Roleplaying skill system
Nov. 19th, 2008 08:37 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
jackIntro
In most mechanics-based roleplaying games, outside of any specific tactical arena (such as combat for a DnD fighter, or talking/trap-finding for a DnD rogue, where abilities tend to be tightly defined and optimised), there's some sort of skill system which mediates acts like "drive fast" or "bluff a bouncer" or "climb a cliff".
In some cases, the skills shouldn't come into play, because the result should obvious, or irrelevant. However, if the GM wants to introduce some forking possibilities, and allow the characters to differentiate themselves mechanically, it comes down to a skill roll in most games.
Many games
Suppose I'm a medium skilled showjumper who wants to ride a horse into battle. This normally goes something like:
* A "riding" skill is defined, describing how hard various tasks are. Eg. Difficulty 10 = "Go for a trot", 15 ="ride a showjumping course", 20 = "Ride into battle and don't fall off", 25 = "leap a horse-and-four across a wide canyon"
* I get 5 points in the ride skill.
* My innate strength and dexterity add some modifier
* I roll a dice, add 5, and try to score more than 15.
There's a few obvious observations:
* You might equally well say "Roll a dice, and try to score less than 5 plus something" or equivalent. The trouble is, it's easy to forget the something, and just assume that every skill check is at "average" difficulty
* Some games adjust the variance by having you roll multiple dice. However, there's generally a fixed relationship between variance and skill, and this is often completely constant. DnD has a constant variance, except when you're unflustered and have no risk from failure, when the variance is zero.
* Some games offer an arbitrary cut off. Eg. if you don't have any skill in lock-picking, you can't pick locks AT ALL. However, this is generally left to GM common sense except at the lowest-level of the ability, and often forgotten.
* The GM is often in the difficult position of making up a difficulty target score for a given activity, and can't get it to be a reasonable level for both a normal human, and the character in question.
In a tactical situation, the difficulties of the taks are often well laid out mechanically, and you want to compare how well different players did, so it makes sense to have an overall "how well you did at that skill" result. But if not, I wonder if it would make more sense to let the modifiers affect the target.
A modest proposal
Suppose "ride skill 5" meant not "25% more chance of succeeding at any task while flustered, and certainty of succeeding at 25% more tasks while calm", but "can normally ride [a showjumping course]".
Ride skill 1: Can go for a trot.
Ride skill 5: Can ride a showjumping course
Ride skill 10: Can ride into battle
Since the skill levels are mostly made up on the spur of the moment anyway, this doesn't make it any harder to adjudicate, and if you want a quick book answer, you can say "stand a 50% chance of succeeding at task of appropriate skill level, gain/lose 25% for each difference in level", and have exactly the previous behaviour.
However, there seem to be a couple of advantages:
* If I'm very much better at a skill than a normal human, you can work out what that means when you choose that skill, rather than trying to define it in terms of how many 0s there are in "a normal person has a 0.0000001% chance of succeeding at..."
* The variance is up front in the die result, and the system encourages to adjudicate on the fly what %s of success are possible. "A normal person goes for a trot? 50% chance of success, 10% chance of hurting themselves" and "Skill 10 goes for a trot? Auto-success." and "If you are trained to ride into battle, and do so? 1-20 = fall off anyway, 20-40 = minor injury, 40-60 = ok, 60-80 = particularly well, etc".
For important, tension-building rolls you can specify the results before-hand, or for minor rolls you can roll the die, get a "good" or "average" or "catastrophic" result, and see what that translates to in the circumstances: some situations are plainly success/fail, others graduated, others mostly all the same except for exceptional success or exceptional failure.
Does it make sense?
In most mechanics-based roleplaying games, outside of any specific tactical arena (such as combat for a DnD fighter, or talking/trap-finding for a DnD rogue, where abilities tend to be tightly defined and optimised), there's some sort of skill system which mediates acts like "drive fast" or "bluff a bouncer" or "climb a cliff".
In some cases, the skills shouldn't come into play, because the result should obvious, or irrelevant. However, if the GM wants to introduce some forking possibilities, and allow the characters to differentiate themselves mechanically, it comes down to a skill roll in most games.
Many games
Suppose I'm a medium skilled showjumper who wants to ride a horse into battle. This normally goes something like:
* A "riding" skill is defined, describing how hard various tasks are. Eg. Difficulty 10 = "Go for a trot", 15 ="ride a showjumping course", 20 = "Ride into battle and don't fall off", 25 = "leap a horse-and-four across a wide canyon"
* I get 5 points in the ride skill.
* My innate strength and dexterity add some modifier
* I roll a dice, add 5, and try to score more than 15.
There's a few obvious observations:
* You might equally well say "Roll a dice, and try to score less than 5 plus something" or equivalent. The trouble is, it's easy to forget the something, and just assume that every skill check is at "average" difficulty
* Some games adjust the variance by having you roll multiple dice. However, there's generally a fixed relationship between variance and skill, and this is often completely constant. DnD has a constant variance, except when you're unflustered and have no risk from failure, when the variance is zero.
* Some games offer an arbitrary cut off. Eg. if you don't have any skill in lock-picking, you can't pick locks AT ALL. However, this is generally left to GM common sense except at the lowest-level of the ability, and often forgotten.
* The GM is often in the difficult position of making up a difficulty target score for a given activity, and can't get it to be a reasonable level for both a normal human, and the character in question.
In a tactical situation, the difficulties of the taks are often well laid out mechanically, and you want to compare how well different players did, so it makes sense to have an overall "how well you did at that skill" result. But if not, I wonder if it would make more sense to let the modifiers affect the target.
A modest proposal
Suppose "ride skill 5" meant not "25% more chance of succeeding at any task while flustered, and certainty of succeeding at 25% more tasks while calm", but "can normally ride [a showjumping course]".
Ride skill 1: Can go for a trot.
Ride skill 5: Can ride a showjumping course
Ride skill 10: Can ride into battle
Since the skill levels are mostly made up on the spur of the moment anyway, this doesn't make it any harder to adjudicate, and if you want a quick book answer, you can say "stand a 50% chance of succeeding at task of appropriate skill level, gain/lose 25% for each difference in level", and have exactly the previous behaviour.
However, there seem to be a couple of advantages:
* If I'm very much better at a skill than a normal human, you can work out what that means when you choose that skill, rather than trying to define it in terms of how many 0s there are in "a normal person has a 0.0000001% chance of succeeding at..."
* The variance is up front in the die result, and the system encourages to adjudicate on the fly what %s of success are possible. "A normal person goes for a trot? 50% chance of success, 10% chance of hurting themselves" and "Skill 10 goes for a trot? Auto-success." and "If you are trained to ride into battle, and do so? 1-20 = fall off anyway, 20-40 = minor injury, 40-60 = ok, 60-80 = particularly well, etc".
For important, tension-building rolls you can specify the results before-hand, or for minor rolls you can roll the die, get a "good" or "average" or "catastrophic" result, and see what that translates to in the circumstances: some situations are plainly success/fail, others graduated, others mostly all the same except for exceptional success or exceptional failure.
Does it make sense?
no subject
Date: 2008-11-19 09:06 pm (UTC)no subject
Date: 2008-11-19 11:28 pm (UTC)no subject
Date: 2008-11-20 12:40 am (UTC)no subject
Date: 2008-11-20 01:16 am (UTC):)
no subject
Date: 2008-11-20 12:57 am (UTC)Of course, there's room for some cinematography. If you want some drama, you can roll to get a critical failure putting on one's trousers ("ow"), or a critical success going to the shop ("On the way..."[1] :)).
[1] Of course, that's mostly just funny, it's normally a bad idea to allow critical successes on everything, or it just gets silly :)
no subject
Date: 2008-11-19 09:13 pm (UTC)http://www.niksula.hut.fi/~juuso/domus/rules/fudge.html