We talk a bit in our book about drawing inferences from small samples. The playoffs in sports are a good example of a small sample that is insufficient to answer the question “who is the best?” Although small samples can be misleading, when the sample points in the direction you wish to go, well, you just have to ignore the rules of statistical inference.
So far the Detroit Tigers, the team I have followed since my childhood in the Motor City, are 20-10. According to ESPN.com, Detroit’s expected winning percentage – based on “Bill James' Pythagorean theorem of baseball: Runs scored [squared] / (Runs scored [squared] + runs allowed [squared])” — is currently the best in Major League Baseball. Given a sample of thirty games, I am prepared as a fan of Detroit to declare that the Tigers are indeed the best team in baseball. Perhaps even more important, if the Tigers “dominance” continues – which I do not believe will happen – then the best team in baseball will once again lag in baseball’s payroll rankings. According to The Hardball Times, the Tigers are out of the top ten in payroll, and pay their players more than $100 million less than the dreaded Yankees.
– DJ
Wages of Wins Journal 
Posted on May 6, 2006 by dberri
0