thief skills, and that's different from the 2d6 morale roll. Contrast that with the d20 system, where skills, saves, spell resistance, and knocking down doors, attacks, and saves all use the same basic d20 + mods vs. a target number mechanic.

Yet in D&D3, I always had a hard time remembering how spell resistance worked, because it looked superficially like the more common d20 rolls, but it had very different parameters (caster level instead of spell level like saves, etc.). It's apparent similarity to the other mechanics is exactly what made it hard to remember, because they made it look too much like other subsystems. Conversely, in B/X D&D, it was always easy to remember what I needed to roll for opening a door or attacking an orc, because the heterogeneous resolutions methods used distinct visual and tactile mnemonics. When you have mechanics that are substantially different, they should look and feel different; when mechanics with fundamental differences look the same, it makes them harder to keep straight.

Quote from: VisionStorm;1129559
Old D&D had non-unified mechanics, not because they worked better to handle these details, but because RPGs were still new and IMO the designers where making things up as they went along, so they didn't think to handle them under a unified mechanic.
Nonsense, OD&D drew from a wargaming tradition more than a century old. Gygax did have a preference for creating new systems for new situations, but the rest of that is just making up stuff to confirm your own bias.

Quote from: VisionStorm;1129559
No system ever calls for you to make rolls to perform routine tasks. To handle more deterministic tasks, like physical sciences or engineering, you could just assign a minimum skill level to attempt certain actions and require a skill roll only during uncertain situations, or if the character is attempting to accomplish something special.
Plenty of systems do have modifiers for trivial tasks, which result in a non-zero chance of failure. And the rest of that is you making up new rules. And if have to make up new rules to prove the existing rules work, that suggests they don't.

Quote
3: It's a game. That doesn't mean that therefore things don't have to make sense, ever. But it does mean that sometimes you're gonna have to make compromises to make things feasible within the context of the game rules. The reason why these points you mention seem to suck in terms of game mechanics is because, IMO, they're impossible to adequately represent in terms of game mechanics and every game is gonna suck at representing these things in a "realistic" matter. But at least with unified mechanics you don't have to over complicate the system just so that you can still utterly fail to realistically represent this wide range of variability in task resolution.
I agree with your basic premise, though I slightly disagree with some of the conclusions. Games are imperfect abstractions, and have to focus on what's important or interesting. But there's quite a range between my "awareness of some of these flaws could make existing systems better" and your dismissing it entirely because it can't match some perfect representative ideal.

Steven Mitchell

  • Hero Member
  • *****
  • S
  • Posts: 3770
Probability Theory and You
« Reply #39 on: May 11, 2020, 01:43:55 PM »
Unified systems are useful only so much as they are not pushed too far.  Exactly what is "too far" depends somewhat on design intent, and thus can't be entirely critiqued in a vacuum. There is nothing that says that a game system needs to have a single system throughout to use unified systems (note plural), however.  Rather, it just needs to settle on a few clear systems that are used appropriately.  

Which is to say that I agree somewhat with Pat and somewhat with Vision Storm on this question. There are multiple good and many more bad ways to model a particular thing.  There are things that probably should or should not be modeled.  There are games with different lists of what should or should not be modeled.  There are things that are theoretically or superficially useful to model, that upon testing or practical use turn out to not be worth the complexity costs in having them.  So I think the argument can only go so far in the abstract.

Pat
BANNED

  • BANNED
  • Hero Member
  • *****
  • ?
  • Posts: 5252
  • Rats do 0 damage
Probability Theory and You
« Reply #40 on: May 11, 2020, 02:23:25 PM »
Quote from: Steven Mitchell;1129575
Which is to say that I agree somewhat with Pat and somewhat with Vision Storm on this question.
Where do you disagree with me? Because in my original post, the only thing I said was that I disliked unified systems. Vision Storm posted a followup that assumed I said all kinds of things I never said, and created a lengthy rebuttal based on those imaginary fabrications. But to make it clear, I didn't say anything that Vision Storm implied, nor do I believe any of that crap. It's quite literally a strawman position, created solely to be knocked down, and in no way represents what I believe.

The reason I'm posting this is because, when someone makes up a lengthy rebuttal based on false premises like that, people casually reading a thread seem to frequently assume the original poster actually holds the strawman position. Even when the original poster makes a later post, as I did, disclaiming it.

I read your post, and don't see anything that disagrees with anything I said. Your position does not seem to be halfway between mine, and Vision Storm's. Which suggests you're ascribing to me the position Vision Storm created, rather than responding to what I actually said. And since I have zero interest in defending a strawman position I don't hold, I want to put a stop to that.

Steven Mitchell

  • Hero Member
  • *****
  • S
  • Posts: 3770
Probability Theory and You
« Reply #41 on: May 11, 2020, 02:36:52 PM »
Quote from: Pat;1129580
Where do you disagree with me? Because in my original post, the only thing I said was that I disliked unified systems. Vision Storm posted a followup that assumed I said all kinds of things I never said, and created a lengthy rebuttal based on those imaginary fabrications. But to make it clear, I didn't say anything that Vision Storm implied, nor do I believe any of that crap. It's quite literally a strawman position, created solely to be knocked down, and in no way represents what I believe.


Quote from: Pat;1129548
...
It's even hard to lump all skills into one category. Some skills are pretty deterministic -- in the physical sciences or engineering, where you can precisely measure, calculate, and calibrate forces, the odds of a buttress or a bridge collapsing are almost infinitesimally small. OTOH, improvising something in a jungle might be a risk, but an engineer would usually know whether it's a risk or whether the materials available are sufficient. And trying to create or discover something completely new might never succeed, no matter how many times you try, and you might not know that for sure until you've been butting your head against a potential new theory for 30 years. Those are all completely different, and that's just discussing one related set of skills. The breakdown, degree of confidence in the result, chances of success, and swinginess are going to be very different in the behaviorial sciences and the arts.

That's why I'm firmly convinced RPGs generally don't do skills well, and also why I dislike unified mechanics. They're an attempt to force symmetry on a very heterogeneous set of processes.


You seem to be implying that unified mechanics implies a single, uniform mechanics for the entire system. My argument is that one doesn't need to go that far to get some of the benefits of unified mechanics. Regardless of how many different types of skill behaviors there are, there is still a finite limit when considering the reasonable range of modeling in games.  (That is, not perfect, but good enough to avoid some of the issues you are stressing in the paragraph above.)  Specifically, it is possible to do some form of unified mechanics without forcing symmetry on different things.  

Where I side with you on the remark above and disagree with Vision Storm is that I think the practical outcome of pushing that theory in most cases is very much likely to be exactly the problem of forcing symmetry where it should not be.  But my objection is more about lazy and thoughtless designers than an inherent problem with unified systems.

Pat
BANNED

  • BANNED
  • Hero Member
  • *****
  • ?
  • Posts: 5252
  • Rats do 0 damage
Probability Theory and You
« Reply #42 on: May 11, 2020, 03:07:03 PM »
Quote from: Steven Mitchell;1129582
You seem to be implying that unified mechanics implies a single, uniform mechanics for the entire system. My argument is that one doesn't need to go that far to get some of the benefits of unified mechanics. Regardless of how many different types of skill behaviors there are, there is still a finite limit when considering the reasonable range of modeling in games.  (That is, not perfect, but good enough to avoid some of the issues you are stressing in the paragraph above.)  Specifically, it is possible to do some form of unified mechanics without forcing symmetry on different things.  

Where I side with you on the remark above and disagree with Vision Storm is that I think the practical outcome of pushing that theory in most cases is very much likely to be exactly the problem of forcing symmetry where it should not be.  But my objection is more about lazy and thoughtless designers than an inherent problem with unified systems.
The d20 system is usually considered a unified system, yet things like the miss chance for concealment and the damage roll don't use the d20 + modifiers vs. a target number. Except for the simplest systems, like storygames with a very narrow focus, I can't think of a single game with a single system for everything. There are always exceptions, so it's not reasonable to assume I'm talking such a hypothetical extreme. It's like something saying they like rules light systems, and someone replying that you need more than one rule to make a game. That's not a good argument, because a game with a single rule isn't within the normal range of expectations. Same applies to unified systems vs. less unified systems.

But the d20 system is a unified system in that a lot of mechanics that work very differently are forced into a similar-seeming d20 roll. For instance, attack rolls operate differently than skills operate differently than saves operate differently than spell resistance, but they all use the same dice roll, add modifiers, and compare them to a target number. That's an example of the kind of unified system I have a problem with, because it's masking some very different things by making them look alike. It's just good user interface design to give reinforcement signals when you change modes, and having different dice or a distinct method of reading them is a good way to do it in RPG design. Conversely, unified mechanics have no real virtue in themselves. Yes, there are benefits to simplicity and grouping similar things, but that's not generally what anyone means when they say "unified mechanic".

Also, note that my mention of unified mechanics in the original post was a quick aside. The bulk of the text you quoted is talking about skills, not unified mechanics. I drew a general comparison to unified mechanics at the end, but that's it.
« Last Edit: May 11, 2020, 03:11:39 PM by Pat »

Trond

  • Hero Member
  • *****
  • Posts: 2743
Probability Theory and You
« Reply #43 on: May 11, 2020, 03:28:17 PM »
Quote from: VisionStorm;1129559
I disagree about unified mechanics for several reasons:

1: I have yet to seen a single RPG with non-unified mechanics that not only also failed to adequately represent these sort of outliers in skill rolls, but that didn't additionally over complicate the entire system by providing widely disparate mechanics to handle everything.  All they accomplish is to make mechanics an inconsistent mess.

Old D&D had non-unified mechanics, not because they worked better to handle these details, but because RPGs were still new and IMO the designers where making things up as they went along, so they didn't think to handle them under a unified mechanic. Other systems did it first. Then when D&D finally did it by 3e it never went back or get revised in later editions because it simply worked better than old D&D ever did. Unified mechanics were the fix for the inconsistent mess that old D&D non-unified mechanics used to be.

2: Skill rolls in most games usually represent ability tests during risky situations of high uncertainty, like the middle of combat or working with inadequate time or materials. No system ever calls for you to make rolls to perform routine tasks. To handle more deterministic tasks, like physical sciences or engineering, you could just assign a minimum skill level to attempt certain actions and require a skill roll only during uncertain situations, or if the character is attempting to accomplish something special.

3: It's a game. That doesn't mean that therefore things don't have to make sense, ever. But it does mean that sometimes you're gonna have to make compromises to make things feasible within the context of the game rules. The reason why these points you mention seem to suck in terms of game mechanics is because, IMO, they're impossible to adequately represent in terms of game mechanics and every game is gonna suck at representing these things in a "realistic" matter. But at least with unified mechanics you don't have to over complicate the system just so that you can still utterly fail to realistically represent this wide range of variability in task resolution.


FWIW I agree with all of the above. I think the biggest mess was actually 1st ed AD&D, but that's just me. Runequest was much better at that time. None of these is perfect, but you'll never get that in a game anyway. But a unified system for skills and maybe stat rolls tend to smooth out a lot of things in RPGs IMO.

VisionStorm

  • Hero Member
  • *****
  • Posts: 2184
Probability Theory and You
« Reply #44 on: May 11, 2020, 03:29:00 PM »
Quote from: Pat;1129570
Skill systems with unified mechanics also fail to represent those outliers, so that's not a valid criticism.


Except that the criticism is that you're singling out one approach as the root of the problem when in reality both approaches fail. And if both approaches fail then the one you're singling out can't be the root of the problem--the issue must be something else. My position, beyond that base criticism is that if both approaches are doomed to fail regardless then the most simple, straightforward and/or effective (in terms of simulation) approach should be the one used (or at least the one preferable in most circumstances).

Quote from: Pat;1129570
To provide a positive example, consider B/X D&D: The d20 attack roll is different from the generic skill (or "do things in the dungeon") d6 roll, which is different from the % thief skills, and that's different from the 2d6 morale roll.


That's exactly what I meant by inconsistent mechanics. Every single time you want to do something in the system you have to use a completely different set of mechanics that often doesn't account for character abilities (just rolls a 1 or 2 on a d6) or address any of the issues you were pointing out that unified mechanics fail to address. They're just different almost for the sake of being different, and still don't simulate reality close to 1/1. If anything they simulate it even less by often not accounting for character ability at all.

Quote from: Pat;1129570
Contrast that with the d20 system, where skills, saves, spell resistance, and knocking down doors, attacks, and saves all use the same basic d20 + mods vs. a target number mechanic.

Yet in D&D3, I always had a hard time remembering how spell resistance worked, because it looked superficially like the more common d20 rolls, but it had very different parameters (caster level instead of spell level like saves, etc.).


That's an extremely minor issue (what's the huge difference between a caster level and a combat modifier or a skill level?), and not not even an issue with unified mechanics, but an issue with D&D 3e+ mechanics not being unified enough. If spell casting (along with every other task in the game) was handled as a skill instead of relying on character classes or dragging around these carryover mechanics from older editions like caster levels, THAC0/Combat Modifiers and whatnot that wouldn't be an issue. A spell resistance roll would just be a magic skill roll using the opponent's magic resistance as difficulty. An attack would just be a skill roll against the opponent's defensive skill. A saving throw would just be a skill roll against the negative condition's difficulty, etc.

This is mostly how 5e does it, by the way, which is also the most successful edition in D&D history. I wonder if those two things are related (and I doubt throwing a bone to the OSR had much to do with it cuz most people who play 5e probably don't know WTF B/X D&D even is--they play it cuz it's finally easy enough for normies to grasp).

You're just pointing out minor inconsistencies that are an issue precisely because they're inconsistent rather than being truly unified mechanics.

Quote from: Pat;1129570
Nonsense, OD&D drew from a wargaming tradition more than a century old. Gygax did have a preference for creating new systems for new situations, but the rest of that is just making up stuff to confirm your own bias.


A tradition that dealt with a different style of game that didn't involve in-depth incursions into a simulated world, where stuff like character skills and handling a broad range of tasks becomes relevant.

Quote from: Pat;1129570
Plenty of systems do have modifiers for trivial tasks, which result in a non-zero chance of failure.


In my experience those are usually included for the sake of completeness, given that those difficulty modifiers may become relevant in situations where a skill roll might still be necessary, such as when trying otherwise trivial tasks in combat, or when complications that may penalize an otherwise extremely low difficulty roll are in effect. But those same systems usually include a note somewhere that you don't need to roll every single time a task that might be covered by a skill comes up. You only need to roll when dramatically appropriate. This is pretty much standard fare in RPGs.

Quote from: Pat;1129570
And the rest of that is you making up new rules. And if have to make up new rules to prove the existing rules work, that suggests they don't.


Rules that don't exist in systems with non-unified mechanics either, because they deal with specialized circumstances that are hard to incorporate into the game. Except that if I was going to make them up for a skill-based system using unified mechanics I could make them up on the fly by simply assigning a minimum skill/ability level as a requirement. With non-unified mechanics I would either need to fall back on a minor concession like "roll 2 or less on a d6 and forget about your character's abilities affecting the roll" or make the whole thing up from scratch.