I take every player with 300 or more PA in 2006 and determine his BB/PA, K/AB and HR/(AB-K). This process is repeated for every player with 300 PA in 2007, 8, 9 and 10 as well. All of these are lumped into one big group. I find the standard deviation in BB/PA, K/AB, and HR/(AB-K). A player’s “z-score” for BB/PA is his BB/PA-lgBB/PA in the season in question divided by the standard deviation of BB/PA derived from this process. Likewise for the other to rates.

]]>Can you describe this a little more. Standard deviation of what?

Thanks

]]>Some might think he’s in good company with that comp list. Personally, I think that comp list is lucky to be compared with him. Just my present-day bias though…

]]>Very nice article.

]]>