philly.com
Forget youth: Phillies need to diversify their offensive skill set
Success at hitting requires the mastery of two opponents. The first opponent, the pitcher, is ever-changing. It is a variable that presents one of the most severe limitations of the evaluative abilities of even the most advanced statistics. A home run off of Chris Volstad counts the same as a home run off of Tim Lincecum. There are some ways to measure performance against pitchers of varying strength, which we will look at in a future blog post. For the most part, though, any of the standard measures we use to quantify overall success fail to accurately reflect how a batter might perform in a specific situation against a specific pitcher. This holds true for all statistics, from basic unit tabulations like home runs to ratios like on base percentage to complex approximations like WAR. For instance, all of those numbers say that Ryan Howard should be at a heavy disadvantage against a left-handed starter like Johan Santana, yet Howard is 8-for-25 with four home runs, two walks and five strikeouts against the former Cy Young winner. Those same numbers suggest Howard should hold a heavy advantage against a righthander like Tim Redding, yet Redding has held the first base man to three hits and three walks with 11 strikeouts in 28 career at-bats. From a statistical perspective, these sample sizes are too small to hold much of a predictive value. But Major League Baseball does not take sample size into consideration when it doles out championship rings. Four plate appearances against Chris Carpenter is a small sample size. Seven innings against Edwin Jackson is a small sample size. A Best-of-Five series is a small sample size. In the end, the only thing that matters is the result.