A Skill System Idea

[GSHamster]

Thanks to all the discussion lately about skill systems lately, I've had an idea for a skill system, and I'm presenting it to you guys for feedback. It's slightly influenced by the system in Mass Effect 2.

Background

I usually DM, and the problem I've always had is setting appropriate DCs. Starting from a static base and adding modifiers often ends up in an unexpected place, and my skill checks never seemed to hit the sweet spot. They always seemed to end up too high or too low.

So I'm trying to come at it from a different angle. Instead of asking "How difficult is this task?" and coming up with a numeric answer, let's ask, "What type of person could complete this task?" It seems easier to me to say that "anyone can do it" or "only a highly-skilled person could do it".

System

Each skill has four ranks:

Rank - Cost - DC
Untrained - default - DC 10
Trained - 1 point - DC 12
Excellent - 3 points - DC 14
Master - 5 points - DC 16

All characters start Untrained by default. Purchasing higher ranks costs an increasing number of skill points. Spending 1 point makes the character Trained. Then spending 3 more points moves her up to Excellent.

Untrained people have no formal training in the skill. They rely on natural ability to complete tasks.

Trained people have some training in the skill. The majority of people with this skill are at this level.

Maybe 1 or 2 people in a large town or city have Excellent skill level.

Masters are the best of the best. Maybe a handful of people in the entire country will be at this level.

Task Resolution

Tasks or challenges are then linked to the skill rank. The DM determines what level of skill this task requires. Perhaps only an Excellent climber could climb this particular cliff face. If the character has that skill rank, she makes an Ability check against the DC listed in the table above. The DM chooses the relevant ability based on the task.

A character with a higher rank automatically succeeds at the task without fear of failure.

A character with a rank one level lower can attempt the task with a -10 penalty. Characters with ranks below that cannot succeed at the task.

Example:

The PCs need to scale a sheer cliff face at night. They decide to send Anne up and have her drop a rope back down to the rest. The DM declares that this particular cliff at night would be challenge for an Excellent climber. If Anne is an Excellent climber, she makes a Dex check against DC 14 to see if she is successful. If Anne is a Master climber, she can easily scale the cliff. If Anne is merely a Trained climber, she can attempt the cliff, but suffers a -10 penalty to her roll (effectively DC 24). If Anne is Untrained, she cannot scale the cliff successfully.

Example Skill:

Medicine (mostly Wis or Int based)

Untrained challenges
- cleaning and bandaging a wound preventing infection
- popping a joint back into place

Trained challenges
- setting and splinting a broken bone properly
- diagnosing common illnesses
- making cures for common and uncommon illnesses
- amputation of limbs

Excellent challenges
- diagnosing rare diseases
- making existing cures for rare diseases
- internal surgery
- saving limbs that most would amputate

Master challenges
- creating a cure for a new disease
- brain or spinal cord surgery

Comments

I think this system is fairly simple. It's non-linear in terms of skill points, making easier to learn a lot of little things, but harder to master one skill. For the DM, there's really only 4 categories for challenges, making it a lot easier to assess correct difficulties. Automatic success at lower challenges makes a player who invests in a higher rank feel that decision paid off, which can be hard to see when you invest 1 point at a time.

I also think it maps better to how we think of skilled people. I think it's easier to think of people as a series of tiers (professional atheletes vs amateur atheletes for example) than it is to see a strictly linear gradient of skill.

 

[AeroDm]

That's pretty darn interesting. A few comments:

I'm not sure the DC has to increase for each category. If you think about a system like 4e, there are easy, medium, and difficult settings to a skill. The DCs are then set so that a character has a 65% chance to succeed at a skill of their appropriate level. That same format is what you are aiming at except you've said that easy has 65%, medium has 55%, and difficult has 45%. It is sort of a double tax (pay once for the skill proficiency and again for reduced chance to succeed).

I would change the penalty to -5 if one category below and -10 if two categories below. The reason is that skills are already very not-inclusive and the game is generally more fun when people can try things. Also, if Anne is an Expert climbing an Expert cliff, we want her to have *some* chance to fail. Let's say it is as little as 20% requiring her to roll a 5 or higher. The -10 penalty then has to roll a 15. At that steep a penalty, few people will bother trying if they aren't as proficient as the cliff. Maybe that's the goal, but I'd prefer the more inclusive approach.

Finally, I'm presuming a 4e like system where characters improve over time. If a 10th level character adds +5 for level, you might need more ways for the DC to scale. Otherwise, on a 30th level game like 4e, the +15 from level alone means a character auto succeeds on anything of their level or lower and really struggles for anything above their level. That sort of binary result won't be very exciting.

 

[GSHamster]

Thanks for the feedback!

I'm not sure the DC has to increase for each category. If you think about a system like 4e, there are easy, medium, and difficult settings to a skill. The DCs are then set so that a character has a 65% chance to succeed at a skill of their appropriate level. That same format is what you are aiming at except you've said that easy has 65%, medium has 55%, and difficult has 45%. It is sort of a double tax (pay once for the skill proficiency and again for reduced chance to succeed).

The increasing DCs is essentially to keep up with increasing ability scores. A Master person probably also has a high natural ability. If they have a +4 or +5 and the DC stayed the same at 10 or so, the check would become roll 5 or higher. Master challenges are tasks which should still challenge a player at that level. After ability scores, the roll should go back to 10-12 and up is a success.

I would change the penalty to -5 if one category below and -10 if two categories below. The reason is that skills are already very not-inclusive and the game is generally more fun when people can try things. Also, if Anne is an Expert climbing an Expert cliff, we want her to have *some* chance to fail. Let's say it is as little as 20% requiring her to roll a 5 or higher. The -10 penalty then has to roll a 15. At that steep a penalty, few people will bother trying if they aren't as proficient as the cliff. Maybe that's the goal, but I'd prefer the more inclusive approach.

This is a good point, but the issue I had is that you get Untrained people having a small chance of doing things they should not be able to do.

One other thing I considered is to add an extra category between Trained and Excellent (Good). If you did that, then I think the two steps away would be more reasonable. The table would look like:

Untrained - default - DC 11
Trained - costs 1 point - DC 12
Good - costs 2 more points - DC 13
Excellent - costs 3 more points - DC 14
Master - costs 4 more points - DC 15

It's similar to the previous table (10 points total to get Master) and a being able to attempt tasks 2 steps away seems more reasonable. Untrained could attempt up to Good, Good could attempt Master challenges.

The issue with this is that it might become harder for the DM to assign DCs. What's the difference between Good and Trained? or Good and Excellent? But maybe that's not an issue, and the increased smoothness between ranks outweighs a little more difficulty in assigning DCs.

Finally, I'm presuming a 4e like system where characters improve over time. If a 10th level character adds +5 for level, you might need more ways for the DC to scale. Otherwise, on a 30th level game like 4e, the +15 from level alone means a character auto succeeds on anything of their level or lower and really struggles for anything above their level. That sort of binary result won't be very exciting.

This wouldn't be a good system to add bonuses from levels. The idea that as character levels they get better at a lot of things should be handled by the costs of each skill rank. It costs a very small amount of points to get to Trained (or Good) so a character can be decent at a lot of things very easily. Put another way, sacrificing one Master skill means that you can get 4 or 5 other skills up to Trained.

 

[AeroDm]

I think your instincts are right; fewer categories makes it easier to find something a home. I also better understand your scaling DCs now that you point out this is not a 4e-esque system with built in raises.

The last area that I see catching you up is the relationship between # skills and # skill points. Some skill systems create a direct linkage between aptitude and actions. This system does that to the extreme because if your aptitude is not within one step of the difficulty of the action, you auto fail. This means that all actions MUST fall under a skill, otherwise if the action is too difficult and no one can be trained, no one can do it.

So you'll need a really robust skill list. Hard, but certainly doable.

But once you've settled on the skill list you need to determine how many skill points to give out. This decision is really important because you need to balance the number of skills with the number of skill points. This is an area where you might want to refine your purchase schedule. For example, if the skill list is the size of 4e, the 1/3/5 might be a bad choice because achieving "master" (9 points) is a tradeoff from gaining "trained" in half the list. A purchase schedule of 1/2/3 would put "master" at 6 points and a third of the list. The converse is also true (large list would benefit from more points so the cost of entry to any one skill is reduced).

Fortunately, these changes are easy to just adjust the points awarded later in the design process. It really isn't a fundamental design issue--and your fundamentals look great.