1.89 Seasons: A Baseball Experiment About Hiring and Human Judgment
Nearly every evaluation system in wide use answers the same question: how does this person compare to people like them? Years of experience, school, sector, title trajectory. A statistical claim about a category, applied to an individual who happens to be standing inside it. I have spent this year arguing that the hierarchy is backwards: the population is a prior, not a verdict, and its weight should move as an individual's own evidence accumulates. This week I stopped arguing and measured.
Not on hiring data, because there is no public hiring data worth the name: I could not find a single public dataset with the repeated, standardized, outcome-linked observations this test requires. So I went to a large public dataset that has exactly that: the Lahman Baseball Database, season by season, the licensed statistical record of a public profession. Baseball is also where a version of this argument played out...
Copyright of this story solely belongs to hackernoon.com. To see the full text click HERE
