By Chris Cwik
Tommy John surgery is a pitcher's most dreaded nightmare. The procedure is performed when a player incurs damage to the ulnar collateral ligament (UCL) in his elbow, and it can sideline a pitcher for at least a year, taking prime seasons away from some of the game's best players. While medical advancements have increased the surgery's success rate, it remains one of the most devastating scenarios a pitcher can face.
Pitching injuries can strike without warning; all it takes is one pitch for a pitcher to tear his UCL completely. With the introduction of baseball database websites and the advent of PITCHf/x, there is more information about each pitcher's individual performance than ever. But while one might assume those advancements would lead to a better understanding of the injury, it's still incredibly tough to predict which pitcher's UCL will require surgical reconstruction.
From 2007 to 2013, at least 122 pitchers underwent Tommy John surgery, according to BaseballHeatMaps.com, and that number has increased in recent years. Nearly one-third of the pitchers who opened the 2013 season in the majors have had the surgery, according to a recent study by Will Carroll, with Stephen Strasburg, Brian Wilson and Brandon Beachy among the most recent. But that list of 122 names should not be considered complete; finding accurate disabled list data can be difficult. MLB.com includes transaction pages, which accurately detail each move made by a club -- but going back to any year prior to 2010 produces spotty results, with the specifics of an injury often missing. Incomplete data for offseason injuries also limited Carroll's study, since we don't have reliable information about a pitcher's most recent game in those cases.
In the end, there were 78 pitchers since 2007 who qualified. The year 2007 is important, because that also marks the introduction of PITCHf/x, and while it was not immediately available in every park, it has since been installed in every major-league stadium and even a few spring training complexes. Like most new technologies, PITCHf/x had some bugs early in the process and needed some fine-tuning, so some of the early data is not the most trustworthy. For instance, pitch movement numbers were altered quite a bit due to some early calibration issues.
The recent study's findings are not all that surprising. Most of the injuries that led to Tommy John surgery were properly qualified as a variation of an elbow injury. The next most prominent affliction was a forearm strain, or a variety of strain. (This is slightly misleading, though, as some teams used the term "forearm" to refer to the elbow.) The two other types of injuries that appeared were either "UCL tear" or "Tommy John," cases in which teams identified the arm injury immediately. The diagnosis is often changed to "Tommy John surgery" once swelling clears and more accurate tests can be performed.
This study looked at four areas of a pitcher's numbers to determine whether Tommy John surgery was predictable: performance, walks, velocity and final pitch. "Performance" looks at the pitcher's last game prior to undergoing surgery, with the classification based mostly on whether his stat line was below expectations for that pitcher that season. "Walks" is self-explanatory; did the pitcher experience a notable loss in control or command during their final start?
PITCHf/x helped with the final two categories. "Velocity" indicates whether a pitchers showed notable drop-offs in velocity immediately before an injury. "Final pitch" attempts to give an idea of whether breaking pitches, which are said to be more taxing on the elbow, contribute to serious UCL damage.
Performance was the area that showed the greatest correlation. A little more than 60 percent of pitchers turned in a bad performance, immediately before undergoing Tommy John surgery. This makes sense; players are rarely placed on the disabled list following a great outing. That did happen with about 25 percent of the group, though -- they either left the game during a great start, or showed no signs of decline prior to requiring Tommy John surgery. The rest of the sample, about 15 percent, performed as if it were a normal appearance.
Though the majority of players who experienced Tommy John surgery performed poorly in their last appearance, only 11.5 percent of pitches saw an uptick in their walk rates in that final game. This may be surprising, as lack of control or command is often associated with an injury. In this sample, it wasn't much of a predictor of Tommy John surgery.
The same can be said about velocity. Nearly 41.5 percent of pitchers experienced a significant velocity loss in their final appearance before undergoing surgery, but 49 percent remained stable. Five pitchers even saw their average velocity rise in their last games, and there were a few instances when a pitcher had been working with reduced velocity but threw a high-velocity fastball on his final pitch. This didn't happen enough to be relevant to the study, but it shows how little velocity readings actually matter in predicting Tommy John surgery.
A large majority of pitchers threw a variation of a fastball with their final pitch. Fifty-eight percent of pitchers finished their last game with either a four-seam fastball, a sinker or a cutter. Thirty-one percent of pitchers left after throwing a breaking ball, either a curve or slider. The other 11 percent threw change-ups and splitters in their last pitches. Earlier research has found that there could be a relationship between throwing breaking balls and having elbow issues, but those won't necessarily be the exact pitches on which a player's elbow finally gets shredded.
While the data provided is useful, in the end it doesn't shed much light on Tommy John surgery. Aside from a pitcher experiencing a bad performance, which is expected of any pitcher placed on the disabled list, the injury remains devastatingly unpredictable based on the stats.
As more accurate data becomes available and more advances are made in this area, there may eventually be a way to determine whether certain pitchers are bigger risks for the surgery. MLB's disabled list data will continue to grow, giving access to more accurate informaotin about injuries. A big change must also happen on the minor-league side, where it's now impossible to track the countless number of minor-league pitchers who undergo Tommy John surgery every year. There's also the issue of overuse, which Carroll mentions in his article. But, he notes, most overuse happens before pitchers even reach the professional level, making it tough to track. Overuse is rampant in college baseball, where there are no limits on pitch counts or mandatory rest periods for pitchers.
Most of these changes will come with time. As more seasons pass, more data will accumulate. Enhancements likely will be made to the PITCHf/x software, making it even more accurate than before. For the moment, though, there is still be no way to predict Tommy John surgery.
* * *
Chris Cwik writes for various baseball sites on the internet, including CBSSports.com and FanGraphs.com. He has also contributed to ESPN and the Hardball Times Baseball Annual. Follow him on Twitter at @Chris_Cwik.