I decided I would look at the performance of starting pitchers on the same team as various "pitching prodigies" for, at most, the first two months of the same season after the pitcher joined the team as compared with their "three year averages" (with some exceptions) using the year before, year after, and same year that the prodigy joined the team. Since defining prodigy is fairly subjective, I relied on an expert in hype and used this list from Fox Sports, choosing to substitute Rick Ankiel for Dave Righetti. Besides Ankiel, the list includes pitchers such as Fernando Valenzuela, Mark Fidrych, and Hideo Nomo. To measure pitcher performance, I chose FIP, a statistic that incorporates events mostly within a pitcher's control-namely, and completely, home runs allowed, BB, IBB, HBP, strikeouts, and IP. I found FIP preferable to the traditional ERA because the latter statistic relies on many significant events outside of the pitcher's control-the defense behind the pitcher, for instance.
Being familiar with some of the major findings of sports statisticians, I was fully expecting the prodigy effect to be imaginary. I was surprised to find that during the "hype window," the two months after the prodigy starts his first game, his fellow starters saw their FIPs fall by an average of 3.4%. Maybe most surprisingly, of the 45 pitchers examined, 29 of them saw a decrease in their FIP during the hype window, for a rate of about 64%.
What could this mean for the Nats? If Nats starters saw their FIPs fall by 3.4% that would boost them up about 5 places in the Major League rankings, from 25th to 20th, moving the Nats ahead of the Phillies and out of the NL East basement-presumably just from having Strasburg on the squad. That is, the Strasburg Effect, though small, could be quite meaningful. Just one more reason to love the hard-throwing righty.
Notes: Besides the issue of actually establishing causation, there are, of course, many other caveats to these numbers. First of all, I did not employ the most scientific of methods for gathering my data and I would be surprised if I did not introduce error in that way. Second, FIP assumes HR/FB rates are under a pitcher's control, something which is most likely false, and thus the changes in FIP likely capture things beyond the pitchers's control. Third, I did not adjust for various other problems such as era, park, and age effects. In support of the reults, however, the sample size of pitchers in the hype window is fairly large, 2397 IP to be precise. This article, however, should be considered in fun and not taken too seriously.