D&D 5E Some thoughts on skills.

[Swarmkeeper]

I agree with you about a living world. (Also, I don't know why some DMs seem to think having chests the PCs can't open, secret doors they can't find, lore and history they don't know, and cool items they can't identify or use, makes for a better game experience. )

However, I do think it is partly a system issue, since (in theory) it doesn't happen in Pathfinder because all* the DCs are written into the rules.

It seems, from looking at 5e conversions of Pathfinder adventures, that you get reasonable results if you take the Pathfinder DC (available for free at Archives of Nethys) and reduce it by 5.

Of course, no system can force the DM to actually use the rules in the first place. (I'm as guilty as anyone of estimating the DCs in Pathfinder rather than taking the time to look them up, but I've been playing the game for a long time so my estimates are usually close enough.)


*obviously not every possible DC, but all the ones that come up with any regularity and you can use them to extrapolate the rest

In defense of 5e, rather than giving us a bunch of DC charts to refer to, they keep it simple by giving guidance on how to set a DC for any particular task:

Difficulty	DC
Very Easy	5
Easy	10
Moderate	15
Hard	20
Very Hard	25
Nearly Impossible	30

They even advise:
If the only DCs you ever use are 10, 15, and 20, your game will run just fine.

Then they had to go ahead and muddy the waters with this language:
If you find yourself thinking, "This task is especially hard," you can use a higher DC, but do so with caution and consider the level of the characters.

The "caution" part is good but the "consider the level of the characters" could be misleading people down the path of DC creep which @ph0rk mentions above.

Also, obligatory mention of: "no one reads the DMG anyway"

 

[Pedantic]

In defense of 5e, rather than giving us a bunch of DC charts to refer to, they keep it simple by giving guidance on how to set a DC for any particular task:

I've concluded that generic difficulty tables are just not effective. They don't provide sufficient constraints on what kind of task falls into each category, and they only really work in games like 5e that have a super small range of DCs to begin with. The best use case for them is the most disempowering one, setting a fixed DC native to the environment or challenge itself, instead of the task being attempted (i.e. this is a level 3 dungeon, so checks are DC 13), because that just encourages players to consistently argue for whatever number on their sheet is highest, instead of trying to leverage a specific ability against the environment.

 

[tetrasodium]

I've concluded that generic difficulty tables are just not effective. They don't provide sufficient constraints on what kind of task falls into each category, and they only really work in games like 5e that have a super small range of DCs to begin with. The best use case for them is the most disempowering one, setting a fixed DC native to the environment or challenge itself, instead of the task being attempted (i.e. this is a level 3 dungeon, so checks are DC 13), because that just encourages players to consistently argue for whatever number on their sheet is highest, instead of trying to leverage a specific ability against the environment.

5e doesn't have a small range though, it goes from level1 possible +0/ to +7 with disadvantage all the way to +18 with advantage and an extra d4 at higher levels. That's almost as wide of a range as 3.x had if not wider once you factor (dis)advantage & the d4.

 

[Quickleaf]

At least according to the poll I put up recently, Skills are the single most agreed upon mechanic for D&D. Thinking about the way 5E uses skills, I would like to propose some ideas for how I think they could be better utilized in the game.
1) Skills need to be defined as a specific thing, not treated as an afterthought of an ability check.
2) Skills need to be disconnected from ability scores. there is already a rule that makes this possible, but it needs to be explicit.
3) there should be a penalty (disadvantage? -2?) for using any skill untrained.
4) Skills need to inherently outweigh ability scores on the d20 roll.
5) the skill list needs to be expanded and more skill points need to be provided to all characters.
6) Backgrounds should determine how many base skill points you get, modified by class.
7) Tool proficiencies need to go away and having or not having the right tools should modify skill checks.

In addition, this is a more general rule, but I think advantage and disadvantage should stack from different sources and cancel each other out on a one for one basis, AND if more than 1 "level" of advantage or disadvantage remains, you roll multiple additional dice and take the bester/worst of all of them.

EDIT: For clarity, I am "liking" posts because I am glad people are engaged in the topic, not necessarily becaaue I agree with you lot.

I like the first part of your number 5 (expanded skill list) and the first part of your number 7 (removing tool profs / fold them into skills).

I'd much rather some deeper design work be done around the foundation and ecosystem in which skills exist. I think more direction regarding the following would be good...

• When does the GM call for a check?
• What exploration or social rules & procedures can this skill reference?
• How does this address pile-on checks in the rules?
• Does a group check make sense for this skill?
• Should a non-proficient character even be able to use this skill let alone Help someone else with it?
• What are the consequences of a failed skill check? For instance, how is a "complicated success" (i.e. yes, but) or a "devastating failure" determined by the GM with this skill system?

Many of these are learned by the GM "on the job", so to speak, but if the goal is to encourage a new generation of GMs, I think getting into these grey areas (whether rulebooks, videos, whatever) is a smart move.

 

[Kinematics]

1) Skills need to be defined as a specific thing, not treated as an afterthought of an ability check.
2) Skills need to be disconnected from ability scores. there is already a rule that makes this possible, but it needs to be explicit.
3) there should be a penalty (disadvantage? -2?) for using any skill untrained.
4) Skills need to inherently outweigh ability scores on the d20 roll.
5) the skill list needs to be expanded and more skill points need to be provided to all characters.
6) Backgrounds should determine how many base skill points you get, modified by class.
7) Tool proficiencies need to go away and having or not having the right tools should modify skill checks.

#1: So skill first, ability second, rather than ability first, skill second? It is, admittedly, the way that most players view things, even if the rules expect differently.

#2 I agree with in principal, but it becomes unwieldy in practice. Perhaps if it was presented as, "This ability score is the most common to use with this skill, so record it as the default for quick reference. Otherwise add the specified ability score, and your proficiency bonus if you have proficiency in the skill." Make it so that it's still easy to use the pre-calculated values, but make sure it doesn't imply that those are the only combinations.

#3 is just the lack of any bonus, as others have said.

#4 needs more thought on the mathematical implications.

#5 I broadly agree with, and I'd combine it with #7. Having proficiency in one of these skills allows you to use any of the noted tool sets. Not having the tool set still allows you to use the skill, but you may be more limited in what you can do, or have a higher DC, depending on your intended action.

• Craft: Smithing, tinkering, woodworking, carpentry, leatherworking, glassblowing, jewelery, pottery, masonry, weaving.
• Alchemy: Cooking, brewing, herbalism, (making) poisons, medicine.
• Expression: Calligraphy, cartography, painting, decorating, sculpting.
• Skullduggery: Lockpicking, trapmaking, forgery, disguise, (administering) poisons.
• Musician: All instruments.
• Travel: Navigation, vehicles (land or water).

Add in Culture and Engineering from Level Up as well.

Craft (or sometimes Engineering or Skullduggery) without a tool set would be McGuyvering. Decorating (under Expression) wouldn't need a tool set per se, but would need appropriate materials. Etc.

Drop the Gaming Set proficiencies entirely. That's just silly. If you know the rules, you can play the game. If you want to cheat at the game, that's Sleight of Hand or Deception. Maybe there are people out there with expertise in mahjongg, but that's really outside the scope of the skill system.

Though, perhaps give the GM a tool to mark someone as super experienced at a very narrow skill (say, sushi chef, or Three Card Monte hustler), of something like +1 per year of practice in the skill, up to +10. It's rather different than the broad and generic proficiency gain that comes with level.

On #6, I'm not sure. I still feel that backgrounds should generally provide skills on their own. Of course, getting rid of tool/instrument proficiencies opens up space for one or two more skills that each character can have (though they might get directed into the generalized tool skills).

 

[ph0rk]

When does the GM call for a check?

Related to this, a list of things that are doable "checkless" for trained people would be nice.

E.g. picking a normal indoor lock, climbing a tree, etc. I know people that can pick a master padlock in a pretty short amount of time - it shouldn't even require a check in that sort of case, unless the check is to determine how long it takes.

A trained athlete should have no trouble (I mean, would a child?) unless it is a particularly difficult tree to climb.

 

[Shiroiken]

One gripe I have with the existing system is ability+proficiency+expertise leads to DC creep which can lock out characters with medium ability scores and only proficiency from activities.

Bounded accuracy should keep the DC creep out of the equation. If the DC keeps pushing up, this is because the DM is being antagonistic, forcing a treadmill where only the best person can succeed. This was exceptionally common in 3E, but this is not the way 5E is meant to be ran.

 

[Reynard]

. This was exceptionally common in 3E, but this is not the way 5E is meant to be ran.

It shouldn't have been. 3.x had objective DCs for specific actions.

 

[tetrasodium]

Bounded accuracy should keep the DC creep out of the equation. If the DC keeps pushing up, this is because the DM is being antagonistic, forcing a treadmill where only the best person can succeed. This was exceptionally common in 3E, but this is not the way 5E is meant to be ran.

Bounded accuracy needs to apply to both sides. It falls apart & farcically ceases to be bounded in any way with bounded DCs & bounded monsters across from unbounded PCs

 

[Pedantic]

Related to this, a list of things that are doable "checkless" for trained people would be nice.

E.g. picking a normal indoor lock, climbing a tree, etc. I know people that can pick a master padlock in a pretty short amount of time - it shouldn't even require a check in that sort of case, unless the check is to determine how long it takes.

A trained athlete should have no trouble (I mean, would a child?) unless it is a particularly difficult tree to climb.

This is best handled by an effective and well written Take 10/Take 20 rule. It provides clear scaling, and an easier answer to what you can do without trying. Ideally, the take 10/take 20 values should be written on character sheets, and abilities that modify those values should be worked effectively into classes and other features.