Loading ...
Sorry, an error occurred while loading the content.

The NY Times article is interesting ....

Expand Messages
  • jsm_44092
    ... As both an athlete and a statistician, it is nice to know that athletes may play like statisticians . The article is a bit of a stretch, however, because
    Message 1 of 1 , Jan 20, 2004
    • 0 Attachment
      ... As both an athlete and a statistician, it is nice to know
      that "athletes may play like statisticians". The article is a bit of
      a stretch, however, because the use of prior information in a very
      general and perhaps vague manner is quite different from a formal
      Bayesian approach. As Steve Stigler, the leading expert on the
      history of statistics, charitably put it in referring to the way
      people use probability: "The idea that it's associated with a
      Bayesian approach is not quite clear".

      Basketball and football teams, in particular, study game films so
      that they can learn the tendencies of their opponents. They will use
      this information in a rather general way when they play those
      opponents. They will not use a formal Bayesian approach.

      Bayes's Rule is straightforward and is simply an application of
      conditional probability. Although computing power has indeed made
      certain Bayesian approaches quite feasible, and such approaches can
      be very useful, I doubt that we are on the verge of seeing a Bayesian
      approach being the "majority viewpoint", as Brad Carlin, a dyed-in-
      the-wool Bayesian, has asserted.
    Your message has been successfully submitted and would be delivered to recipients shortly.