[D&D 3.5] Removing The Entire Skill System

[Ziegander]

Skills are problematic in 3rd Edition, heck, they're problematic in just about every edition of D&D I've ever seen. Instead of dealing with mundane tasks that everyone can do, and forcing some granular system onto the players and the DM, making them track little bonuses that, after a few levels, don't add up to anything and whose payoffs are trivially made irrelevant by class features or spells, I thought, well, what if Skills just no longer existed?

"You're mad!" I hear you cry, "You're an inhuman monster!" Well, obviously one cannot wholesale remove some of the abilities that skills allow players to do from the game. If players could no longer Jump or Swim or Search for Traps, then people would be rightfully angry. The game would become much more combat oriented. So, if I removed the Skills system from D&D entirely, what would I replace it with?

Ability checks. Can someone explain to me what is so complex and gamebreaking about any of the D&D skills that it cannot be resolved by a simple ability check?


To successfully Climb, Grapple, Jump, and Swim, the player resolves a Strength check vs some DC.
[br]
To successfully Balance, Ride, Sneak, Tumble, or Use Rope, the player resolves a Dexterity check vs some DC.
[br]
To successfully Concentrate, the player resolves a Constitution check vs some DC.
[br]
To successfully Comprehend Languages and Decipher Scripts, Craft, Disable Devices, Forge Documents, Identify Magic and Spells, or Open Locks, the player resolves an Intelligence check vs some DC.
[br]
To successfully Listen, Tend Wounds, Search, Sense Motives, Spot, or Survive in the Wild, the player resolves a Wisdom check vs some DC.
[br]
To successfully Bluff, Command or Deceive Magic Items, Gather Information, Give a Performance, Handle Animals, Intimidate, or Use Diplomacy, the player resolves a Charisma check vs some DC.


But the main reason for the existence of a granular Skill System, such as one used in D&D 3.5, is that, given two people of equal Strength scores, one might have never climbed before, while one might be an expert climber. An expert climber may have little to no experience grappling or swimming. So, how would I handle such diversity?

It's a good question. There are many ways I might model that sort of real world diversity, but it is more uncertain how well received they might be. The easiest and simplest way would be to say, "I don't and I don't care. D&D is a game, and does not have to be a direct simulation of the real world." But that's mostly a cop-out. Backgrounds are probably the next most simple solution, and could provide a bonus to certain ability checks which scales with level.

For example, the Miner background might grant a +3 competence bonus to strength checks made for climbing, to dexterity checks made for using rope, and to wisdom checks made for searching an area. This bonus might increase by 1 at 2nd level and at every three levels thereafter (up to +10).

The "skill feats" such as Acrobatic and Self-Sufficient could very well still exist, and grant competence bonuses to skills on par with those granted by Backgrounds. So, the Acrobatic feat could grant a +3 competence bonus to strength checks made for jumping and to dexterity checks made for tumbling, and have the bonus scale as above.

Now, what unforeseen effects might such a wide sweeping change have on D&D 3.5 as a whole? Are there any negative consequences to gameplay that I am not accounting for, other than the obvious, "but I'm not trained in lockpicking anymore," ones? Any other thoughts on the idea?

[robertbevan]

your post made me think. it's an interesting idea. i don't have anything to add to it, because i've been drinking quite a bit tonight. but i'll definitely give it another look tomorrow to see what other people said.

[sonofzeal]

The problem with making everything Ability checks is that ability scores... don't really scale very well. At least not relative to the size of a d20.

Example: my Str 10 Halfling Rogue once beat a Str 18 Human Barbarian in arm-wresting, because I rolled a 16 and he rolled a 10. Our rolls weren't even that far apart, but was enough to swing such an implausible imbalance.

Five levels later, that Barbarian might have 21 Str.... and I still would have beaten him. His performance at feats of strength would barely have climbed at all.

Both of those seem like critical flaws. Characters should improve. And characters good in some area should be significantly better than characters who've never thought twice about it.

[Telonius]

You could give each class a certain number of skills to give some bonus to. Barbarian gets four, Bard gets six, etc, and only at first level. Have them pick that number of skills out of their class skill list. (Or out of the total skill list, if you prefer). Bonus could be class level/3. That way, you give each character's a unique feel, but still have the more streamlined "ability score only" feel.

You'll probably run into some problems with PrC prerequisites (and some Feat prerequisites). Many of them have skill prerequisites to prevent anyone from taking them before 5th (or 10th) level.

[Ashtagon]

Example: my Str 10 Halfling Rogue once beat a Str 18 Human Barbarian in arm-wresting, because I rolled a 16 and he rolled a 10. Our rolls weren't even that far apart, but was enough to swing such an implausible imbalance.

Isn't arm wrestling given as an actual example of something where you just compare ability scores without rolling?

[Ziegander]

The problem with making everything Ability checks is that ability scores... don't really scale very well. At least not relative to the size of a d20.

Example: my Str 10 Halfling Rogue once beat a Str 18 Human Barbarian in arm-wresting, because I rolled a 16 and he rolled a 10. Our rolls weren't even that far apart, but was enough to swing such an implausible imbalance.

Five levels later, that Barbarian might have 21 Str.... and I still would have beaten him. His performance at feats of strength would barely have climbed at all.

Both of those seem like critical flaws. Characters should improve. And characters good in some area should be significantly better than characters who've never thought twice about it.

Yes, I know. Did you read the second half of my post? Where backgrounds and feats can make one character significantly better than another in specialized venues? The bonuses may be somewhat too low, but that would be something that's easy to play with.

[Madeiner]

About ability scores and the imbalance versus a d20 curve, i use the following system for ability checks in my game.

You must roll a d20. A result equal or lower to the corresponding ability score is required to succeed.

You want to remember something? Roll d20. If it's lower or equal to your intelligence score, you succeed.

For hard tasks, apply +2 to +6 to the roll. For easy task, apply -2 to -4.
If the task is "hard", you cannot even try to roll if your ability score is less than 10.

In case you ever need a mirror check, like arm wrestling, both participants roll their d20. The one that scores the bigger difference between the roll and their score, wins.

Halfling with STR 10 rolls a 5. He succedes by 5.
Orc with STR 16 roll a 10. He succeedes by 6, so he wins.

Note that i use this system only for things not covered by skills. But it can be adapted for your system too.

At first it sounds "strange". But as you play it's pretty straightforward, and you can notice that no more your INT 8 fighter remembers things almost as easy a wizard.

Heck, once the wizard hits INT 20, he will remember everything that is not particularly hard, every time (bar the natural 20 auto-failure)

Your str 20 barbarian can lift your average portcullis each time, as it should be, and must roll a 16 or less to lift a heavy portcullis.
A str 10 halfling can try to lift the hard portcullis, and succeed on a 6 or less.
However a str 9 character is not even allowed a roll; he automatically fails.

[JellyPooga]

As has been mentioned, Ability Score checks don't scale well in 3.5 and make for poor improvement. Where in older editions, an Ability Check meant rolling under your score on a d20 (hence the 3-18 range for starting scores and why only superhuman scores went above 20), 3rd edition parsed the Ability down into a comparatively small bonus (compared to the range of a d20). 3.5 also made increasing Ability Scores much easier, which really throws out the idea of the "roll under" mechanic, which would be my instinctive reaction to your idea.

I'm not sure what I'd suggest in it's place, but I can recommend that you go back and look at older and alternative editions which deal with "skill checks" in largely the same way that you suggest. If nothing else, it will give you some ideas on how to refine the system.

The biggest problem I can see with the system as a whole, however, is that it really kicks some Characters in the goolies. Rogues, Scouts, Factotums, heck even some Rangers, Ninja's and Spellthieves will all be looking around bewildered wondering quite where they fit in to the system now that their 'Skillmonkey' role has pretty much disappeared. Then there's Races to consider too; Humans, for example, will be left with a bunch of Skill points they're not sure what to do with (because they can't spend them). I'm aware that D&D 3.5 is not the most 'balanced' game out there, but some of the appearance of 'balance' comes from the amount of skill points a class gets. Throwing this out is not something that should be done lightly.

[sonofzeal]

Yes, I know. Did you read the second half of my post? Where backgrounds and feats can make one character significantly better than another in specialized venues? The bonuses may be somewhat too low, but that would be something that's easy to play with.
I read the second half, but it still leaves the vast majority uncovered. A Str 18 should be far more capable at feats of strength. A Dex 18 character shouldn't be failing trivial DC 10 checks a quarter of the time.

Frankly, I don't see any advantage at all of doing it this way. The skill system is imperfect, but I really don't understand what problem you think you're fixing. Small modifiers don't add to much... so let's get rid of the big modifiers too?!?

[Ziegander]

I read the second half, but it still leaves the vast majority uncovered. A Str 18 should be far more capable at feats of strength. A Dex 18 character shouldn't be failing trivial DC 10 checks a quarter of the time.

You're being really vague in your criticism. I don't know what you're trying to say.


Frankly, I don't see any advantage at all of doing it this way. The skill system is imperfect, but I really don't understand what problem you think you're fixing. Small modifiers don't add to much... so let's get rid of the big modifiers too?!?

My point is that D&D 3.5 treats class skills, skill points per level, and skill ranks as a BIG DEAL. When they're not. This causes lots of problems, vast imbalances, sure, but also pointless bookkeeping. So my solution is, get rid of skills, and replace them with something that is acknowledged as being a very small deal. Making and resolving ability checks is not, and is not supposed to be a big deal.

The advantage to this approach is that, classes no longer have to be balanced against each other, based on skill points per level, or class skills. The Knock spell, while possibly being too low level, is no longer such a grievous offender of sensibilities when it invalidates a mere fraction of what an Intelligence check can do, rather than invalidating a whole character concept.

[sonofzeal]

You're being really vague in your criticism. I don't know what you're trying to say.
Under your system, unless someone has a specific background like "Miner", a Str 20 hero is only marginally more competent as strength-based activities than a Str 10 commoner. I want my heroes to be skillful and exceptional and be able to do things well within their specialties. If I have a strong character, I want him to be mighty. If I have a dextrous character, I want them to be lithe. Flubbing skillchecks like a total newbie takes away from that considerably, but that's what they're left with under your system.




My point is that D&D 3.5 treats class skills, skill points per level, and skill ranks as a BIG DEAL. When they're not. This causes lots of problems, vast imbalances, sure, but also pointless bookkeeping. So my solution is, get rid of skills, and replace them with something that is acknowledged as being a very small deal. Making and resolving ability checks is not, and is not supposed to be a big deal.

The advantage to this approach is that, classes no longer have to be balanced against each other, based on skill points per level, or class skills. The Knock spell, while possibly being too low level, is no longer such a grievous offender of sensibilities when it invalidates a mere fraction of what an Intelligence check can do, rather than invalidating a whole character concept.
But now magic becomes even more essential because the non-magic solutions are now worthless.

Skills aren't a huge deal. But then again, the system's pricing of them isn't all that high either. It's possible to pump skillchecks quickly and easily precisely because they're not priced high. And, in almost every campaign I've been in, skill checks have been useful. Spot/Listen get called a lot and can help get the drop on enemies, stealth checks come up fairly frequently, acrobatic checks are nice in lower level games and even in some mid-level games, and there's a nice feeling of smooth competence when you've got that +20 skill mod and can consistently roll superhuman scores. My lvl 10 character who can throw out 30-ish skill checks reliably isn't all that much more powerful for it, but it still feels nice. It's like in a movie when you see someone assemble a rifle quickly and smoothly, and you know they know what they're doing. High skill checks add that slickness factor, and really don't cost that much to get.

[Seerow]

Do you realize you've described 5e's "skill system" 100%?

I'll refer you to the threads discussing 5e for the arguments against it. Because I don't feel like making the arguments all over again.

[Ziegander]

Under your system, unless someone has a specific background like "Miner", a Str 20 hero is only marginally more competent as strength-based activities than a Str 10 commoner. I want my heroes to be skillful and exceptional and be able to do things well within their specialties. If I have a strong character, I want him to be mighty. If I have a dextrous character, I want them to be lithe. Flubbing skillchecks like a total newbie takes away from that considerably, but that's what they're left with under your system.

This is a complaint against how d20 games resolve ability checks in general, not exactly a complaint against my system specifically.

In D&D, yes, a character with Str 20 is only 25% more "mighty" than a character with Str 10, except in terms of carrying capacity and ability to kill stuff with a two-handed weapon. My system doesn't change that.

In my system, yes, if you compare two characters, both with no background (aka no training) in Tumbling, and one has Dexterity 20 while the other has Dexterity 10, the former character will only succeed +25% more often than the latter. What's so wrong with that? Neither of them have any experience in the activity.

Putting it another way, how does 3.5's standard skill points system help your Strength 20 character to feel "more mighty" than my proposed removal of said skill system?


But now magic becomes even more essential because the non-magic solutions are now worthless.

The non-magic solutions are worthless using the current D&D 3.5 skill system. A Rogue doesn't gain any appreciable advantage over another character because of his 8 skill points per level, especially when he puts those skill points in "typical Rogue stuff" like Climb, Disable Device, Hide, Listen, Move Silently, Open Locks, Search, and Spot.


My lvl 10 character who can throw out 30-ish skill checks reliably isn't all that much more powerful for it, but it still feels nice. It's like in a movie when you see someone assemble a rifle quickly and smoothly, and you know they know what they're doing. High skill checks add that slickness factor, and really don't cost that much to get.

But nothing about the standard D&D 3.5 skills system is tied inherently to this. Stuff like class features, feats, synergy bonuses, masterwork tools, magic item bonuses, and spells all can add up to provide large bonuses to skill checks, without the baggage of class skills, skill points per level, and skill ranks.

Obviously I haven't fiddled with check DCs, but given my proposal to ditch the skill system entirely, and replace it with ability checks, I would need to change the DCs on lots of the standard skill uses. So, using my system, someone could still quickly and smoothly assemble a rifle, and you can still know that that person knows what they are doing because either they have the background for it, or the skill feat for it, and they can reliably hit that higher DC for faster assembly.

As far as Backgrounds are concerned, I would probably stick to very general rules that allow players to make them up as part of character creation. Choose any three "skills," get a +3 (scaling to +10) bonus on them. Choose any five, get a +2 bonus (scaling to +7). Choose any seven, get a +1 bonus (scaling to +5). That sort of thing.

@Seerow: Yeah, you're pretty much right, though, in adapting it to 3.5, scaling with level is thrown in for good measure. So not 100%, but 95-99%, sure. I'm also not sure what the heck the rules are in 5th Edition for specific "skills" like Tend Wounds or Command/Deceive Item, but I would make sure rules for such ability checks are spelled out (or rather, they already are I just have to fiddle with the DCs).

[jackattack]

Under the current system, I can buy any skill I want to fit my character background. And I decide how good I am at it by how many points I choose to spend, so I am better at some things than I am at others.

Under the proposed system, I have to take bundles of skills, and my skill choices determine my character background (instead of the other way around). And my skill level is determined solely by my rolled stats, so I am as good at every STR-based skill I have as I am at every other STR-based skill I have.

Sorry, I prefer the current system over the proposed system.

[Deox]

If you have a chance, take a look at Star Wars Saga Edition's skills and rules. Simplified list, you're either trained or not, and you can focus.

For an example, let us take an acrobatics check (tumble would fall under here). You get 1/2 your character level as a bonus to this roll, if you're trained in acrobatics, you get another +5 and if you're focused, an additional +5. So, a 10th level character has a +15 to acrobatics (+5 from 1/2 CL, +5 Trained, +5 Focused).

This method really streamlines skill checks, makes the skillful characters still skillful, and makes you still want a decent intelligence so you can be trained in more off the bat.

Won't get rid of the skills entirely, but at least this is a step in the direction I think you're wanting to go.

[sonofzeal]

This is a complaint against how d20 games resolve ability checks in general, not exactly a complaint against my system specifically.
Ability checks suck. Skill checks suck considerably less... but still suck a bit, so your solution is to erase them and make everything into the more sucky version? Seriously? How is that not a complaint against your system?



The non-magic solutions are worthless using the current D&D 3.5 skill system. A Rogue doesn't gain any appreciable advantage over another character because of his 8 skill points per level, especially when he puts those skill points in "typical Rogue stuff" like Climb, Disable Device, Hide, Listen, Move Silently, Open Locks, Search, and Spot.
I really don't think we've been playing the same game. Skill checks are rarely essential, but they've been useful in just about every game I've been in except pure dungeoncrawls, and even there they have moments. I've give some examples:

- I once used a Jump and a Tumble to get my Barbarian past the bodyguards and right up in the face of the enemy spellcaster.

- While DMing, I once had a player use Balance checks to run up a rope and get to a high guarded ledge.

- Spot and Listen checks routinely allow us to detect enemies earlier, and potentially get a Surprise Round against them, or at least prevent them getting a Surprise Round against us.

[ericgrau]

This system should work fine since skills are mostly separate from the rest of the system. I only see one issue which is often a pitfall when DMs set the DC: chance of failure. Skills aren't that great,but DMs feel compelled to include a chance of failure even for the best PC. This is a problem when the skill has a minor benefit upon success and a major negative drawback upon failure. High DCs make smart PCs ditch skills entierely.

Two ways to handle this:
1. Forget the skill. Skills aren't that great so PCs should use spells and magic items 95% of the time. Spider climb, etc. Or remove it and roleplay it out instead. You might not do this for every skill, but there are articles on the benefits of completely removing some overused skills like spot and social skills and relying on narrative and roleplay instead.

2. Set DCs according to risk and ability.
Ability:
Level 1 high ability modifier: +4
Level 1 moderate ability modifier: +2
Level 1 low ability modifier: -1
Level 20 high ability modifier: +12
Level 20 moderate ability modifier: +5
Level 20 low ability modifier: +1

Risk:
If there is high risk, those with a high ability modifier should auto-pass the DCs at a reasonable level (DC 5 or 10). If there is no risk, they should need to roll a 5 while those with moderate ability might need a 10 (DC 10 or 15). This does not count background bonuses and feats, which may make someone excel much earlier even at the special uses of a skill (DC 15). i.e., what used to be DC 30 checks like tumbling through multiple foes.

Opposed checks wouldn't change much.

[ThiagoMartell]

That's basically what D&D Next does.

[KillianHawkeye]

I disagree that the skill system is useless. At least, certain skill-based classes such as the Ranger and Rogue can make good use of it if they don't spread themselves too thin.

Example: I have an 8th level Ranger/Scout. He can make almost any Spot, Listen, Hide, or Move Silently check even if he rolls less than 5.

[Libertad]

I've heard that D&D variant Castles and Crusades done away with skills and feats in favor of a stronger emphasis on race and class. Your race and class determine which abilities are "Prime," and it's easier for characters to succeed on actions which fall under their Prime stat.

[NichG]

I've been thinking about the issues with the 3.5 skill system. To me, the bigger problem is that the net bonus from magical items, spells, etc can be four or more times higher than the actual investment in the skill in a high-OP game, but won't be in a lower OP game. That means that there's a huge variance you have to plan for in designing the system, and in the high-OP case the investment of ranks is basically irrelevant.

So instead, my thought is 'keep skills, but divide the effects for each skill up into a set of minimum ranks'. For example, Hide:

- 0-4 ranks: You can hide in situations with total cover/concealment, or at large distances (100ft+).
- 5-9 ranks: You can now hide when you have partial cover/concealment, but you cannot move while hidden.
- 10-14 ranks: You can now hide with partial cover/concealment and can hide when moving from one source of cover to another, so long as you end your move with cover.
- 15+ ranks: Now you can take re-hide when taking a ranged attack (standard Sniping rules), and can attempt to hide at full speed.

This helps it not be a pointless choice. The details still matter, since any opposed check will use the modifiers not the ranks.

[KillianHawkeye]

So instead, my thought is 'keep skills, but divide the effects for each skill up into a set of minimum ranks'. For example, Hide:

- 0-4 ranks: You can hide in situations with total cover/concealment, or at large distances (100ft+).
- 5-9 ranks: You can now hide when you have partial cover/concealment, but you cannot move while hidden.
- 10-14 ranks: You can now hide with partial cover/concealment and can hide when moving from one source of cover to another, so long as you end your move with cover.
- 15+ ranks: Now you can take re-hide when taking a ranged attack (standard Sniping rules), and can attempt to hide at full speed.

This helps it not be a pointless choice. The details still matter, since any opposed check will use the modifiers not the ranks.

I kind of like this idea, but it'll take some work to convert all the skills. You may have to incorporate all the expanded skill uses from splatbooks to get enough interesting options.

[Tenno Seremel]

With a d20 roll you might have better luck with attribute scores instead.

[NichG]

I kind of like this idea, but it'll take some work to convert all the skills. You may have to incorporate all the expanded skill uses from splatbooks to get enough interesting options.

I don't think the work would be more than a few hours from the designer point of view. But it'd be several sessions for the players to get used to the new system, which could kill the idea.

I'm phasing this in in my own games, introducing new uses of existing skills that require certain minimal ranks. If my group gets used to it, I'll start converting over the base stuff.

[HouseRules]

Use AD&D 2E Thief's Skill limitations.
In 2E, the maximum number of skill points per level to any one skill is 15; that's 15% change.
The equivalent in 3E would be to limit a maximum of 3 skill points per skill per level; 3 ranks for class skills, 1.5 ranks for cross-class skills.

[Vrock Bait]

Honestly, I don’t think this is a good idea at all. I think the main problems you’re citing, like having to keep track of things, are completely fixed in 5th Edition, where you barely have to deal with skills.

If you actually like the system, but think it’s imperfect, then the better solution is to add more skill synergies and combine different useless skills. For example, Spot + Listen, Disable Device + Open Lock, etc.

Because while the casual player doesn’t like keeping track of too many things, that’s why most of them play 5e. I enjoy, as a munchkin, being able to boost my Jumps astronomically. I enjoy, as a roleplayer, to assign Profession(fisherman) to my fighter to mechanically demonstrate his background, low skill points or not.

And really, what you’re doing is just, firstly, completely disregarding the effects of training for the natural ability of a character, and secondly, getting rid of awesome features like skill tricks. I understand, you might not like keeping track of things, but that’s what 5e is for. 3.5e is complicated, and that’s on purpose.

[Evoker]

Am I missing something, or is this an 8-year thread necro? That's probably the biggest number of years between posts I've ever seen on any forum.

[Saintheart]

You're not missing anything.

Arise thread, long since dead
Haunt the forums from which thou first were spawned...

[jdizzlean]

The Mod Life Crisis:Sadly, there isn't a skill in existence that can defeat thread necromancy, not even with a natural 20.