from Amazon
Friday, January 30, 2009
Michael Schell
Ok, I finally got the book, and started reading it last night. My kind of book so far. He also sounds like he’d fit right at home among the readers here.
I’m going to have problems with the underlying assumptions (that Schell explicitly discusses). The first is that he’s treating the mean as equal for each time period. Only among MLB fans is the average player in 1908 equal to the average player in 2008. In every other sport, the greatest players of all time are virtually all within the last fifty years. But, not in baseball. Anyway, this is not a slight on Schell, as he simply does what most people do, and treat the average as being the same year-to-year.
The second one, though, is a problem. A big problem. He forces the distribution of players around that mean to be the same, year to year. (Or he will anyway. I’m only on page 11 so far.) This would kind of make sense if the types of players you have in baseball is the same every year. But, this is not true. There were gazelles in MLB in the 1970s and 80s, while the 1990s and 2000s saw a rise in power hitters. I can’t believe that the 1900s and the 1930s had the same kind of players. Not to mention that you had 16 teams at one point, and then 30 now. You have expansion that may or may not have kept up with population growth. It certainly has not in the last 40 years, as the number of US-born players is the same over the last 40 years, with the increase number being handled by foreign-born players. Then of course you have the influx of Black players over the last 60 years.
But, worse than that, worse than forcing the distribution to be the same, is that he does it at a component level. Doubles+triples and walks and home runs are treated independently, as if you can fix the distribution to be the same era-to-era without accounting for the fact that there is a relationship between each component. Indeed, the component-level transformations seems to be the heart of the entire process. He would have been better off doing it in a “binary” fashion, the way we’ve discussed the issue in the past, as far back as when Voros introduced DIPS.
But, Schell does explicitly make his assumptions clear, which allows a reader like me to point out the flaws so readily. Weird as it sounds, I’m going to enjoy reading the book, and I have a feeling I’m going to recommend it as well. Feel free to post your own thoughts or reviews, and if my initial impressions are invalid.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date