Andy Sonnanstine and the Need to Work Ahead

In 2008, Andy Sonnanstine had the best season according WAR for any Rays pitcher ever not named Scott Kazmir or James Shields.  In 2009, it seemed nothing went his way. RJ has looked at overall pitch location and whiff rates by pitch type with nothing concrete emerging. It's time for a new theory for the Pitch F(x) team to attempt to verify.

For a "command" pitcher like Sonny, working ahead in the count is crucial. In 2008, his first pitch strike% was 64%, good for 10^th among all qualified starts in Major League Baseball.  This season that % dropped to 56.8%. His overall strike % had a lesser decline from 67.1% to 64.7%. While he was still a "strike thrower," the 35.3% balls may have been coming earlier in counts.  In 2008, 35.7% of Batters Faced began the count 1-0. This figure jumped to 43.2% in 2009. What effect does this have on hitters? They are less likely to chase when ahead in the count. Sonny's O-Swing % fell from 25.4% in 2008 all the way to 21.0% this season. Falling behind can be a very dangerous thing when you have a career contact in the zone % of 89.4%.   

Working ahead in the count has always been important for Sonny, but it has always been something he has been good at.  Consider Sonny's career splits in batting average and slugging % based on hitter/neutral/pitcher counts. We will consider hitters counts as 1-0, 2-0, 3-0, 2-1, 3-1. Pitcher counts include 0-1, 0-2, 1-2, 2-2. Neutral counts are 0-0, 1-1, 3-2:

Sonny has always been somewhat of a contact pitcher with a consistently higher BABIP. His strength is command, arm angles, multiple pitches and a low walk rate.  It's not surprising to see higher numbers across the board for Sonny relative to league average numbers. However a career BA/SLG/ISO split of .371/.661/.290 on hitter counts mandates that Sonny work ahead of the hitters.  How many Plate Appearances ended on what type of counts in 2009?

Both pitcher and neutral counts declined in favor of an increase in hitter counts. This is a disaster for Sonny. The last thing I wanted to look at today is the year over year change in 3 ball count results:

Sonny was a total failure at battling back in the count in 2009. He walked 34.3% of Full Counts, almost twice the previous season while his full count strikeouts fell from 24.3% to 5.7%. None of the 5.7% started out at 3-0 or 3-1. How much of this dramatic shift was a conscious managerial decision based on opponents batting success in hitting counts versus Sonny or his own sub-conscious nibbling cannot be determined. What is clear is that Sonny has always had a higher BABIP and ISO than the average pitcher and the difference is exaggerated on hitter counts. We know Sonny fell behind in the count more frequently in 2009 putting himself in dangerous situations for himself. Finally, Sonny walked a much greater percentage of hitters once he allowed 3 balls in 2009. I'd be curious what Pitch F(x) would tell us about first pitch selection, location, and opponent reaction.