Pitch F/X Audit: Washington Nationals
October 21, 2008 2 Comments
With only the World Series left to play, the 2008 major league baseball season is very quickly going to come to its official close. Last year, Pizza Cutter ran a series titled “Sabermetric Year in Review,” during which he wrote detailed pieces examining each team from various statistical perspectives. With the advancements made in Pitch F/X data I thought it would be interesting to add to the year in review gameplan by “auditing” each team from a few different angles. As I mentioned last week, this is a work in progress and I am more than willing to solicit feedback regarding changes, additions, or subtractions, but the ultimate goal here is to examine the repertoires of different pitchers and how they potentially change either with men on base, against different handed batters, or as the game progresses.
I will be following the same reverse pythagorean order as Pizza Cutter, so whichever team he profiles, expect to see a Pitch F/X year in review for that team the subsequent day. That means, to kick off this series, we will be looking at the Washington Nationals. I am not going to be compiling data for every pitcher that logged an inning this past season, but rather the pitchers who had significant playing time and/or are going to be big parts of the staff moving forward. There are going to be some pitchers not profiled due to lack of significant data, which, for the Nationals, means no Shawn Hill numbers this time around.
The Nationals pitchers that will be discussed here are: John Lannan, Tim Redding, Odalis Perez, Jason Bergmann, Collin Balester, Charlie Manning, Saul Rivera, Jesus Colome, and Joel Hanrahan. There will be four sections of data, each with a summary and a few links to tables of data. I cannot stress enough that if any of you readers have ideas for new segments, or suggestions of different ways to look at the pitchers from here on out, please do not hesitate to contact me, as my goal here is to offer you information that may otherwise be very difficult or time-consuming to compile yourselves.
Overall Velocity and Frequency
Before we get into any crazy splits, let’s take a look at the overall usage patterns of different pitches for the staff, as well as the average velocities for those pitches. To view the frequency and velocity data, click here.
Amongst the starting pitchers, Lannan, Redding, and Balester all threw predominantly fastballs, coming just short of accounting for 2/3 of their deliveries with the heater. Jason Bergmann threw 55% fastballs, mixing it up with 21% curveballs, 17% sliders and 6% changeups. He occasionally mixed in a splitter as well. Odalis Perez, however, was very interesting, as he threw a fastball, slider, and changeup, but mixed these deliveries much moreso than his rotation counterparts. The fastball accounted for just under 42% of his total pitches thrown, with the slider being delivered 30% to go along with 28% changeups.
His 30% sliders were not the most on the rotation, though, as Tim Redding barely edged Perez out with 31% sliders. Redding utilized two pitches throughout the season, with the fastball and slider accounting for 94% of his total pitches thrown. Perhaps he would be better suited for the bullpen with only two pitches, as starters tend to need that third pitch to succeed. Speaking of the bullpen, you should notice that three of the five pitchers covered were just like Redding, with the fastball and slider encompassing just about all of their pitches. Shell, who made 39 appearances with a 2.16 ERA, mixed his pitches a bit more, incoporating a slider and changeup every so often to complement his fastball-curveball tandem. Saul Rivera virtually split his pitches evenly, with 34% fastballs, 32% sliders, and 33% changeups.
When looking at how the pitching staff split pitch frequencies and results between different handed hitters, I opted to go with the four primary starters: Lannan, Redding, Perez, and Bergmann, as their splits would all feature significant sample sizes of pitches thrown. To view the data for these four starters, click here.
Lannan threw his fastball much more often against righties, while the other three were more balanced in their usage of the heater against lefties and righties. In fact, Jason Bergmann had the second biggest fastball usage discrepancy, and his was under five percent of a dropoff against righties. Lannan was able to get more swings and misses against like-handed lefties than righties, as well as more foul balls. Redding had the same results, but against like-handed righties.
Odalis Perez recorded the highest percentage of swings and misses against left-handed hitters, but did not fare as well against opposite-handed righties. Other than his called strike and swinging strike rates, though, all other results stayed virtually the same, meaning that swings and misses were replaced with no swings, but strikes still resulting. Bergmann received just 5.9% swinging strikes from lefties but doubled that to 11.8% against righties, the highest of any Nationals starter against either LHH or RHH. Lefties also put 22.5% of his offerings in play.
Splits Throughout the Game
Something I have always been curious about, but never had the means to research before is how, or if, the repertoire of pitchers changes as the game progresses. For instance, will a pitcher abandon his curveball after the first few innings and throw a majority of fastballs? Or will he split his deliveries more evenly as his “stuff” dies out from fatigue? Or, perhaps, nothing changes. To see how some of the repertoires shifted, click here.
Overall, Nationals starting pitchers did not get deep into games too often, as the sample sizes from the 7th inning onwards are extremely small. Lannan utilized the slider and changeup virtually the same percentage of the time from innings 1-3 to 4-6, but replaced 5-6% of fastballs with curveballs. In the few outings that ventured deeper, he threw more sliders than before. Redding did not differentiate his deliveries too often, sticking with the fastball-slider tandem. Balester’s only real shift came in innings 4-6, when he replaced 4% of his fastballs with a slight uptick in curveballs and ~3.5% of changeups.
Odalis Perez’s splits were fairly consistent as the game progressed, slightly decreasing his already reduced fastball usage in the later innings. Bergmann threw close to 58% fastballs in the first three innings, but in innings 4-6, spread the wealth amongst offspeed pitches by increasing his usage of curveballs and changeups. Later on in the games, however, he barely threw any changeups.
Bases Empty/Men On
Another interesting aspect of pitch frequency or the percentage of certain pitches being thrown deals with what occurs with runners on base as opposed to the bases empty. Perhaps the hard-throwing reliever is more reluctant to throw his slider with runners on out of fear that it might bounce away as a wild pitch. For this split, I used all ten pitchers, and to view the results, click here.
With the bags empty, Lannan threw his fastball much more, while he increased usage of breaking and offspeed pitches with some ducks on the pong. The same can be said of Redding, who dropped from 68% fastballs with nobody on to 61% fastballs with bags occupied, simultaneously increasing his slider usage from 25% to 31%. Balester’s split was even more dramatic, as he threw 70% fastballs with nobody on but just 59% out of the stretch. This theme was again repeated with both Odalis Perez and Jason Bergmann, meaning the entire starting rotation relied less on the fastball with runners on base.
The bullpen experienced slightly different results. Hanrahan, Shell, and Saul Rivera each threw less fastballs with runners on than not, but Jesus Colome and Charlie Manning went in reverse, relying moreso on the heater with ducks on the pond. Then again, Manning and Colome were two pitchers who threw 64%+ fastballs, and Colome does throw 95 mph, so his usage is not too surprising. Manning does not throw that hard, however, averaging around 87 mph, so his reliance on the fastball seems to be successful due to the effectiveness of his slider.
A starting rotation of Lannan, Redding, Perez, Bergmann, and Balester is probably not intimidating to several AAA teams, let alone teams in the major leagues, so they are going to need to bring in or develop some pitching prospects. Lannan is at best a #3 starter, so being relied upon as an ace is a big no-no. Redding will need a solid third pitch to stick around for a few more years, and Odalis is nothing more than a stopgap veteran to provide “leadership.” Joel Hanrahan looks like the real deal as a reliever, and Steven Shell and Saul Rivera did not look half bad either. Manning had very poor peripherals this past season, but he is young and could develop. Colome is only sticking around for the same reason Jorge Julio is consistently given work: he throws hard. The Nationals’ bullpen isn’t as big of a problem as their rotation, especially considering that Chad Cordero should be back in action next season. Their rotation, like their offense and defense, though, needs significant work.
Another aspect of the Pitch F/X data I am compiling but did not report just yet is the standard deviation of release points. My hope is that, when a few more of the teams are audited, I can post league leaders in several different categories, such as release point consistency, swinging strike percentage against left-handed and right-handed hitters and such. If anybody has ideas for more categories to include, let me know.