NOW LIVE! Today's the day you meet your new best friend. You don’t have to leave Wolfy behind... In 'Pets & Sidekicks' your companions level up with you!

Reply to thread

Message: <blockquote data-quote="pemerton" data-source="post: 6851664" data-attributes="member: 42582">I didn't say that it is.I said that the number 100 is twice the number 50. Which it is.Similarly, 18 STR is not six times 3 STR, though 18 is six times 3.Why are you stating my point back to me as if I need it explained?Once that number is assigned to an ability score, it is some kind of measure. For instance, p 7 of the 5e Basic PDF tells us that "The Ability Score Summary table provides a quick reference for what qualities are measured by each ability". I think it is obvious that this is not a measurement of a determinable property (ie it is not like a measure of length, mass, duration, etc) but it is a type of ranking measure where the steps in the ranking are demarcated by a rough notion of frequency in the population.No. It's like saying that there's .5% of people capable of finishing first in a race, and that if you roll an 18 on 3d6 for you SPEED stat than your PC is one of those fast people.And not only is that something that it makes sense to say, it's something that is actually done in at least some RPG stat generation systems.I don't understand your repeated concern with means and SDs. No one else in this thread is talking about them, as best I can tell.A roll of 3d6 will produce one of 16 possible natural number results. Those results can be put in a ranking (corresponding to the counting sequence of those numbers, from 3 to 18). Each of those numbers also has a likelihood of occurring.Without getting too far into the philosophy of probability, I think it is acceptable to treat those likelihoods as corresponding, at least in rough terms, to frequencies within the population. That is to say: given that the result of 18 is the highest result in the ranking; and has a likelihood of occurrence of about 0.05%, it is quite reasonable to stipulate that a character with an 18 INT has the degree of intelligence that is found in the most intelligent 0.05% of the population.And similarly for each other score, all the way down to 3 where (given that it is the lowest ranking) we stipulate that an intelligence of 3 corresponds to the degree of intelligence that is found in the least intelligent 0.05% of the population.The above exercise has nothing to do with means and standard deviations. It is simply about probabilities, calculated using simple combinatorics.This is the exercise that  [MENTION=23751]Maxperson[/MENTION] is performing when equating a certain INT score to a certain IQ. The exercise is in no way based on any view about the mean or the SD of any sets of (actual or notional) data. And  [MENTION=6787503]Hriston[/MENTION]'s objection is quite simple: namely, that Maxperson is stipulating that the roll of 18 corresponds to a degree of intelligence found in far less than 0.05% of the population, and hence that Maxperson is carrying out the stipulation task in a manner that contradicts the parameters that he has set for himself. This criticism also is in no way based on any view about the mean or the SD of any sets of data.</blockquote>

Verification