Making the Skill System Work

[3catcircus]

Ok, I've been reading about the skill system stuff - complex skills, etc., and I thought back to the original D20 system used for TW:2K v2.2

Right now, with 3.x, the skill system assumes a DC of 5, 10, 15, 20, 25, etc. Nice 25% increments (25% assuming each digit on a d20 is 5%).

This makes it difficult for a DM to determine just what is a difficult or easy DC for a given skill check.

Why not take a hint from TW:2K and base the DC on the (N)PCs' own skill levels?

Easy = Skill / 4
Average = Skill / 2
Difficult = Skill
Formidable = Skill * 2
Impossible = Skill * 4

So - a PC with 5 ranks in the Hide skill and a ability mod of +3 give a total skill of 8.

An easy task (hiding in pitch dark while wearing a blacksuit and not moving) requires you to beat a DC of 2. An average task (same situation but no blacksuit) requires you to beat DC 4. A difficult task, hiding in the shadows of a dungeon while not moving, is a DC 8. A formidable task (hiding in moonlit shadows while moving) is a DC 16. An impossible task (hiding in broad daylight while jumping up and down) is a DC 32.

This seems to make a lot more sense. It equates the chance of success directly to a given (N)PCs skill level, rather than an arbitrary DC. A person with lots of skill will find an "easy" task to be easy while another person with very low skill levels will find an "easy" task to be more difficult. This seems to mimic real-life a bit more. For example, taking a corner at high speed will be much easier for a police officer or race car driver than for a teenager with a learner's permit because of the relative skill levels.

Thoughts?

 

[gizmo33]

This is backwards from what I thought. I thought that difficulty should be assessed independent of the skill of the PC. When I'm trying to decide what the DC for something should be (barring other guidelines), if I decide that it would be difficult for a normal person, that would fix the DC (say at 18), that way I'm not penalizing someone with lots of skill points. Sure, for someone with a +18 modifier to the skill, the task becomes pretty easy for that individual, but I think in that case the modifier speaks for itself and a classification system isn't a very useful tool. What I wouldn't like (and what you might not be saying) is to decide on the DC of the wall after I know the skill level of the person making the attempt. The wall doesn't suddenly get more difficult to climb when experienced climbers approach it (hmmm, though this is a fantasy world...)

 

[Aust Diamondew]

Bad idea. For the reason's listed by Gizmo.

And thats where the wonkiness just begins.
A character with a 0 modifier in a skill will be up against a DC 0 skill check for anything.
A character with +20 in a skill would have a harder time than a character with +10 in a skill at an Formidable task. The +20 character would need a natural 20 to hit his DC of 40. The charater with plus 10 would need to roll a 10 or better to hit his DC of 20.

 

[Twowolves]

In the system the OP is talking about, it's a flat d20 "roll under" style. Someone with a "+20" modifier would need to roll:
>5 (Impossible)
>10 (Formidible)
>20 (Difficult)
>40 (Average)
>80 (Easy)
on a 1d20.

Not the same as "roll 1d20+ mods >= DC" currently in use in 3.X ed.

Still wouldn't work for D&D as it is now (stat bumps become way more important than skill ranks, and any system that multiplies base numbers while keeping a limited die roll range gets wonky fast).

 

[Zurai]

Uh, then what's the point of even having skill ranks? Under your system, more ranks = harder to perform hard tasks. Also, it scales way, way, WAY too quickly.

Example:

Person A has a total skill modifier of +1. His "Formidable" task will be a DC 2 and an automatic success (even a roll of 1 will get him 1+1=2).

Person B has a total skill modifier of +21. His "Formidable" task will be a DC 42 and will be quite literally impossible for him to succeed at (even a roll of 20 will get him 21+20=41).

Person B was a complete moron for putting ranks in the skill, which completely negates the entire point of a skill system.

 

[Zurai]

In the system the OP is talking about, it's a flat d20 "roll under" style. Someone with a "+20" modifier would need to roll:
>5 (Impossible)
>10 (Formidible)
>20 (Difficult)
>40 (Average)
>80 (Easy)
on a 1d20.

Not the same as "roll 1d20+ mods >= DC" currently in use in 3.X ed.

No... no it isn't. I quote from the OP:

An easy task (hiding in pitch dark while wearing a blacksuit and not moving) requires you to beat a DC of 2. An average task (same situation but no blacksuit) requires you to beat DC 4. A difficult task, hiding in the shadows of a dungeon while not moving, is a DC 8. A formidable task (hiding in moonlit shadows while moving) is a DC 16. An impossible task (hiding in broad daylight while jumping up and down) is a DC 32.

The DC increases as the "difficulty" of the task increases. Your method is the inverse (and still suffers identical problems, by the way).

 

[fuindordm]

It's a nice idea, but it doesn't work for d20+skill > DC.

I've seen it work well in games with percentage-based skills, something like

For a skill rated at S%, your %chance of success is

Routine task : Sx2 (writing an essay, climbing a tree)
Challenging task: S (writing an original story, climbing a steeply sloped rock face)
Difficult task: S/2 (writing a good novel, climbing a vertical rock face without tools)
Extraordinary: S/4 (writing a masterpiece, climbing Matterhorn without tools)

 

[Twowolves]

No... no it isn't. I quote from the OP:



The DC increases as the "difficulty" of the task increases. Your method is the inverse (and still suffers identical problems, by the way).

I must have skimmed past his examples when I read that he was trying to translate the old GDW House system to d20. What I described is that old system. I have no idea what the OP was trying to come up with, but it's not the same.

In that old system, you added your Stat (ranging from 1 to 16, but averaging around 6 or 7) and added your Skill level (1-5 I think). That was your target to roll under on a Difficult task with a flat 1d20 roll. Easier tasks doubled or quadrupled your base number (still rolling under on 1d20), harder tasks halved or quartered it.

It's not "MY" method at all, I just thought I was helping clarify what the OP meant. But I guess I didn't. Furthermore, I in no way endorse either version of this system for D&D.

 

[3catcircus]

I guess I didn't articulate it as well as I thought...

As was mentioned by Twowolves, my inspiration was the TW:2K's flat d20 "roll under," where the standard task was "Difficult" and the DC was equal to your skill level.

So - in that system, with abilities ranging from (IIRC) 1-10, a person might have a Jump skill of 8 (Str 5 + 3 ranks in the skill) and would have to roll under an 8 on d20 to succeed at a Difficult task. To succeed at an "Average" task, he'd have to roll under 16.

What I am advocating is an inverse type of d20 + skill "roll over." Str 18 (+4 modifier) and 6 ranks in Jump would still net you 10 total ranks in Jump, for example.

What I didn't articulate was that a task that someone with a skill of 5 would consider a "Difficult" task would be an "Easy" task for someone with a skill of 10.

So: for someone with 6 ranks in Jump, jumping 5 feet straight up might be a "Difficult" task requiring you to beat a 10 by rolling at least a 2 (DC = Skill level of 10, and unless you roll a 1, you should succeed). For someone with 10 ranks in Jump, jumping 5 feet straight up might be an "average" task requiring you to beat a 7 (DC = Skill level of 14/2), while jumping 10 feet straight up might be considered "Difficult," requiring you to beat a 14.

I think this more closely mimics real-life - e.g. running a 4 minute mile is gonna be "impossible" for me, might be "difficult" for a college baseball player, and might be "easy" for a world-record holder of the fastest mile (which is, btw, 3:43.13, set by Morocco's Hicham El Guerrouj on 7 July 1999 in Rome...)

Don't know if it is worth the trouble, but *anything* has to be better than the arbitrary static DCs set for skills, especially with the woefully inadequate number of examples that currently exist in the 3.5 PHB.

 

[Zurai]

Don't know if it is worth the trouble, but *anything* has to be better than the arbitrary static DCs set for skills, especially with the woefully inadequate number of examples that currently exist in the 3.5 PHB.

Actually, scaling DCs either require hideously complicated rules or just plain break. Static DCs are the simplest, easiest, and most logical way to handle skills.

Your example can very, very easily be simulated by a series of DC 15 checks.