Teaching Stats with Sports

FROM: Chance News #79. A great FREE publication.

GO TO: http://www.causeweb.org/wiki/chance/...Chance_News_79

At Moneyball U, what are the odds? by Alan Schwarz, New York Times, 4 November 2011


The title is a reference to the movie "Moneyball," which features the role of probability and statistics in the world of baseball. Neverthless, the story leads with the comment that "Watching a baseball telecast may not be the best way to learn basic probability." Indeed, when a good hitter has gone hitless in his last 10 at-bats, announcers can't resist saying that he is now "due for a hit." The appeal is to a mythical Law of Averages that will even things out in the short run, but but of course the real Law of Large Numbers promised no such thing. Similarly, fans will offer a variety of explanations for the much-discussed sophomore jinx, a phenomenon which can generally be accounted by regression to the mean.
Given that many students can be readily engaged by such conversations, college courses have appeared that teach statistical concepts in the context of sports. The article describes such offerings at Stanford, Ohio State, Bowling Green, Louisiana Tech, and James Madison University, among others. It is a mistake, however, to assume that all students are sports fans. The article relates an anecdote from the James Madison course. When the professor asked if we should be surprised that the last 14 opening coin tosses at the Super Bowl have all been won by the N.F.C., one student asked "What's the N.F.C.?" Fortunately, this student still seemed to appreciate the lighter atmosphere of the class. Or, as Prof. Jim Cochran of Louisiana Tech professor observes, "You want them to demand data, to look for evidence, to test hypotheses. You have students who would otherwise not come close to this discipline. It’s very valuable."
The article catalogs a variety of probability and statistics techniques that are readily illustrated with sports data. We were intrigued by a reference to Simpson's paradox, which "helps explain why the Cincinnati Reds, despite having the best record in the National League during the strike-split 1981 season, didn’t make the playoffs." Details can be found in Prof. Cochran's article Bowie Kuhn's worst wightmare (INFORMS Transactions on Education, Vol 5, No 1, Sept 2004). This is a sophisticated discussion which applies integer programming to demonstrate the possibility aggregation paradoxes. As Cochran states in the abstract, "although the case deals with a baseball-related problem, it is relatively self-contained and requires no understanding of how baseball is played, so students who are unfamiliar with the sport will not be seriously disadvantaged."