Reply to thread

Message

<blockquote data-quote="AbdulAlhazred" data-source="post: 9026033" data-attributes="member: 82106">My understanding is that ACTUAL efforts have been made in terms of quantifying these sorts of things. In fact there's rather a rich literature. Its not ALL probabilistic, but Monte Carlo methods would certainly be a likely approach in terms of picking an outcome. However, as the entire sad history of 'Intelligence Snafus' indicates, even very well-funded and highly trained organizations with much motivation have a poor track record at predicting outcomes in these sorts of complex situations. I won't refer to any current events, but you needn't look far to see the leadership of modern nations massively miscalculating basic facts about their neighbor's behavior... What I gather from this is that, at least beyond some fairly limited threshold, the sheer number of variables and lack of quantified values for them alone will preclude forecasting that is much better than pure guesswork and just adjudicating something that 'seems likely'. Certainly if the CIA cannot get these sorts of things right, your local dungeon master is completely overmatched by the task. So in terms of deployment within RPGs, IMHO this sort of idea is a non-starter. We make stuff up, and we give it some basic plausibility. So the question really is; what are the actual criteria that are used to decide between plausible outcomes A and B? I would volunteer that those are going to depend on the agenda of the game in question.</blockquote>

[QUOTE="AbdulAlhazred, post: 9026033, member: 82106"] My understanding is that ACTUAL efforts have been made in terms of quantifying these sorts of things. In fact there's rather a rich literature. Its not ALL probabilistic, but Monte Carlo methods would certainly be a likely approach in terms of picking an outcome. However, as the entire sad history of 'Intelligence Snafus' indicates, even very well-funded and highly trained organizations with much motivation have a poor track record at predicting outcomes in these sorts of complex situations. I won't refer to any current events, but you needn't look far to see the leadership of modern nations massively miscalculating basic facts about their neighbor's behavior... What I gather from this is that, at least beyond some fairly limited threshold, the sheer number of variables and lack of quantified values for them alone will preclude forecasting that is much better than pure guesswork and just adjudicating something that 'seems likely'. Certainly if the CIA cannot get these sorts of things right, your local dungeon master is completely overmatched by the task. So in terms of deployment within RPGs, IMHO this sort of idea is a non-starter. We make stuff up, and we give it some basic plausibility. So the question really is; what are the actual criteria that are used to decide between plausible outcomes A and B? I would volunteer that those are going to depend on the agenda of the game in question. [/QUOTE]

Verification