September 3, 2013
A Closer Look at College Pitch Counts and Injuries
Most of our writers didn't enter the world sporting an @baseballprospectus.com address; with a few exceptions, they started out somewhere else. In an effort to up your reading pleasure while tipping our caps to some of the most illuminating work being done elsewhere on the internet, we'll be yielding the stage once a week to the best and brightest baseball writers, researchers and thinkers from outside of the BP umbrella. If you'd like to nominate a guest contributor (including yourself), please drop us a line.
Dustin Palmateer once played division III junior college baseball, finishing with a career batting average below the Mendoza Line. He now writes about the game. You can reach him via email.
Once a rallying cry for the sabermetric cause, pitch count analysis has subsided in recent years, largely because major league organizations have fully embraced a better-safe-than-sorry approach. Gone are the days of a 20-year-old Kerry Wood surpassing 120 pitches eight times in his rookie season. By contrast, Florida Marlins rookie sensation Jose Fernandez hasn’t been allowed to throw 110 pitches in a single outing this year, and he didn’t reach 100 until mid-June.
Tim Lincecum’s recent 148 pitch outing still piqued our curiosity, if only because of its rarity. A starting pitcher has thrown 148 or more pitches 65 times since 1990, and 92 percent of those outings occurred in the ’90s. Last season there were only four outings of 130 or more pitches, and 74 that surpassed 120. In 1998, Wood’s rookie year, 133 starts exceeded 130 pitches and an astounding—by today’s standards, at least—498 lasted until at least 120.
While predicting pitcher injuries remains a difficult task—incorporating considerations such as workload, pitch types, mechanics, body type, recovery time, and numerous other variables—all parties (well, most, anyway) seem to agree: fewer pitches is probably better, especially for younger pitchers who have yet to establish what kind of workload they can handle.
In this new pitch-count-sensitive age, all of a pitcher’s starts once he’s drafted and sent to the minors are closely monitored. Stephen Strasburg’s early career provides a good example of the modern-day development of a pitching prospect. Strasburg’s brief minor league stint in 2010 saw him average just 71 pitches per start in 11 outings. He quickly graduated to the majors and averaged just 82 pitches per start for the Washington Nationals in his rookie season, reaching seven innings just twice in 13 starts and never cracking the 100-pitch mark.
In 2012, his first full season after returning from Tommy John Surgery (despite the delicate handling), Strasburg was put on an innings limit and shut down in mid-September. He finished the year with 159 1/3 innings pitched and just eight of 23 starts with 100 or more pitches. Without their ace, the NL East champion Nationals went on to lose the NLDS to the St. Louis Cardinals in five games.
Before a pitcher is drafted, however, major-league teams are forced to simply observe their development, unable to alter pitch counts or mechanics, as high school and college coaches have full control. While it would be unfair to vilify all amateur coaches for having little regard for the long-term health of their pitchers, it’s certainly true that their goals don’t always align with those of a professional organization.
Austin Wood was a reliever with the Texas Longhorns in 2009, at least until May 30th of that year. Inserted into the seventh inning of an NCAA tournament game against Boston College, Wood would go on to pitch 13 scoreless innings, registering 169 pitches in his historic performance. He had thrown a paltry 30 pitches the previous day. Overshadowed BC reliever Mike Belfiore pitched 9 2/3 scoreless innings of his own before departing after throwing 129 pitches. Texas won 3-2 in 25 innings.
A starter-turned-closer, Wood was a mid-level prospect heading into the ’09 draft and was taken in the fifth round by the Detroit Tigers. Wood’s professional career was soon derailed by shoulder surgery that caused the left-hander to miss all but 1 1/3 innings of the 2010 season. He returned to post solid numbers out of the pen in 2011 at Double-A Erie, but abruptly retired from baseball in 2012. It’s not too much of a stretch to connect Wood’s 169-pitch effort to the shoulder surgery that took place just one season after that heroic performance (although Wood felt that the outing had little to do with his surgery and had only kind words for his former UT coach Augie Garrido).
Belfiore was more highly touted than Wood in 2009 and went 45th overall to the Arizona Diamondbacks in the draft. He was originally converted into a starter by Arizona, but returned to the bullpen in 2011. Currently with Triple-A Norfolk in the Orioles organization, the 24-year-old Belfiore appears poised to at least earn his first cup of coffee with Baltimore. Unlike Wood, Belfiore hasn’t spent any significant time on the disabled list.
What’s what makes analyzing pitcher usage so difficult: two different pitchers, both clearly overworked on a spring night back in 2009, one quickly encountering career-altering arm trouble and the other staying largely healthy while finding moderate success on the mound. But what if, instead of just a couple of handpicked cases, we could look at a larger sample of college pitchers? Do overworked college pitchers consistently break down with injuries in pro ball? Do more conservatively managed arms remain healthy in their journey to the major leagues?
Thanks to the indispensable college statistics site Boyd’s World and injury data provided by BP’s own Corey Dawkins, we can attempt to answer those questions. To measure pitcher usage in college, we will use Pitcher Abuse Points (PAP), developed in the late-’90s by Rany Jazayerli and later revised into PAP^3 (the version used at Boyd’s World and in this article). PAP is based on the idea that every pitch thrown over the 100-pitch threshold increases potential damage to a pitcher’s arm, and that extremely high pitch count outings, presumably when a pitcher is tiring, are of the most dangerous variety.
In this article we’ll look at the top 15 starting pitchers taken in each draft from 2003 through 2009, and then compare their workload in college to their injury track record in professional baseball (both the minor and major leagues). A more industrious study might look at all starters taken in each draft over a longer time period, but we’re limited by how much PAP data we have for college pitchers (Boyd’s World has data from 2002-2010) and accurate injury data that can be extremely difficult to come by, especially at the minor league level (thanks again to Corey on this front).
First we’ll split our sample of 104 pitchers into thirds, based on their workload in college (sorted by PAP/start):
The following two graphs show all pitchers in the sample, comparing their PAP/start in college against injuries (both two-and five-year year spans) in pro ball:
It’s clear, from both the table and the two graphs above, that there is little correlation between workload in college (based on PAP/start) and days missed due to injury (for what it’s worth, using total PAP rather than PAP/start yields similar results). In fact, there is actually a very slight negative correlation for PAP/start compared with both two-year injuries (-.064) and five-year injuries (-.093).
What’s perhaps most interesting is that the relationship between the overworked pitchers and the middle tier follows the trend we might expect. The two groups missed almost the same amount of time due to injury in the two-year time period, but the middle tier missed seven fewer days over five professional seasons and significantly less time over their careers. Things get confounding with the lightly worked group, however, as its members miss significantly more days due to injury than both of the other groups, providing us with little comfort that our team’s newly drafted Future Ace who was rarely allowed to crack 100 or 110 pitches in college is any better off than one who was used like Old Hoss Radbourn.
To put their workloads into perspective, the overworked group of pitchers reached at least 121 pitches in 30 percent of their starts, while the lightly worked group did so in just under six percent of their outings. However, as the chart below shows, the lightly worked group sustained five more severe injuries (100 or more days missed from one injury) over their first five pro seasons than the workhorses:
Category V start—133-plus pitches
We’ll discuss these somewhat contradictory findings later, but for now let’s focus on some specific pitchers. The following table lists the top 10 pitchers based on PAP/start along with their injury data:
That kind of profile in college is cringeworthy by today’s standards, and there were some questions both about Verlander’s ability to repeat his delivery and that extremely ambitious workload going into the 2004 draft. Taken second overall by the Detroit Tigers and signed for just over $3 million, Verlander has answered all questions in his journey to becoming one of the most durable, prolific pitchers in the game.
Verlander has thrown at least 220 innings over the past four seasons, leading the majors in PAP over that period. Despite all of those innings and pitches dating back to his college days, the righty has remained remarkably healthy. His injury track record reads more like that of a hardened video gamer (blister, laceration [broken callus], blister) than that of a major-league pitcher. Verlander’s most significant setback came in 2005, when he missed the end of the minor league season (34 days) after being shut down due to shoulder tightness.
As in Verlander’s case, there were plenty of questions surrounding Washington right-hander Tim Lincecum heading into the 2006 draft. The combination of Lincecum’s strange mechanics, slight frame, and substantial college workload seemed like a potential recipe for disaster. Over 1,300 major-league innings later, with two Cy Young awards under his belt, Lincecum has yet to sustain any type of serious arm injury, missing time only from minor complications like blisters, contusions from batted balls, and assorted illnesses.
Boyd’s World provided us with game logs for a number of years, including Lincecum’s final two college seasons, giving us an opportunity to compare his pitch counts from college, the minors, and the majors:
The first group consists of Lincecum’s starts during his final two college seasons. Those starts break down as follows:
*Note that a number of the pitch counts used in this article were estimated, as not all college teams track or list pitch counts.
Lincecum’s highest pitch count game came in a 6-4 complete game victory against Arizona State on May 20th, 2005. He fanned 13 that night, walking seven, while facing 40 batters in a 165-pitch performance. Lincecum was also worked hard down the stretch of his junior season, throwing at least 100 pitches in his final six college starts and reaching the 130 mark three times over that span.
Once they assumed control over his outings, the San Francisco Giants quickly harnessed Lincecum’s pitch counts. He made two starts for Low-A Salem-Keizer, throwing just 59 pitches in four innings of work, and was quickly promoted to High-A San Jose, where he was allowed to reach the mid-80s by the end of the ’06 season. He made five starts at Triple-A Fresno in 2007, averaging 94 pitches per start and reaching 100 twice.
At the major-league level in 2007, Lincecum made 24 starts and averaged 99 pitches. The Giants unleashed him a little more in 2008, as he averaged 108 pitches per start and was allowed to throw more than 130 twice. He threw a season-high 138 pitches in a late-season shutout in San Diego (the scene of his 148 pitch no-hitter this year), a start that was bookended by 127- and 118-pitch outings. Still, a more mature Lincecum has reached 120 pitches in only 21 (10 percent) of his major league starts compared to 13 times (39 percent) over his final two seasons at Washington.
The following table lists the bottom 10 pitchers based on PAP/start, with accompanying injury data:
Andrew Brackman was used cautiously in his junior campaign at NC State, but he was ultimately forced to miss the 2007 ACC and NCAA tournament due to lingering elbow issues. In the 12 starts for which we have data, Brackman reached 100 pitches in just 50 percent of them and reached 120 pitches only once.
Brackman was selected by the New York Yankees at the end of the first round in 2007, falling largely due to concerns over the elbow. Shortly after signing with the Yankees, Brackman was sent to visit Dr. James Andrews. Tommy John surgery was scheduled for August, forcing Brackman to miss the entire 2008 season.
Brackman was obviously damaged goods at the end of his college career, nursed along because the arm simply couldn’t handle a Verlander or Lincecumesque type of workload. The Tommy John surgery wasn’t directly related to the low pitch counts in college; it was inevitable either way. More on this later.
Over his final two seasons at Long Beach State, Vance Worley reached the 100-pitch plateau in just nine of the 19 starts we have data for, throwing 120 or more only once. Worley has largely remained healthy in professional ball, missing only limited time due to bone chips in his elbow. He was used responsibly in college and developed into an effective major league starter, avoiding any serious injury setbacks—a feather in the caps of college pitch count hounds.
As is often the case, research into a somewhat foreign subject often reveals more questions than answers. While it’s clear that pitch counts in college may not have as large of an impact on future health as we might expect, it’s difficult to make too many sweeping assertions based on the findings of this article.
First, we’re dealing with a limited sample of 104 pitchers taken in the first few rounds of mid-2000s drafts. While it’s certainly possible to expand and improve a study on this subject, as mentioned previously, we are currently limited by the amount of pitch count data we have as well as injury data for minor leaguers. Perhaps a larger scale study incorporating more pitchers over more years would produce different results.
There are other things that may help explain the confounding results. The overworked group of pitchers were taken 31st overall on average and signed for $1.6 million. The lightly worked group were taken 41st overall and signed for $1.3 million. On average, the pitchers who were allowed to throw more pitches in college were better pitchers, perhaps more mechanically sound with better pitchers’ builds and thus less likely to break down due to injuries.
On a similar note, and probably more important, pitchers who were worked hard in college were probably healthy, while pitchers who were used conservatively were more likely to be dealing with injury issues. The Andrew Brackman case mentioned in the article is a perfect example. College coaches need to be given some credit for working guys who can handle it while being more cautious with pitchers who come with arm trouble or durability concerns.
The next time your team takes a highly touted college starter, remember that his college pitch counts may not tell you as much as you’d think about how much time he’ll be spending on the disabled list. While it’s probably never a plus to see a pitcher used like Verlander or Lincecum were in college, despite their subsequent careers, a heavy college workload doesn’t guarantee an injury-riddled professional career. On the other hand, while a bunch of sub-100 pitch counts may look preferable on the surface, it can also suggest that the pitcher is already dealing with injury or durability issues.
It wouldn’t be surprising if major-league teams already had a good handle on predicting injuries for amateur pitching prospects. They have a lot at stake, as signing bonuses have continued to rise throughout the years, despite the efforts of the new CBA. Hopefully we can make strides to get a better understanding of this subject ourselves; after all, we have a lot at stake too. You can’t be a major player in your 20-team dynasty league with half of your minor-league pitching staff on the DL.
Thanks to Boyd’s World and Corey Dawkins for pitch count and injury data. This article would not have been possible without their help. Also thanks to Baseball-Reference and Baseball America for stats and draft data.