By Frank Zhu
When MRI scans in September showed Angels two-way star Shohei Ohtani would need Tommy John surgery, it was the final blow to an Angels team that had seen injuries derail their hot start which saw them leading the AL West 40 games into the 2018 season. After a near-comical amount of injuries which saw 4 of 6 starters injured, including losing starting pitcher JC Ramirez for the season with a UCL tear, and 15 players on the disabled list, the Angels sputtered to 4th place in the AL West, 17 games behind the Oakland Athletics for the 2nd Wild Card spot.
Immediately after the season, however, the focus shifted to Ohtani, and how Tommy John surgery would affect his future performance. For those unfamiliar with Tommy John surgery, it is a procedure to repair the Ulnar Collateral Ligament (UCL), which is located on the inside of the elbow. The action of overhead throwing places stress on the UCL, which over time can tear.
Before 1974, when the procedure was first done, a UCL tear ended a pitcher’s career. Now, Tommy John surgery allows pitchers to return to the mound after a year of rehabilitation, which can prolong careers. However, Tommy John surgery is also viewed as harming a player’s future value, both in terms of trades and free agency contracts. Discerning the impact of surgery on pitcher performance is especially relevant today, as an increase in pitch velocity has led to more UCL injuries– 25% of active MLB pitchers have undergone Tommy John surgery.
As a result of these concerns, I wanted to test if Tommy John surgery leads to a decline in player performance. To answer this question, I used data from MLBPlayerAnalysis for a list of all Tommy John surgeries coupled with pitching statistics from Lahman’s Baseball Database. For each of the 423 pitchers who underwent Tommy John surgery, I took their average ERA from the 3 seasons preceding the surgery and compared it to their average ERA from the 3 seasons after surgery, excluding seasons where a pitcher faced less than 100 batters, then ran a t-test on the data (p <.05). The t-test tells if there is a statistically significant difference in the mean ERA between two populations, in this case pitchers before and after surgery. The null hypothesis was a pitcher’s average ERA stayed constant before and after surgery.
After analyzing the data, the average ERA of pitchers after Tommy John increased from 3.98 to 4.41, an average increase of 0.43. The p-value for the t-test was .0087, which is less than .05, indicating there is enough evidence to reject the null hypothesis. Therefore, the change in pitchers’ ERA after Tommy John surgery is statistically significant, and pitchers’ performances are negatively affected by Tommy John surgery.
As shown in the chart, the change in pitchers’ ERA is centered at around a 30% increase in ERA (Positive % changes in ERA are worse), with several data points skewing the data right. This indicates that pitchers’ performances, on average, tend to be negatively affected in seasons after surgery. Furthermore, to attempt to account for the right skew of the data, I calculated the median percent ERA change of the data, which was a 7.7% increase. The mean, median, and t-test all show pitcher performance decreases after Tommy John surgery.
Based on this analysis, one factor I thought which could be related to performance decline was the number of batters faced. If pitchers’ performance after Tommy John surgery declined, their injury could affect their durability as well. To examine this, I used the number of batters faced in a season before and after a pitcher’s surgery.
When comparing the average number of batters a pitcher faced (Minimum of 100 batters faced in a season) before and after surgery, the average number of batters decreased from 443 batters a season before surgery to 320 batters after surgery, a decrease of 123 batters, or a 27.8% decrease. This change was statically significant with a p-value of 9.43e-06. Furthermore, the median percent decrease in batters faced after surgery was 29.4%.
As seen in the graph of change of average batters faced in a season (Positive % indicates more batters faced after surgery), the data is again skewed right, indicating some pitchers pitch many more innings in the seasons after Tommy John surgery. However, on average, pitchers are less durable after surgery.
Based on the analysis, it seems that teams are right to shy away from pitchers with a history of Tommy John surgery – while surgery may prolong a pitcher’s career, a pitcher’s utility drastically decreases – their performance decreases, shown by a 30% uptick in ERA, and their durability throughout the season also decreases, with the average pitcher facing 27% fewer batters. One hypothesis for this could be because pitchers who were starters before surgery become relievers after surgery.
This leads to a few interesting points. First, this study evaluates a pitcher’s performance immediately after returning from surgery, not the longevity of their careers after surgery. Second, it would be fascinating to see if a player who has Tommy John surgery is more vulnerable to further injury. Third, further articles could evaluate a pitcher’s performances using more advanced metrics, such as percentage of pitches thrown for strikes or average pitch velocity. Pitch velocity could be especially interesting, as pitchers could be throwing slower to “protect” their arms, leading to increased durability but decreased performance.
However, one potential flaw with the data is that while recuperating from surgery, pitchers age, so part of this decline in performance could be due to a decline in simply athletic ability, as pitchers tend to decline as they get older. In this study, the age of surgery ranged from 23 years old to 42 years old(!), with an average age of 29.2 years.
If you have any questions for Frank about this article, please feel free to reach out to him at frank_zhu@college.harvard.edu