I once thought that having a skill system in a class-based D&D could work, it just hadn't yet.  Now, having tried to "fix" skills in 3E and having dealt with them in 4E and 5E--and not been impressed with the options prior to 3E, such as the RC optional proficiencies--now I'm finally convinced they don't work.  I know some of you said that earlier, but some things a person has to learn for themselves.

I love skills-based games, just not in my D&D.

I don't really get how skills don't work with class systems or in 5e in particular--other than perhaps the list not being perfect or well-defined, which can always be homebrewed or adapted to the setting. But mechanically, 5e's skills are the most simple and straightforward of any edition, and work exactly like an attack, save or any other ability check. That’s pretty much one of the savings graces of the system for me.

Other than S'mon's mention of Medicine being useless as written (which I see as an issue of implementation, rather than poor core mechanics), I haven't really seen any clear examples of how exactly skills fail, esp. not as a concept.

Quote from: VisionStorm;1143438I don't really get how skills don't work with class systems or in 5e in particular--other than perhaps the list not being perfect or well-defined, which can always be homebrewed or adapted to the setting. But mechanically, 5e's skills are the most simple and straightforward of any edition, and work exactly like an attack, save or any other ability check. That's pretty much one of the savings graces of the system for me.

Other than S'mon's mention of Medicine being useless as written (which I see as an issue of implementation, rather than poor core mechanics), I haven't really seen any clear examples of how exactly skills fail, esp. not as a concept.

It's a difficult thing to see from a theory perspective.  Not saying this is comprehensive, but part of it for me was realizing that there aren't many (or maybe even any) large problems with skills in D&D, but there are a series of smaller pitfalls.  Only trouble is that in practice, the smaller things combine into some unholy mess worse than you would expect from the individual pieces.  For example:

- Magic isn't skills based.  It leaves a gap in the game that monkeys with  the distribution of skills versus spell casting.  
- At the broad scope at which D&D classes operate, there just aren't that many relevant skills.  Not even as many as 5E uses.
- Perception as a skill in D&D is all kinds of screwed up in a way that probably deserves its own topic (i.e. it's complicated and will be a wide-ranging discussion of multiple parts of game design)--and that's assuming that one is in the group that thinks "perception" should even be a thing.

More broadly, skills and improvements to them operate with a different mix of breadth and depth compared to class mechanics.  It's easier to think about this in a skill-based game.  Take GURPS for example.  You can have a campaign in GURPS with point totals selected, and then put in min and max ranges at which characters are expected to operate.  Depending on where you set those ranges, you'll get a different kind of game.  Set them low enough, you'll encourage more breadth.  Set them higher, you'll encourage specialization.  And so on.  With D&D, the classes already set the depth expectations but will also have an effect on the breadth of the characters as well. 5E kind of, sort of, works OK in this regard because we know the rogue is going to be able to sneak around, the bard can win friends, etc.  So it doesn't matter all that much that Medicine doesn't do a lot.  

It's like a cheap but functional pair of shoes.  It's not ideal, but not entirely cramping your feet the moment you put them on.  Wear them all day long while standing on your feet at a presentation, you start to notice you need a better pair.  That's how 5E skills function.

Quote from: Steven Mitchell;1143450- Magic isn't skills based. It leaves a gap in the game that monkeys with the distribution of skills versus spell casting.
- At the broad scope at which D&D classes operate, there just aren't that many relevant skills. Not even as many as 5E uses.
- Perception as a skill in D&D is all kinds of screwed up in a way that probably deserves its own topic (i.e. it's complicated and will be a wide-ranging discussion of multiple parts of game design)--and that's assuming that one is in the group that thinks "perception" should even be a thing.

Out of these three points I relate with the first one the most because it's an issue of mine with D&D as well, in that I would prefer a skill-based magic system. But that's an issue of D&D's magic system rather than a problem with skills per se. It would also be almost an entire topic to itself about how to handle magic in general and using skills in particular.

I'm not sure what you mean by the second point, and I'm not sure to what extent classes impact the number of applicable skills in the game. Though, I could probably reduce skills into even broader categories of activities than 5e's current list, and in a way I already have in my own system, which uses broad skills (or "disciplines") as a basis for all actions, that are more general than 5e skills. But I'm not sure if you're talking about making skills more general or something else.

As for the third point, my issue with Perception in 5e (and even earlier editions of D&D) is that there are way too many variations of what's essentially the ability notice things. This hankers back to earlier editions of the game where the ability to merely notice a trap (and nothing else) was its own class-exclusive ability. Then this was turned into the "Search" skill in 3e, along with a separate Spot skill to notice sneaky characters (which is apparently a completely separate function from noticing things through "Search"), yet another Listen skill also to notice sneaky characters (cuz that apparently takes two fucking skills) and yet a fourth Sense Motive skill to notice when people are full of shit (the only one of these functions that arguably deserves a separate skill IMO, and even then that's just a subset of noticing things, which could just all be covered by a single skill).

By 5e notice skills have been morphed into Investigation, Insight and Perception. Out of these three skills the only one I would use for sure is Perception. I have read nuanced explanations about the distinction between Investigation and Perception, but frankly that's just bullshit. It doesn't take a completely and utterly separate skillet to notice things in passing vs noticing them by actively searching. That's just the same thing. Insight at least is arguably distinct from Perception/Investigation, since it relies more on intuition rather than physically spotting things. But even then that's just a subset of noticing things, as I mentioned in the paragraph above. Which is why I folded all of those functions into a single Perception discipline in my own system, and treated Intuition (Insight) and Observation (Perception/Investigation) as specialties.

Quote from: VisionStorm;1143499Out of these three points I relate with the first one the most because it's an issue of mine with D&D as well, in that I would prefer a skill-based magic system.

There actually are skill based magic systems for D&D 3e . But I ended up preferring a 'trait' based system.

Love me some 5E bound accuracy and advantage/disadvantage, but it took me a while to put my finger on what bothered me about skills.

I like how skills in 5E are "distilled" into essential parts, but the issue is that when designing an encounter (be it combat or otherwise), I can engineer possible skill checks that make sense. Here a PC can do an Intelligence (History). Here's the DC table. PCs don't have History? They can still do straight-up Int. That's why it's Intelligence (History). Etc.

However, when I'm winging it because my players are being players, like 90% of the time at my open-world table, coming up with a table in my head doesn't work. Thus the entire skill check mechanic for me falls flat. I've played with the universal DC rates of easy, medium, hard, nearly impossible, but find myself wishing there was a better way.

Put me also in the I love skills-based games, just not in my D&D camp.

Quote from: Shrieking Banshee;1143509There actually are skill based magic systems for D&D 3e . But I ended up preferring a 'trait' based system.

Yeah, I'm aware of a handful of alternatives, including some that use the existing D&D spell casting system, but replace spell slots with skill checks vs a DC based on the spell's level to determine whether the character manages to cast the spell (I think Lion & Dragon uses a similar mechanic, but I got the one I've used somewhere else). I also like the idea of tying access to different types of magic to feats (or traits, advantages, whatever), and I tend to think of skills and feats as being interrelated, but that's a whole other can of worms about alternate magic systems and how to best handle magic through skills, comparable to the way how other types of activities can also be handled through skills.

Quote from: Anthony Pacheco;1143514However, when I'm winging it because my players are being players, like 90% of the time at my open-world table, coming up with a table in my head doesn't work. Thus the entire skill check mechanic for me falls flat. I've played with the universal DC rates of easy, medium, hard, nearly impossible, but find myself wishing there was a better way.

How difficult can it be to remember to use DC 15 for all rolls by default, then adjust that by increments of 5 up or down (10, 15, 20, etc.) if necessary? That's already the simplest way it could possibly be, and it's easier to eyeball than an opponent's AC or save DC, which almost invariably require some calculation, and would always be used even if you somehow managed to get rid of skills in the game.