Sunday, May 02, 2010

New Stats vs Old Stats

This past off-season, the Yankees acquired CF Curtis Granderson from Detroit and sent one of their promising prospects, Austin Jackson, to Detroit. Now Jackson is playing in the majors and had a great April, or perhaps, a lucky April.

By conventional stats, he has done very well. A summary can be found in this ESPN article, where he is hitting with a BA of .356 and has 37 hits (leads the AL).

But we look at things another way, and maybe this is nothing to get too excited about. Of all sports, baseball is one where people always warn about sample size. The Pinstriped Bible pointed this out a few weeks ago with Jackson. Jackson is striking out at a rate of 29%, which is not good, and he's on pace for roughly 200 K's over the year. His BABIP (batting average for balls in play) is .514. The BABIP is the % of balls that are hit that end up as hits and not outs. Jackson's .514 is extremely high (Ichiro has a BABIP of .357 for his career). So we're saying here that Jackson had a very lucky April but not necessarily one that will predict his future.

I like what the new statistical approach is suggesting here. We already know that BA and hits are stats with some limitations. The newer stats also have limitations but I think it will be interesting to see what happens with Jackson the rest of the year in these terms. I'm a little surprised that, which usually posts more in-depth insight, ran the Jackson has a BA of .356, therefore the trade is a bust story.

No comments: