[My As Yet Unnamed RPG] Skill system

[John Griffith]

At the heart of my system are the characters that drive the story; and the key players are intended to be the heros, PCs. As the focus of my game is intended to be the characters I wanted players to have as much control during the character building and improvement phases of the game as possible. If someone asked me the question: "So if I use your system, what kind of character can I build?" my answer to them would be: "Anything you can imagine, in time." The character building system and the character improvement system uses the same principle: a skill set.

Skill set
A skill set is simply a term describing the entire breadth and width of a character's abilities. Not everything a character can do is represented by a skill, for instance climbing a ladder, or tying a shoe, etc. Skills such as these are considered to have already been mastered by a character and they are not replicated by the skill system. Equally, I wanted the skill system, the whole of my game system, to encourage roleplaying--that is to encourage creativity, critical thinking skills, problem solving, and such like. Therefore my system intends to replicate only those skills that the player himself cannot mimic. In application what this means is my system does allow characters to take skills that mimic: knowledge, wisdom, charisma, diplomacy, bartering, intelligence, wit, charm, etc. This means that skills a character takes are always either skills describing complex physical tasks or supernatural abilities.

Miscellaneous skills
This is one of three categories that describe all skills. The Miscellaneous skill category includes any and all skills that mimic physical activities that are not considered combat-related or magical in nature. I have skill lists in my core rules, however my system is not about lists of skills. The skill lists I provide serve merely as examples. Players can select from the provided skill lists, but they are encouraged to build their own skills. Building a skill is quite simple, there is one rule that guides the player/GM:

1. The parameters of the skill must be defined in concrete terms with a clearly defined cause and effect.

The skill system is designed to break down professions (e.g. a lock smith) into component parts, each part being its own skill. The intended effect of this is to:

a) Create characters that can be on a sliding scale from highly focused and specialized in one or a few areas of expertise to characters who are jack-of-all-trades being able to do a little bit of everything.
b) Simulate the dedication it takes for experts to become singularly skilled at their artform.
c) Offer players, freedom to pick only those skills that they want their character to be able to perform.
d) Create non-combat characters who are as talented and valued as combat characters.
e) Encourage characters whose ability to think, learn, create, and problem solve are not predicated on a finite number.

Task resolution
A character's ability in each Miscellaneous skill is measured by a descriptor ranging from clueless (untrained) to Neophyte (beginner) all the way up to Grandmaster (highest level of skill). There are twenty-six descriptors in all counting clueless which in effect is descriptor zero. The difficulty of a task is measured by thirty-one descriptors ranging from simple (easiest) all the way up to Impossible (hardest). Impossible tasks are impossible to complete unless the character has some form of additional assistance other than their skill--this assistance may come in the form of a special tool, or innate talent, or magical assistance. The success or failure of a task is decided by a percentile roll. The character's level of skill is matched on a chart with the difficulty of the task which determines a target number. The percentile roll must be less than or equal to this target number for success.

What decides task difficulty?
Some tasks may have a base difficulty attached to them, for instance performing a handstand, while other tasks may have a variable difficulty attached to them, for instance picking a lock. In this case the difficulty will be decided by the complexity of the lock which will differ based on the value of what it is protecting. For many tasks the difficulty will be based in part or in whole on the environment. Using the handstand example. If conditions are perfect the difficulty to perform a perfect handstand might be one level. However, what if a strong wind is blowing and the ground is not level? In this case the strength of the wind might add three levels of difficulty and the sloping ground might add two levels of difficulty which would raise the overall difficulty to six. All of the skills are affected by relevant environmental conditions. For some skills the difficulty of a given task is predicated solely or almost solely on environmental conditions. Take the the ability to sneak about without making noise as an example. Certainly the skill level of the sneak comes into play, but what affects the difficulty of the task--sneaking past that guard? What kind of equipment is the sneak wearing? Does any of it clank or jingle as they move? What kind of clothing is the sneak wearing? What kind of boots? Are their boots hard soled or stiff? Do they squeak? What sort of surface is the sneak moving over? Pebbles? Sand? Snow? Carpet? Grass? Cobblestones? Will their footsteps echo? Is there is any background noise? Perhaps running water or a crackling fire? Is the guard distracted? In conversation with another guard maybe? Or half asleep? Or drunk? Maybe he's a little hard of hearing? All of these factors are taken into account in determining the difficulty of the task. Now it sounds like a lot of information to consider, and it is. I use a base modifier for the task difficulty for key elements. For instance moving over carpet may increase the difficulty of the task by only a little whereas moving over snow may increase it by a lot. Generally there will only be between three and five key conditions that will modify the difficulty.

Success/Failure
Depending on the nature of the task, failure may or may not penalize the character. Success in a task always nets a character an amount of experience based on the difficulty of the task in proportion to their skill level. If a skill is too easy for a character they may not get any experience at all. If a character fails a task but does not fail by a wide margin they still earn experience for the attempt, but a smaller amount than they would have earned for success. This mechanic is designed to model the fact that we can learn from our mistakes, however I did not want to reward failure equally well or better than success and encourage failure.

Questions? Comments? Like it? Hate it?

[TonyLB]

All of these factors are taken into account in determining the difficulty of the task.

By whom, and for what purpose?

By the GM, whose goal is to oppose the players in any way he can get away with?  By the players, whose goal is to provide their characters with a level of adversity that will earn them the praise of their peers without getting the character killed?  By other players, whose aim is to apply social pressure against actions they don't like and social encouragement for actions they do like?

[Eero Tuovinen]

You should be asking questions of us, not the other way around. After all, surely you had some reason to post these ideas here?

That said, I find the basic philosophy screwy: if you only model skills the players can't mimic, then you should only have supernatural skills, surely? Unless you're playing with somehow limited people most other things they can mimic quite fine, so fighting, thief skills, craft skills, languages and such are right out...

But that's not really relevant; if magic/fighting/craft is the split you're happy with, then you should go with it. My personal experience is that this kind of fiddling with the ultimate skill system is a pretty common rpg designer hobby. Only rarely do I see fruitful games growing out of it, though, because the designer fails to make the skill system become an engine in a fruitful gaming experience. In practice all these elaborate skill systems seem to play just like GURPS, my own early tinkerings included.

If you're interested in seeing fruitful skill systems in action, I recommend The Shadow of Yesterday and Runequest. You'll notice that these don't work because of their detailed (or not-so-detailed) realism, but because they motivate interesting play interactions by tying the skill system into something meaningful (massive game advantage for relatively small investment in the case of TSOY, religious cult progression in Runequest).

[John Griffith]

All of these factors are taken into account in determining the difficulty of the task.

By whom, and for what purpose?

By the GM, whose goal is to oppose the players in any way he can get away with?  By the players, whose goal is to provide their characters with a level of adversity that will earn them the praise of their peers without getting the character killed?  By other players, whose aim is to apply social pressure against actions they don't like and social encouragement for actions they do like?

Thanks for your comments.

Yes, the GM determines the difficulty of the task. As I said it is often influenced by environmental conditions. Is it the GM's goal to oppose the players in any way he can get away with? I would hope not. Are you suggesting that the mechanics open up the system to abuse by GMs or even encourages abuse by GMs? I'd like to understand the point you are making with that statement.

Players can influence the task difficulty to some degree through roleplaying, but they do not set it:
a) wearing appropriate clothing and using appropriate equipment for the specific task,
b) describing the way in which they perform the task
c) using the environment, time of day, natural terrain, etc., to their advantage

[Adam Dray]

John, I think you'll find with more reading on the Forge that Tony's pointed questions were not accusations, but really well-formed questions designed to understand your design priorities.

That said, there is nothing wrong with designing a game where the GM does everything he can within the rules to oppose the players. In fact, games like Dogs in the Vineyard and My Life with Master work best when the GM is not pulling punches and the game designs allow the GM to play balls-out "against" the players without ruining things. Games poorer for design put the GM in a position where they have to carefully balance adequate challenge with keeping the players alive/happy/etc.

Furthermore, players crave that all-out challenge in certain kinds of games. When I play Dogs or MLwM, I want the GM to assail me ruthlessly because the game puts the real choices in my hands, regardless of what the GM does to me.

It seems to me that you're trying to write a skill system that captures reality to the Nth degree. Do you think game play will be better for it?

The skill system is designed to break down professions (e.g. a lock smith) into component parts, each part being its own skill. The intended effect of this is to:

a) Create characters that can be on a sliding scale from highly focused and specialized in one or a few areas of expertise to characters who are jack-of-all-trades being able to do a little bit of everything.
b) Simulate the dedication it takes for experts to become singularly skilled at their artform.
c) Offer players, freedom to pick only those skills that they want their character to be able to perform.
d) Create non-combat characters who are as talented and valued as combat characters.
e) Encourage characters whose ability to think, learn, create, and problem solve are not predicated on a finite number.

Let's break this down a bit. Goal (b) seems to relate to desire for "realism." Goal (c) seems to relate to player wish-fulfillment. That is, give the player the tools to create the character they really want to play. The "only" in that sentence confuses me, however. Goal (d) seems a design priority to de-emphasize combat play (probably relative to your prior experience with D&D or some other game).

Goal (e) confounds me a bit. Characters don't do any of those things. Players do them for their characters. We can pretend that characters do those things. Even so, you have removed any modeling of those things from your design -- your system "intends to replicate only those skills that the player himself cannot mimic." Do you mean to say that because you don't have skills for things like Diplomacy and thus leave all diplomacy up to the player convincing a GM that he has been sufficiently diplomatic, that player's character's abilities are "not predicated on a finite number" (ability score or skill number)?

Why do this? Does it stem from a belief that if you codify social things, that players won't role-play them? The ol' "role-play, don't roll-play" adage?

Do you realize that it means that your game does not allow a player of average intelligence or average charisma to play a character of high intelligence or high charisma? How does that meet your goal to let a player play any character he wants, in time?

[TonyLB]

Yes, the GM determines the difficulty of the task. As I said it is often influenced by environmental conditions. Is it the GM's goal to oppose the players in any way he can get away with? I would hope not. Are you suggesting that the mechanics open up the system to abuse by GMs or even encourages abuse by GMs? I'd like to understand the point you are making with that statement.

My point in asking the question is to see whether you have an answer to it, and if so what that answer is.

So, you've told me one thing that the GM's agenda isn't (outright unbridled competition).  Now, can you tell me what the GM's agenda is?  What's he supposed to be doing, and how does doing it reliably contribute to the game?

[John Griffith]

I wrote a bunch of comments in response to your first few paragraphs but when I got to your last paragraph I had to delete them all. Read on...

I accept your premise that removing social skills from the game could prevent a player from playing a character with intelligence or charm beyond their capability to roleplay it. That does indeed defeat my goal of giving players freedom to play any character they can imagine. I'll be. So either I have to redesign my premise, or, social skills need to be in. Wow. Thank you. You have no idea how insightful that question was.

Let me ask you this:

Based on that decision--to include social skills--would it then follow in your mind that a character's attributes; i.e. strength, speed, etc., would also need to include things like intelligence, etc? My thought is yes. As I have removed that from my design I will have to put it back in, which really is not difficult to do because the mechanics of the system works identically for all associate skills. But I'm interested to hear your thoughts on this.

[Adam Dray]

I urge you not to ignore Tony's questions. He can provide a lot of insight into things, too.

Would I also include abilities like intelligence, etc.?  Well, that depends.

If you goal is to provide a system that can describe any conceivable character in game terms, you'll need game terms that span the vast space of human creativity. There are two ways to do that: 1) create an attribute or skill for everything conceivable, or 2) use freeform traits. The first is an impossible ideal. You always end up leaving something out and thus your player can't create his high-functioning autistic character or whatever. The second requires more of players but gives them ultimate freedom to create their own ability score and skill names, essentially. You write "High functioning autistic 7" on your sheet and that gives it teeth in the game. Nowhere do the rules explain how to interpret that, so you need rules that help players interpret things on their own, or at least to help them come to some resolution when there are disagreements. One potential downside of freeform traits is that not all traits are written equally. For example, "Good at throwing stars 10" probably isn't as useful as "Ninja 2."

I'm curious to see your responses to my other comments, and to Tony's.

[John Griffith]

My point in asking the question is to see whether you have an answer to it, and if so what that answer is.

So, you've told me one thing that the GM's agenda isn't (outright unbridled competition).  Now, can you tell me what the GM's agenda is?  What's he supposed to be doing, and how does doing it reliably contribute to the game?

The GM is supposed to extrapolate what key conditions present at the time the character attempts their task have a significant impact on the character's chance of success.

If the GM does this extrapolation reliably he presents a challenge for the player to overcome that allows him to contribute in a unique fashion to the achievement of a group goal.

The overall idea is to offer every player a chance to shine, with the net result being to encourage a team effort in accomplishing overall goals.

That second question was a really good question by the way.

[John Griffith]

I urge you not to ignore Tony's questions. He can provide a lot of insight into things, too.

Would I also include abilities like intelligence, etc.?  Well, that depends.

If you goal is to provide a system that can describe any conceivable character in game terms, you'll need game terms that span the vast space of human creativity. There are two ways to do that: 1) create an attribute or skill for everything conceivable, or 2) use freeform traits. The first is an impossible ideal. You always end up leaving something out and thus your player can't create his high-functioning autistic character or whatever. The second requires more of players but gives them ultimate freedom to create their own ability score and skill names, essentially. You write "High functioning autistic 7" on your sheet and that gives it teeth in the game. Nowhere do the rules explain how to interpret that, so you need rules that help players interpret things on their own, or at least to help them come to some resolution when there are disagreements. One potential downside of freeform traits is that not all traits are written equally. For example, "Good at throwing stars 10" probably isn't as useful as "Ninja 2."

I'm curious to see your responses to my other comments, and to Tony's.

Adam I wrote some responses to your other comments but threw them out after I accepted the premise that my skill system was not supporting my design goals.

The free form idea is interesting. Presently in my design a player say beginning from scratch and building a character would choose their own skills piece by piece, really very like your freeform concept I think. Every character begins as a clean slate. That is there are no character templates. If they are designing their own skill, rather than using one I have custom made they are required to define it in terms of cause and effect. i.e. "How do I perform this skill?" (cause) and "If I perform this skill what happens?" (effect). When they define the skill they are defining what category it belongs to. Their character's ability to use the skill effectively is a separate decision. There is a finite scale for measuring ability. This concept works for all skills irregardless of their cause and effect. Does that sound similar in design to the freeform idea?

[Adam Dray]

You probably ought to read the following design patterns in the free book Design Patterns of Successful Role-Playing Games, by John Kirk III: Attribute, Skill, and Trait.

Before we really go much further about redesigning your skill system, can you tell me a bit about the resolution system and how these skills will be used? Do you understand the difference between a task resolution system and a conflict resolution system? If so, does your system resemble one of those?

[John Griffith]

Here's my follow up to those other questions you raised Adam.

It seems to me that you're trying to write a skill system that captures reality to the Nth degree. Do you think game play will be better for it?

I guess that is not an unfair statement given the number of factors I suggest including in setting the conditions for pass/fail though that is not my intention. My purpose with this part of the design is to attempt to explain why such and such task is difficult or easy to succeed at. I like to know "why?" this works rather than just "how do I make it work?" Does that make a lick of sense to you? What are the any potential pitfalls--as you see them--to this mechanic, i.e. does it put too much emphasis on success/failure?

Let's break this down a bit. Goal (b) seems to relate to desire for "realism." Goal (c) seems to relate to player wish-fulfillment. That is, give the player the tools to create the character they really want to play. The "only" in that sentence confuses me, however. Goal (d) seems a design priority to de-emphasize combat play (probably relative to your prior experience with D&D or some other game).

Gaol (b). A desire for realism? In some ways yes. Most of all I am trying to present a sharply defined contrast between specialization and non-specialization.

Goal (c). You're pretty much on the money with this one. The "only" may have been a misplaced word, it was essentially a loose reference to the fact that I do not use character templates. Every character is built from the ground up.

Goal (d). If de-emphasizing combat play means bringing its importance in the overall scheme of the game more in line with other parts of roleplaying then yes, that is absolutely my intention. I don't want any particular part of my system to overshadow any other part of my system. I want players to decide for themselves what is important to their character and then choose those skills that reflect that decision.

Goal (e) confounds me a bit. Characters don't do any of those things. Players do them for their characters. We can pretend that characters do those things.

You're right, goal (e) was incorrectly worded. I wrote characters when I meant players.

Even so, you have removed any modeling of those things from your design -- your system "intends to replicate only those skills that the player himself cannot mimic." Do you mean to say that because you don't have skills for things like Diplomacy and thus leave all diplomacy up to the player convincing a GM that he has been sufficiently diplomatic, that player's character's abilities are "not predicated on a finite number" (ability score or skill number)?

With respect to this I don't have a lot to say except you are correct. I cannot pick an choose which "abilities" for lack of a better word to model and still meet my goal of giving players total creative liscense over their character. Fortunately this can be easily fixed, I think.

I've got to say this is tough to do but revealing and useful. I suspect you've done this more than once with a budding gung-ho designer :)

[TonyLB]

The GM is supposed to extrapolate what key conditions present at the time the character attempts their task have a significant impact on the character's chance of success.

So, for instance, if a thief is running across rooftops then the GM determines how slippery and badly made they are, based on what's been narrated about the environment, yes?  If, for example, it's already been narrated that it's raining then the difficulty is higher, because the slates are slippery with rain.  Fine.  Good.

Now who originally narrated that it was raining?  I'm guessing "the GM."  So, really, the GM isn't a judge of some external world ... he's narrating difficulties, and then assigning them a number, but he's responsible for many (if not all) stages of that process.  He's choosing the difficulties to apply against a given player's efforts, from the first steps of narrating them through to the final step of applying a number as a penalty.

What I'm asking is "On what basis does he choose those difficulties?"  Here's some examples of systems that explicitly answer that question:

• My Life with Master:  The Master chooses difficulties with an eye toward crushing the spirit of his minions ... sometimes by making things impossibly difficult, sometimes by making the things the minions don't want to do ludicrously easy.
• Dogs in the Vineyard:  The GM chooses from a limited set of resources with the intention of persuading players to take as much Fallout as he can convince them to.
• Capes:  All players provide adversity with the intention of making a conflict so hard that other players need to take a moral stake in order to be strong enough to defy them.

So, back to my example of the thief.  Here's how a GM in the various systems would choose whether or not it was raining:

• MLWM:  If the minion is climbing toward the window of a beloved in order to give her flowers then it's raining ... and he gets struck by lightning.  If he's climbing toward the window of a beloved because the Master has ordered him to kidnap her for devious experiments then the sky is clear and the footing reliable.
• DitV:  Is the difference in difficulty going to convince the player to take some Fallout dice?  Then it's raining.  Is it going to convince them to just Give, and avoid further difficulty?  Then the sky is clear.
• Capes:  Is the other player utterly determined that, no matter what, they'll climb the roof?  Then it's raining ... and climbing this roof is going to make him late for his date with the girl next door.  Is the player willing to lose the conflict, rather than make a moral stand?  Then the sky is clear.

Does this help make it more clear what I'm trying to get at?

[John Griffith]

You probably ought to read the following design patterns in the free book Design Patterns of Successful Role-Playing Games, by John Kirk III: Attribute, Skill, and Trait.

Before we really go much further about redesigning your skill system, can you tell me a bit about the resolution system and how these skills will be used? Do you understand the difference between a task resolution system and a conflict resolution system? If so, does your system resemble one of those?

Well admitting that I had to look up a definition for task resolution vs. conflict resolution, yes I understand the difference.

My system resembles task resolution.

For example, to use an rpg cliche:

A character is exploring a dungeon and comes across a door that is locked. They want to move through the door with as little fuss as possible so they get out their lock picks and go to work on the door. They have the skill Lock picking which allows them to use a set of special tools to manipulate a door's lock and open or close it. The lock on the door will have a set level of complexity describing how difficult it is for the lock to be opened without a key. The relationship between the character's level of ability in Lock picking and the complexity of the skill determines a target percentile. The character makes a percentile role. If the role is equal to or less than the target percentile the lock opens.

[John Griffith]

Tony, yes I see what you are getting at. And you make a good point. Here's my answer or at least my attempt to answer.

Taking into account every single factor that would influence success/failure would be not just arduous but also impossible. The idea is to choose the key elements that influence success/failure. These key elements come from a pool, essentially a list. If I take your example of the thief running across the rooftops:

In this case as you point out the GM is indeed narrating many of the difficulties. He is narrating how steep the roof is, whether there is enough wind to hamper the thief's progress, the natural slipperiness of the roofing material, and whether or not it is raining making it even slicker. In these instances if a decision had not been made prior to the thief's decision to run across the rooftops it would become a very subjective decision at the whim of the GM. If the GM had previously determined that rooftops in this town are flat and made of coarse slate, that today it was storming hard--both wind and rain--then it would be a less subjective decision. To be honest I'm not sure that there is an answer to the subjective decision question as it seems that it would be up to the GM. Do they want to make this task easier or harder? And why? Is there an answer to that question?