Here’s a statistical analysis  of the career of Roger Clemens, purporting to show that his career trajectory is similar to that of Randy Johnson, Curt Shilling and Nolan Ryan, three people who presumably did not take steroids. The point is that it’s not THAT unusual to be successful in your 40’s and that Clemens’s performance didn’t spike when people think it did. The analysis is done by the agents who represent Clemens in contract negotiations.
Is it a convincing analysis or an example of how to lie with statistics? I’m only reporting on this one. You decide. I like the charts, though. Nice use of color to make the case.
UPDATE: J.C. Bradbury does his own analysis of Clemens  (HT: Matt C.) and shows that Clemens strikeout and home runs relative to the rest of the league declined with age as you might expect if he weren’t a user (or if steroids make no difference). There’s no obvious spike relative to the rest of the league. Interesting. He does not look at the key questions of whether the decline is slower than for other pitchers or relative to the past. I would assume that all players are working harder to stay in the major leagues because the returns are higher to baseball than they used to be. On the other hand, younger players are working harder to make it into the league, so maybe there’s no difference.