The NY Times article is interesting ....
- ... 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.