Alan Schwarz has one of his typically fine pieces in the Times today. The subject is Eric Walker, who argues that the so-called steroid era has not produced a discernible increase in power numbers across baseball, despite appearances. He blames a juiced ball for an anomalous surge in 1993-94, and uses a statistic called Power Factor (PF), or total bases per hit, to prove his thesis. PF has been steady during the steroid period. I'm wont to believe the steroid effect, whatever it might be, is over-stated by the layman, but Walker's methodology, as explained by Schwarz at least, is unconvincing. PF doesn't seem to me a reasonable indication of anything, given the complexity of the system it models and the immense number of moving parts involved (stadium size, expansion, etc.) One of the problems with steroids is that we're never going to know the true effects, because we're never going to have a controlled scientific study of what they can or can't do. That said, when a complex stat so patently fails to explain observed phenomena (Bonds, McGwire, etc.), then you need to look a bit more closely at your own methodology.
This is no defense of the Mitchell Report and subsequent Congressional furor. I can't imagine anything more sad than subpoening Chuck Knoblauch to testify against his will before grandstanding politicians. Just when it seems things couldn't get much worse, well, there always seems to be another level to which the game can sink. Ugh.
[Ed Note: Please take the time to read through the comments, as Mr. Walker took the time to respond to this post, and to some of the comments in the thread.]