Has Jonathan Sanchez Been the Same Pitcher?

As much as we at FanGraphs and other analytically-flavored websites preach against it, ERA is still the most commonly used measure of pitching performance (I would hope that we as a collective baseball fan base have dropped W-L record down on such a list). Whether it’s a Hall of Fame debate (Jack Morris) or recent transactions (Matt Garza), it’s almost natural for us to first look at a pitcher’s ERA in order to make a quick evaluation. Jonathan Sanchez has improved his ERA dramatically in the past few seasons, going from a 5.01 ERA in 2008 to a 4.24 ERA in 2009 to a 3.07 ERA in 2010. When looking at such numbers, my first inclination is that he got better every season, but I will also remember to look at peripheral statistics and stats such as FIP and WAR.

For the most part, Sanchez’s rate statistics have been relatively similar each season while his BABIP and strand rate numbers have decreased. Both of these conditions suggest that Sanchez was hit hard by bad luck in the past and has since benefited from regression to the mean. Take a look at Sanchez’s 2008-2010 seasons below:

2008: .327 BABIP, 67.5% LOB%, 5.01 ERA || 8.94 K/9, 4.27 BB/9, 0.80 HR/9, 4.14 xFIP
2009: .290 BABIP, 72.6% LOB%, 4.24 ERA || 9.75 K/9, 4.85 BB/9, 1.05 HR/9, 4.19 xFIP
2010: .262 BABIP, 79.5% LOB%, 3.07 ERA || 9.54 K/9, 4.47 BB/9, 0.98 HR/9, 4.11 xFIP

Since it appears that since Sanchez’s rate statistics and xFIP have remained pretty much the same, we can attribute Sanchez’s change in ERA to a turn of good fortunes in the form of BABIP and strand rate. I’m not convinced just yet that BABIP and strand rate are the sole explanations for his improved results.

Somewhere along the way, Sanchez may have made an adjustment in his pitch selection. Based on his season-by-season pitch type values, his fastball and changeup have improved run values in 2010.

Let’s investigate Sanchez’s pitch results based on pitch type. MLBAM classifies a slider and a curve ball for Sanchez, but the behavior of the two pitches was too similar for me to be comfortable with splitting them up, so I put both sets of pitches into one bucket. I’ve categorized this slider-curve hybrid as a slider.

FF	Usage	Ball%	CStr%	Foul%	InPlay%	SwStr%
2008:	76.8%	38.2%	18.0%	19.0%	15.5%	9.3%
2009:	68.3%	37.8%	21.0%	16.8%	15.3%	8.9%
2010:	64.4%	37.2%	22.4%	18.7%	14.9%	6.7%
SL	Usage	Ball%	CStr%	Foul%	InPlay%	SwStr%
2008:	13.4%	43.5%	6.9%	15.4%	15.7%	18.5%
2009:	23.0%	40.9%	15.8%	13.7%	14.5%	14.8%
2010:	18.9%	41.8%	12.1%	14.6%	14.8%	16.7%
CH	Usage	Ball%	CStr%	Foul%	InPlay%	SwStr%
2008:	9.6%	31.4%	15.9%	16.7%	21.3%	14.7%
2009:	8.7%	37.7%	10.9%	18.2%	19.4%	13.8%
2010:	16.7%	38.6%	12.0%	16.0%	19.5%	13.8%

There doesn’t seem to be any particular trend that jumps out here, except that Sanchez has been relying less on his fastball and more on his slider and changeup. The way in which Sanchez gets strikes differs by pitch type, but, as we saw earlier, Sanchez maintained a relatively good K rate of about nine strikeouts per nine innings. There also seems to be a decrease in fastballs put in play per pitch (InPlay% for FF), but the difference isn’t large enough to be significant.

A look at Sanchez’s batted ball numbers reveals that his line drive percentage has also decreased (21.4% in 2008, 16.2% in 2009, 14.8% in 2010) while his ground ball percentage has remained relatively constant. Perhaps Sanchez has been able to decrease his BABIP each season by relying less on his fastball and more on the breaking ball and off-speed pitches. It’s possible that the decrease in line drive rate is attributable to the decrease in fastball usage, and hence the decrease in balls in play off of the fastball. But I found that a higher percentage of Sanchez’s changeups put in play were line drives in 2008 and 2009, a percentage that was cut in half in 2010 (LD% of 2008-09 changeups was 30.9%, 15.9% in 2010). This combined with Sanchez’s increased changeup usage in 2010 (up to 16.7% in 2010) could be the main factor in his decreased BABIP in 2010.

Yet I don’t think Sanchez’s decreased BABIP is sustainable. I’m inclined to think while it’s interesting to look at a pitcher’s batted ball results by pitch type in order to examine past performance, it is probably not a good predictor of future results. So while Sanchez produced different results in ERA with different pitch selections, his decreased BABIP and strand rate, but consistent rate statistics and constant xFIP, suggest that he has been, more or less, the same pitcher since his first full season as a starter. Barring a dramatic turnaround in his high walk rate, expect an increase in Sanchez’s ERA in 2011, possibly by a full run.