Harvard professor "proves" that Barry Bonds is having the best offensive season ever.

This guy came up with a formula that attempts to answer the question, "How many runs would a lineup of 9 incarnations of X player produce per 9 innings?" The professor states repeatedly that this formula is not an estimate. Of course it's an estimate. All of probability is estimating.

In addition, the article neglects to mention whether the stat passed the ultimate test of any runs created formula: inputting statistics for a whole team and comparing the output to the actual number of runs. (They did it for the league and said it was close.)