Statistics (and statistics)
Posted: Sat Jan 14, 2012 2:05 pm
Hey guys,
So I've been wondering lately what exactly our stats mean. The effective range for a PC is between 35 and 100, but without some context, it's hard to extrapolate the numbers on score into actual feats.
I think there's a common feeling that 50 is 'average', which if so provides a starting point, but there are still some big questions there. Even knowing an average doesn't define the issue if you're not sure of the variation you can expect within the population.
So I'd like to have some benchmarks for what various levels in various stats mean, in terms that might make sense to people. Both descriptions, and information that's in familiar numerical scales when possible (the easy carrying ability of someone with a certain strength, a rough IQ equivalent for int - yes I know IQ is flawed, but it's instinctively clear to most people due to experience.)
Does anyone else think this would be helpful/want to weigh in on how they conceptualize stats?
HERE THERE BE ENTIRELY OPTIONAL MATH:
It's often considered a default expectation that variables are normally distributed - i.e, that they fit what's known as the Bell curve, with a lot of people in the middle and not a lot of people at the ends. Now, this doesn't fit our PCs, of course - but I subscribe to the theory that PCs are meant to represent exceptional, extraordinary individuals, hence why so much happens to them compared to Joe Peasant.
So, assuming a normal distribution, I think of 50 as average across the whole population, with perhaps standard deviations of 20 - what this would mean would be that 84% of people would be below a 70, and 95% of people would be below a 90. For stats nerds, yes, this does mean that 100 would be only better than 99.38% of people, but hey, let's not sweat the small stuff. ;)
So I've been wondering lately what exactly our stats mean. The effective range for a PC is between 35 and 100, but without some context, it's hard to extrapolate the numbers on score into actual feats.
I think there's a common feeling that 50 is 'average', which if so provides a starting point, but there are still some big questions there. Even knowing an average doesn't define the issue if you're not sure of the variation you can expect within the population.
So I'd like to have some benchmarks for what various levels in various stats mean, in terms that might make sense to people. Both descriptions, and information that's in familiar numerical scales when possible (the easy carrying ability of someone with a certain strength, a rough IQ equivalent for int - yes I know IQ is flawed, but it's instinctively clear to most people due to experience.)
Does anyone else think this would be helpful/want to weigh in on how they conceptualize stats?
HERE THERE BE ENTIRELY OPTIONAL MATH:
It's often considered a default expectation that variables are normally distributed - i.e, that they fit what's known as the Bell curve, with a lot of people in the middle and not a lot of people at the ends. Now, this doesn't fit our PCs, of course - but I subscribe to the theory that PCs are meant to represent exceptional, extraordinary individuals, hence why so much happens to them compared to Joe Peasant.
So, assuming a normal distribution, I think of 50 as average across the whole population, with perhaps standard deviations of 20 - what this would mean would be that 84% of people would be below a 70, and 95% of people would be below a 90. For stats nerds, yes, this does mean that 100 would be only better than 99.38% of people, but hey, let's not sweat the small stuff. ;)