prior probability

Did you update your priors after Game 1 of the Finals when Kyrie Irving joined his teammates Kevin Love and Anderson Varejao with a season-ending injury?

Love and luck (both good and bad) are in the air …

Via the excellent kottke, check out this video featuring Lonnie Mimms, a self-described computer geek who collects old Apple computers and is creating an “Apple Pop-Up Museum” in an abandoned CompUSA store. “It’s never complete,” he says about his massive collection of vintage Apple electronic products …

That is the title of our most recent work in progress in which we combine the art of legal rhetoric and reasoning (applying facts to rules) with the mathematics and logic of fuzzy sets. In brief, a “fuzzy” or multi-valued concept is a quantitative variable that lies within a probable range. Variables like temperature or height, for example, could be “high,” “medium,” or “low.” Likewise, whether or not a legal rule (or legal standard) applies to a given fact pattern is a often matter of degree, rather than a clear-cut or definitive “yes” or “no.” By way of comparison, consider the basic legal doctrines of “intent,” “negligence,” and “ordinary care” in the law of torts. We would argue that such legal concepts are inherently fuzzy or multi-valued in a mathematical or logical sense. Whether a particular defendant acted with the requisite intent or was negligent in a given situation is often a matter of degree. Thus, what if judges and juries had the option of assigning a real number to rate the truth of a moving party’s legal and factual claims in civil or criminal cases. This real number must lie within a range from 0 to 1, where 0 means that there is no evidence in support of the plaintiff or prosecutor’s case, and 1 means that he has proved his case with absolute or complete certainty, while values greater than 0 but less than 1 represent various levels of truth of the moving party’s case. In other words, when deliberating on whether a plaintiff or prosecutor has met his burden of proof or otherwise proven his case, we would propose that a judge or jury render a “fuzzy verdict” by assigning a numerical value to the strength or weakness of the moving party’s case … This is just a work in progress, but are we on the right track or totally off base?