Orthopedics

Feature Article Supplemental Data

Advanced Baseball Metrics Indicate Significant Decline in MLB Pitcher Value After Tommy John Surgery

Ryan S. Selley, MD; Daniel A. Portney, MD; Cort D. Lawton, MD; Michael D. Shockley, MD; Robert A. Christian, MD; Matthew D. Saltzman, MD; Wellington K. Hsu, MD

Abstract

There are conflicting reports regarding the efficacy of Major League Baseball (MLB) pitchers following medial ulnar collateral ligament reconstruction (UCLR). As advanced baseball metrics have revolutionized the way general managers define pitchers' value, the authors believe that these should be used to measure clinical outcomes after UCLR. All MLB pitchers who underwent UCLR from April 1, 1991, through July 1, 2016, were compiled (N=253). Pitcher demographics and statistics were collected for up to 3 full seasons preceding and following surgery. Data for pitching controls (all MLB [AMLB] pitchers) were also collected (N=14,756). Prior to surgery, pitchers with UCLR were significantly better than the AMLB pitchers in nearly all advanced value-based statistics: higher wins above replacement (WAR; 0.93 vs 0.78; 95% confidence interval [CI], 0.80–1.06), lower fielding independent pitching (FIP; 4.23 vs 4.51; 95% CI, 4.12–4.34), lower expected fielding independent pitching (xFIP; 4.17 vs 4.38; 95% CI, 4.06–4.28), higher win probability added (WPA; 0.21 vs 0.05; 95% CI, 0.1–0.32), and higher leverage index (pLI; 1.03 vs 0.96; 95% CI, 0.99–1.06). Pitchers who returned to play after UCLR demonstrated significantly lower value with worse WAR, FIP, WPA, and pLI (P<.05). Conversely, after excluding pitchers who failed to achieve a sustained return to play of greater than 1 year, there were no significant decreases in value after surgery. The authors conclude that, prior to injury, pitchers who have UCLR are more valuable than average MLB pitchers. However, UCLR pitchers perform worse when compared with their presurgical values. This may be affected by players not being able to continue their careers for more than 1 year. [Orthopedics. 2019; 42(6):349–354.]

Abstract

There are conflicting reports regarding the efficacy of Major League Baseball (MLB) pitchers following medial ulnar collateral ligament reconstruction (UCLR). As advanced baseball metrics have revolutionized the way general managers define pitchers' value, the authors believe that these should be used to measure clinical outcomes after UCLR. All MLB pitchers who underwent UCLR from April 1, 1991, through July 1, 2016, were compiled (N=253). Pitcher demographics and statistics were collected for up to 3 full seasons preceding and following surgery. Data for pitching controls (all MLB [AMLB] pitchers) were also collected (N=14,756). Prior to surgery, pitchers with UCLR were significantly better than the AMLB pitchers in nearly all advanced value-based statistics: higher wins above replacement (WAR; 0.93 vs 0.78; 95% confidence interval [CI], 0.80–1.06), lower fielding independent pitching (FIP; 4.23 vs 4.51; 95% CI, 4.12–4.34), lower expected fielding independent pitching (xFIP; 4.17 vs 4.38; 95% CI, 4.06–4.28), higher win probability added (WPA; 0.21 vs 0.05; 95% CI, 0.1–0.32), and higher leverage index (pLI; 1.03 vs 0.96; 95% CI, 0.99–1.06). Pitchers who returned to play after UCLR demonstrated significantly lower value with worse WAR, FIP, WPA, and pLI (P<.05). Conversely, after excluding pitchers who failed to achieve a sustained return to play of greater than 1 year, there were no significant decreases in value after surgery. The authors conclude that, prior to injury, pitchers who have UCLR are more valuable than average MLB pitchers. However, UCLR pitchers perform worse when compared with their presurgical values. This may be affected by players not being able to continue their careers for more than 1 year. [Orthopedics. 2019; 42(6):349–354.]

Ulnar collateral ligament reconstruction (UCLR) is one of the most common surgical procedures performed among professional baseball pitchers. Much of the reported success after treatment has been based on a high return to play rate (94%).1 Pitching performance often has been touted as better on return from surgery based on basic statistics such as earned run average (ERA) and walks plus hits per inning pitched (WHIP).2 Other studies have reported decreases in pitching velocity after UCLR, but this has not been significantly lower when compared with age-matched controls.3 These data have led to media reports concluding that UCLR pitchers have improved performance after surgery when compared with preoperatively.4

Conversely, other authors have reported significantly worse results, with only 67% of pitchers returning to the same level of competition postoperatively and significant declines in performance metrics.5 Lansdown and Feeley6 concluded that pitchers demonstrate a significant decrease in mean velocity, fastball velocity, and fastball percentage after reconstruction. Notably, pitchers who have ulnar collateral ligament (UCL) injury demonstrate higher-velocity fastballs than controls, which has been identified as a risk factor for UCL injury.7 Finally, Keller et al8 reported that pitchers requiring UCLR pitch a higher percentage of fastballs than controls (46.7% vs 39.4%).

Advanced statistical-based measures have been recently used to more accurately quantify a pitcher's value, such as the leverage index (pLI). This metric quantitatively measures the critical nature of a moment in a game. For example, a high pLI would be characterized by the scenario with a tie game, bases loaded, and 2 outs in the 9th inning.9 Pitchers with a pLI above 1 are used in decisive scenarios such as this that require more dependability, reliability, and consistency, whereas those with a value less than 1 are used less in these situations. These values can help to provide a more accurate qualitative value of pitchers using quantitative measurements. Whereas initial literature focused on traditional statistics to assess the performance of pitchers with UCLR, Gray and Cook10 advocated the need for advanced metric analysis as it pertains to performance assessment in the medical literature. The inherent flaw of such traditional statistics (ERA, batting average against, hits, runs given up) is that other variables, such as run support and strength of the opposing team, unpredictably influence these measures. The use of these basic metrics as outcome measures to ascertain the success of a player is inadequate. This study sought to use advanced baseball metrics to define the value of pitchers both before and after UCLR.

Materials and Methods

This study was exempt from institutional review board approval because all information was obtained from publicly available resources.

Preoperative/Risk Analysis

A total of 358 Major League Baseball (MLB) pitchers were identified who had undergone UCLR between April 1, 1991, and July 1, 2016, via a publicly available database11 and were verified via public media records through a previously established protocol.12,13 Pitchers who underwent revision procedures (n=43) were excluded. Additionally, only seasons in which pitchers accumulated 6 or more innings pitched (if before 2002) or 100 pitches (if after 2002, when pitch count data were made available) were included. Of the 292 remaining pitchers, pitching statistics were collected for up to 3 full seasons preceding and following surgery (Figure 1).

Study flow indicating players and years collected and exclusion and inclusion criteria. Abbreviations: MLB, Major League Baseball; UCL, ulnar collateral ligament; UCLR, ulnar collateral ligament reconstruction.

Figure 1:

Study flow indicating players and years collected and exclusion and inclusion criteria. Abbreviations: MLB, Major League Baseball; UCL, ulnar collateral ligament; UCLR, ulnar collateral ligament reconstruction.

Each subject's demographic data, including age, height, weight, throwing arm, MLB debut date, country of origin, high school state, and college state, were collected from publicly available records. Furthermore, surgery-specific data, including surgery and return date, were collected. The index season was identified as the season during which each player had surgery. If the surgery occurred during the offseason, the index year was defined as the following season to include data from the previously completed season.

Statistics of included pitchers were compiled and categorized: basic pitching (wins, losses, innings pitched, pitches, ERA, WHIP, hits per 9 innings [H/9], home runs per 9 innings [HR/9], average), advanced value-based pitching (fielding independent pitching [FIP], expected fielding independent pitching [xFIP*], tERA*, wins above replacement [WAR], win probability added [WPA], pLI, pace**), and pitching power statistics (strikes per 9 innings [K/9], bases on balls per 9 innings [BB/9], fastball%*, fastball velocity*, contact%*, strike zone%*). Statistics followed by * and ** were collected beginning in the 2002 and 2007 seasons, respectively.

Specific advanced pitching statistics included WAR, FIP, xFIP, WPA, and pLI. Wins above replacement estimates the number of wins a player has been worth to his team compared with a standard replacement pitcher, FIP is an estimate of a pitcher's ERA while controlling for factors such as defense or pure chance, xFIP adds values that better predict future performance, WPA captures a pitcher's contribution to the team's win expectancy, and pLI is a measure of critical game scenarios.14

A control cohort was created with all MLB pitchers from the 1988 to 2015 seasons with the same exclusion criteria (N=14,756) (AMLB). Ninety-five percent confidence intervals were generated for the UCLR pre- and post-UCLR averages and compared with the overall MLB averages. All variables were analyzed and clustered based on correlations among variables. One representative variable for each of the clusters was selected based on its distance to intra- and inter-cluster variables. These selected variables were further evaluated based on Spearman's and Hoeffding's D coefficiency with injury. Those ranked lowest in both of these coefficients were excluded. The remaining variables were subsequently evaluated as predictors of injury using logistic regression models. Predictive variables with statistically significant effects on injury (P<.05) were kept in the model using forward selection methods, and standardized parameter estimates were generated. Analyses were performed using R 3.3.1 software (R Foundation for Statistical Computing, Vienna, Austria).

Pre- vs Postsurgical Analysis

Of the 208 players who returned to MLB following UCLR, 192 had 100 or more pitches or 6 or more innings pitched in at least 1 of 3 years prior to surgery and after return. All statistics for these players were compared. Each player served as his own control for preand post-UCLR performance. A 2-tailed paired t test was used to evaluate for significant differences. Significance was accepted with P<.05. Further, players were again analyzed by excluding those who failed to pitch more than 1 year after injury (n=44) to assess if longevity had an effect on value. The remaining pitchers who were able to pitch into years 2 and 3 after UCLR were analyzed via the same methods (Figure 1).

Return to Play Analysis

Of the 292 pitchers with primary UCLR, 253 had surgery prior to June 2014 and were included for analysis. This cutoff date was chosen to allow adequate time to return from surgery, as data collection was stopped after July 2016. To analyze only pitchers who achieved a meaningful return as defined by the authors' inclusion criteria, pitchers with 100 or more pitches (2002 to present) or 6 or more innings in a season (before 2002) were included. On calculation of sustained return to play in years 2 and 3 after surgery, the cohort of pitchers was reduced to those with surgery prior to May 2013 and May 2012, respectively (Figure 2).

Study flow demonstrating return to play calculation by year after surgery. Qualifying return to play was defined as more than 6 innings pitched before 2002 or more than 100 pitches thrown after 2002 in a season. Abbreviations: MLB, Major League Baseball; UCL, ulnar collateral ligament; UCLR, ulnar collateral ligament reconstruction.

Figure 2:

Study flow demonstrating return to play calculation by year after surgery. Qualifying return to play was defined as more than 6 innings pitched before 2002 or more than 100 pitches thrown after 2002 in a season. Abbreviations: MLB, Major League Baseball; UCL, ulnar collateral ligament; UCLR, ulnar collateral ligament reconstruction.

Results

A total of 292 pitchers undergoing primary UCLR met inclusion criteria. The mean age at MLB debut was 23.9±2.1 years, and pitchers had surgery at the mean age of 28.6±3.9 years (Table 1).15 Players who required UCLR performed at a significantly higher level than controls prior to surgery. They threw significantly faster velocity (91.8 mph vs 90.7 mph; 95% confidence interval, 91.39–92.13) compared with AMLB pitchers. Advanced value-based statistics also demonstrated significantly greater value for the UCL compared with the average pitcher, with significantly higher WAR, lower FIP, lower xFIP, higher WPA, and higher pLI (Table A, available in the online version of the article). Number of pitches (1249±857), strike percentage (K%; 19%±0.06), and ERA (4.24±1.3) were statistically significant positive predictors of UCL injury (standardized parameter estimates: 0.1283, 0.108, and −0.1793, respectively).

Demographic Data for MLB Pitchers With UCLR

Table 1:

Demographic Data for MLB Pitchers With UCLR

Pre- and Post-Surgery Descriptive Statistics and Comparison to MLB average

Table A.

Pre- and Post-Surgery Descriptive Statistics and Comparison to MLB average

Pitchers who underwent UCLR who returned to play and had sufficient data for paired comparison (n=192) demonstrated significantly decreased statistical performance in numerous metrics, with significantly decreased innings and total pitches per year (P<.05). Pitchers gave up significantly more hits (P<.05) and home runs (P<.05) per 9 innings following return to play. Earned run average was also significantly worse following return to play (P<.05). Additionally, pitchers threw slower fastballs (91.8 mph vs 90.9 mph; P<.05), had a lower fastball percentage (63.9% vs 60.3%; P<.05), and were less likely to throw the ball in the strike zone (P<.05) after surgery (Table B, available in the online version of the article).

Paired Outcomes Data

Table B:

Paired Outcomes Data

Advanced statistics help researchers precisely characterize pitcher performance. Most notably following UCLR, pitchers demonstrated a decline in value in numerous dimensions. Fielding independent pitching was significantly worse following return (P<.05), along with WAR (P<.05), WPA (P<.05), and pLI (P<.05). However, these findings were affected by a cohort of players who failed to achieve a sustained return to play of more than 1 year (n=44). When analyzing only pitchers who were able to achieve a sustained return to play (n=148), there were no longer significant differences in pre- vs post-UCLR values, which indicates that longevity after surgery can affect these parameters (Table C, available in the online version of the article).

Post-UCL, Years 2 and 3 Paired Outcomes Data

Table C:

Post-UCL, Years 2 and 3 Paired Outcomes Data

Discussion

Ulnar collateral ligament reconstruction has been extensively studied. Different performance measures being used to assess outcome has resulted in controversy regarding efficacy after return to play.2,5,10 Studies that rely on rudimentary statistics dependent on subjective scorers' determination of errors and team defense may not accurately assess pitcher value without contribution from the surrounding environment. There are a limited number of studies using advanced pitching metrics to assess UCLR outcomes in MLB pitchers.6,16,17 Advanced statistical measures demonstrate important distinctions that cannot be determined from basic metrics.

The data of the current study evaluating advanced metrics for UCLR pitchers suggest that value is affected, at least in part, by surgical reconstruction. First, FIP was significantly worse after surgery but not significantly worse than the AMLB pitcher.6 Furthermore, WPA, which is a cumulative statistic that measures how often a player's actions have increased a team's odds of winning, is significantly decreased after surgery. Evaluation of pLI, which is a measure of the critical nature of a particular moment in the game,9 demonstrates a significantly higher score than the AMLB pitcher prior to surgery but a significantly decreased score compared with presurgical levels following UCLR.

Outcomes following UCLR have varied in the literature. Erickson et al2 reported a return to play rate, defined as 1 game in the MLB, for UCLR pitchers of 83%. Moreover, they found a 97% return to play rate in all professional leagues, with only 2.8% unable to return to high-level play. Makhni et al5 found less convincing results. In their series, only 67% of established pitchers returned to the same level of competition postoperatively and 57% of established pitchers returned to the disabled list due to injuries in the throwing arm. Furthermore, they found significant declines in basic performance metrics after surgery compared with preinjury levels, although these were not statistically different from age-related declines in age-matched controls.5 The recent availability of pitching data has allowed researchers to analyze pitch types, pitch velocity, and pitching-specific outcome metrics. Multiple authors have found a significant decrease in mean velocity, fastball velocity, and fastball percentage after reconstruction.3,6 This further challenges the notion that pitchers return to their preinjury state following UCLR.

On the basis of the evaluation of all advanced value-based metrics as a whole, the current authors conclude that pitchers who sustain a UCL injury are more valuable than the average MLB pitcher, with significantly better WAR, FIP, xFIP, WPA, and pLI (Table C). Furthermore, these statistics demonstrate significant decline in value after surgical reconstruction of UCL when considering the entire cohort of pitchers. Notably, when pitchers who retire within 1 year after injury are excluded, this effect is no longer evident. This could indicate that the subset of pitchers who were unable to achieve a sustained return represent a failed reconstruction or those who chose UCLR in anticipation of finishing their career. In light of these findings, the current authors suggest evaluating UCLR pitchers in separate cohorts. Of the 292 original pitchers included for preoperative analysis, 253 pitchers had time to return to play. Of this cohort, 75.1% (190 of 253) achieved a qualifying return to play. When excluding pitchers who fail to return more than 1 year, the qualifying return to play decreases to 61% (144 of 236). This suggests that the failure rate could be as high as 39% as defined by length of career after surgery. This finding may indicate that the first year after UCLR may be a critical time to evaluate how a pitcher responds to the differences in elbow mechanics after surgery. If a pitcher responds with appropriate performance and endurance after 1 year in MLB, then his value may not be significantly different from presurgical measures. This study was not able to determine the effects of strict postsurgical pitch counts on this phenomenon, although the authors are actively investigating this within their research group.

Several studies have analyzed specific player factors that might place players at risk for UCL injury.7,8,18 A more recent study by Whiteside et al18 found that fewer days between games, smaller repertoire of pitches, less pronounced horizontal release location, smaller stature, greater mean pitch speed, and greater mean pitch counts per game were significant predictors for UCLR. Similarly, the current authors' data indicate that UCLR pitchers throw more and faster fastballs, more pitches, and more innings overall prior to injury than the average MLB pitcher. The current authors found ERA, total pitches thrown, and K% to be most predictive of UCL injury.

This study had multiple limitations. It relied only on publicly available information and reporting of injuries, which is limited by the accuracy of the reporting. Furthermore, postoperative performance may be affected by surgeon skill, reconstruction method, other associated elbow/shoulder pathology, kinetic chain issues, and rehabilitation philosophy. This study was also a retrospective case series that compared pitchers with UCLR with the AMLB as well as with their preoperative performance. The results could have been affected by age-related decline being attributed to a decline in value, although the comparison with the AMLB control group should have offset this possibility. Additionally, this decline was no longer observed when excluding pitchers who failed to achieve a sustained return of more than 1 year, suggesting an alternate explanation for the decline in value. Similarly, variations in the amount of data available both pre- and postoperatively for an individual pitcher make it difficult to place players into discrete categories for both return to play and value analysis. As is the case with previous publications from the authors' research group, reporting errors that may exist were hopefully mitigated with the sample size of this study. Finally, there are numerous external factors that can affect an individual player's performance and return to play, including defensive team support, salary caps, and other unrelated injuries. Despite these limitations, the authors believe that the use of value-based statistics is helpful in the analysis of MLB pitchers following UCLR.

Conclusion

To the authors' knowledge, this is the largest and most specific value-based analysis of MLB UCLR pitchers. Using pertinent and more appropriate value-based pitching statistics, the authors concluded that pitchers who sustain a UCL injury are more valuable and perform at a higher level than the average MLB pitcher. Notably, when considering the entire cohort of pitchers, they are significantly less valuable after UCLR than prior to surgery. However, this effect was eliminated when analyzing players who were able to maintain a sustained return of more than 1 year. Only 61% of pitchers were able to maintain a sustained return of more than 1 year. These data suggest that when assessing performance, players should not be analyzed in one cohort but instead delineated based on the length of their career after surgery. The authors believe that these findings should be used to guide health practitioners in the preoperative counseling of players who require UCLR.

References

  1. Koh JL, Schafer MF, Keuter G, Hsu JE. Ulnar collateral ligament reconstruction in elite throwing athletes. Arthroscopy. 2006;22(11):1187–1191. doi:10.1016/j.arthro.2006.07.024 [CrossRef]17084295
  2. Erickson BJ, Gupta AK, Harris JD, et al. Rate of return to pitching and performance after Tommy John surgery in Major League Baseball pitchers. Am J Sports Med. 2014;42(3):536–543. doi:10.1177/0363546513510890 [CrossRef]
  3. Jiang JJ, Leland JM. Analysis of pitching velocity in Major League Baseball players before and after ulnar collateral ligament reconstruction. Am J Sports Med. 2014;42(4):880–885. doi:10.1177/0363546513519072 [CrossRef]24496506
  4. Conte SA, Hodgins JL, ElAttrache NS, Patterson-Flynn N, Ahmad CS. Media perceptions of Tommy John surgery. Phys Sportsmed. 2015;43(4):375–380. doi:10.1080/00913847.2015.1077098 [CrossRef]26307904
  5. Makhni EC, Lee RW, Morrow ZS, Gualtieri AP, Gorroochurn P, Ahmad CS. Performance, return to competition, and reinjury after Tommy John surgery in Major League Baseball pitchers: a review of 147 cases. Am J Sports Med. 2014;42(6):1323–1332. doi:10.1177/0363546514528864 [CrossRef]24705898
  6. Lansdown DA, Feeley BT. The effect of ulnar collateral ligament reconstruction on pitch velocity in Major League Baseball pitchers. Orthop J Sports Med. 2014;2(2):2325967114522592. doi:10.1177/2325967114522592 [CrossRef]26535301
  7. DeFroda SF, Kriz PK, Hall AM, Zurakowski D, Fadale PD. Risk stratification for ulnar collateral ligament injury in Major League Baseball players: a retrospective study from 2007 to 2014. Orthop J Sports Med. 2016;4(2):2325967115627126. doi:10.1177/2325967115627126 [CrossRef]26848482
  8. Keller RA, Marshall NE, Guest JM, Okoroha KR, Jung EK, Moutzouros V. Major League Baseball pitch velocity and pitch type associated with risk of ulnar collateral ligament injury. J Shoulder Elbow Surg. 2016;25(4):671–675. doi:10.1016/j.jse.2015.12.027 [CrossRef]26995458
  9. Slowinski S. LI. http://www.fangraphs.com/library/misc/li. Accessed October 1, 2016.
  10. Gray AD, Cook JL. Return to pitching after Tommy John surgery: letter to the editor. Am J Sports Med. 2014;42(12):NP54. doi:10.1177/0363546514559883 [CrossRef]25428958
  11. Roegele J. Tommy John surgery list. https://docs.google.com/spreadsheets/d/1gQujXQQGOVNaiuwSN680Hq-FDVsCwvN-3AazykOBON0/edit#gid=0. Accessed October 1, 2016.
  12. Minhas SV, Kester BS, Larkin KE, Hsu WK. The effect of an orthopaedic surgical procedure in the National Basketball Association. Am J Sports Med. 2016;44(4):1056–1061. doi:10.1177/0363546515623028 [CrossRef]26801923
  13. Mai HT, Alvarez AP, Freshman RD, et al. The NFL Orthopaedic Surgery Outcomes Database (NO-SOD): the effect of common orthopaedic procedures on football careers. Am J Sports Med. 2016;44(9):2255–2262. doi:10.1177/0363546516651426 [CrossRef]27311414
  14. Baseball statistics and analysis. http://www.fangraphs.com. Accessed October 1, 2016.
  15. Erickson BJ, Harris JD, Tetreault M, Bush-Joseph C, Cohen M, Romeo AA. Is Tommy John surgery performed more frequently in Major League Baseball pitchers from warm weather areas?Orthop J Sports Med. 2014;2(10):2325967114553916. doi:10.1177/2325967114553916 [CrossRef]
  16. Liu JN, Garcia GH, Conte S, ElAttrache N, Altchek DW, Dines JS. Outcomes in revision Tommy John surgery in Major League Baseball pitchers. J Shoulder Elbow Surg. 2016;25(1):90–97. doi:10.1016/j.jse.2015.08.040 [CrossRef]
  17. Marshall NE, Keller RA, Lynch JR, Bey MJ, Moutzouros V. Pitching performance and longevity after revision ulnar collateral ligament reconstruction in Major League Baseball pitchers. Am J Sports Med. 2015;43(5):1051–1056. doi:10.1177/0363546515579636 [CrossRef]25862037
  18. Whiteside D, Martini DN, Lepley AS, Zernicke RF, Goulet GC. Predictors of ulnar collateral ligament reconstruction in Major League Baseball pitchers. Am J Sports Med. 2016;44(9):2202–2209. doi:10.1177/0363546516643812 [CrossRef]27159303

Demographic Data for MLB Pitchers With UCLR

CharacteristicValue
Age at surgery, mean±SD (range) y28.6±3.9 (21.3–48.0)
Height, mean±SD (range), in74.3±2.1 (67.0–79.0)
Weight, mean±SD (range), lb211±22 (160–280)
Body mass index, mean±SD (range), kg/m226.8±2.5 (20.8–34.4)
Draft round,a mean±SD (range)8.8±10.0 (1–56)
MLB debut age, mean±SD (range),y23.9±2.1 (18.9–38.2)
Throwing arm, No. (% of total) (n=292)
  Right214 (73.3)
  Left78 (26.7)
Country of origin, No. (% of total) (n=292)
  United States220 (75.3)
  Foreign born72 (24.7)
High school state, No. (% of total) (n=220)
  North116 (52.7)
  South104 (47.3)
College state,b No. (% of total) (n=165)
  North84 (50.3)
  South87 (52.7)

Pre- and Post-Surgery Descriptive Statistics and Comparison to MLB average

Pre-Surgery (n=292)Post-Surgery (n=208)MLB Average
Mean ± St. Dev (Min–Max)95% CIMean ± St. Dev (Min–Max)95% CI(n=14756)
Basic Pitching Statistics

  W4.59 ± 3.94 (0–17)[4.14 – 5.04]3.46 ± 3.35 (0–18)[3.01 – 3.92]4.38 ± 4.60
  L4.41 ± 3.3 (0–13)[4.03 – 4.79]3.63 ± 2.82 (0–12)[3.24 – 4.01]4.36 ± 3.94
  IP81.1 ± 56.0 (6.1–234.1)[74.7 – 87.6]62.0 ± 46.3 (5.0–222.1)[55.7 – 68.3]77.7 ± 64.0
  Pitches*1249 ± 857 (115–3348)[1132 – 1366]1031 ± 768 (112–3386)[916 – 1145]1201 ± 976
  ERA4.24 ± 1.3 (0.69–11.37)[4.09 – 4.39]4.81 ± 1.6 (0.–11.32)[4.60 – 5.03]4.67 ± 2.05
  WHIP1.41 ± 0.25 (0.81–2.67)[1.38 – 1.43]1.48 ± 0.28 (0.91–2.6)[1.44 – 1.52]1.46 ± 0.33
  H/98.94 ± 1.67 (4.23–15.28)[8.75 – 9.13]9.46 ± 1.86 (4.77–17.18)[9.21 – 9.71]9.35 ± 2.27
  HR/90.99 ± 0.44 (0.–3.24)[0.94 – 1.04]1.13 ± 0.57 (0.–3.95)[1.05 – 1.21]1.08 ± 0.68
  Avg.253 ± .036 (.135–.385)[0.249 – 0.257].263 ± .038 (.152–0.391)[0.258 – 0.268].261 ± .046

Advanced Pitching Specific Statistics

  FIP4.23 ± 0.96 (0.62–8.38)[4.12 – 4.34]4.58 ± 1.12 (2.22–8.44)[4.43 – 4.73]4.51 ± 1.34
  xFIP*4.17 ± 0.78 (1.41–7.25)[4.06 – 4.28]4.39 ± 0.77 (2.4–7.79)[4.28 – 4.51]4.38 ± 0.92
  tERA*4.52 ± 1.06 (1.59–8.81)[4.37 – 4.66]4.81 ± 1.17 (2.19–9.64)[4.63 – 4.98]4.82 ± 1.48
  WAR0.93 ± 1.13 (−0.5–5.77)[0.80 – 1.06]0.56 ± 1.01 (−0.8–5.17)[0.42 – 0.70]0.78 ± 1.36
  WPA0.21 ± 0.97 (−1.9–3.64)[0.10 – 0.32]−0.03 ± 0.83 (−2.25–3.26)[−0.14 – 0.08]0.05 ± 1.29
  pLI1.03 ± 0.32 (0.13–1.98)[0.99 – 1.06]0.94 ± 0.28 (0.08–2.05)[0.91 – 0.98]0.96 ± 0.36
  Pace**22.2 ± 2.06 (17.7–30.7)[21.86 – 22.53]22.56 ± 2.28 (18.1–29.8)[22.16 – 22.96]22.32 ± 2.38

Pitching Power Statistics and Strikeouts

  K/97.29 ± 1.99 (2.2–12.66)[7.06 – 7.52]7.03 ± 1.99 (1.74–14.22)[6.76 – 7.30]6.68 ± 2.14
  BB/93.72 ± 1.37 (1.04–11.67)[3.56 – 3.87]3.84 ± 1.54 (1.16–11.32)[3.63 – 4.05]3.78 ± 1.65
  FB%*63.4 ± 9.5 (21.4–88.9)[62.1% – 64.7%]60.6 ± 9.9 (31.1–85.0)[59.1% – 62.1%]61.2 ± 11.6
  FB velocity*91.8 ± 2.7 (81.2–99.3)[91.39 – 92.13]91.0 ± 2.9 (78.6–97.7)[90.51 – 91.39]90.7 ± 3.1
  Contact %*78.6 ± 4.7 (63.2–89.4)[78.0% – 79.2%]79.5 ± 4.6 (65.5–88.7)[78.8% – 80.2%]79.8 ± 5.2
  Strike Zone %*48.2 ± 4.8 (34.9–63.5)[47.6% – 48.9%]47.4 ± 4.7 (33.3–60.2)[46.7% – 48.2%]48.1 ± 5.2

Paired Outcomes Data

Pre-Surgery (n=192)Post-Surgery (n=192)p-value
Mean ± St. Dev (Min–Max)95% CIMean ± St. Dev (Min–Max)95% CI
Basic Pitching Statistics

  W5.17 ± 4.16 (0.0–17.0)4.58–5.763.45 ± 3.31 (0.0–17.7)2.98–3.920.12E-07
  L4.94 ± 3.34 (0.0–13.0)4.47–5.413.65 ± 2.79 (0.0–12.0)3.26–4.040.14E-06
  IP90.0 ± 57.8 (6.1–234.1)81.8–98.261.8 ± 45.7 (5.0–222.1)55.3–68.30.15E-10
  Pitches*1393 ± 854 (125–3348)1242–15441026 ± 759 (114–3386)892–11600.51E-07
  ERA4.37 ± 1.39 (0.75–11.37)4.17–4.574.83 ± 1.59 (1.58–11.32)4.61–5.050.64E-03
  WHIP1.42 ± 0.26 (0.91–2.67)1.38–1.461.47 ± 0.27 (0.91–2.60)1.43–1.510.063
  H/99.07 ± 1.76 (4.50–15.28)8.82–9.329.38 ± 1.82 (4.77–17.18)9.12–9.640.046
  HR/91.00 ± 0.44 (0.00–3.24)0.94–1.061.13 ± 0.57 (0.00–3.95)1.05–1.210.007
  Avg.255 ± .038 (.143–.385)0.250–0.260.261 ± .037 (.152–.389)0.256–0.2660.064

Advanced Pitching Specific Statistics

  FIP4.28 ± 0.95 (2.11–8.38)4.15–4.414.57 ± 1.10 (2.22–7.99)4.41–4.730.001
  xFIP*4.20 ± 0.78 (2.19–6.29)4.06–4.344.38 ± 0.73 (2.40–6.65)4.25–4.510.056
  tERA*4.55 ± 1.06 (2.50–8.81)4.36–4.744.77 ± 1.12 (2.19–8.49)4.57–4.970.191
  WAR1.13 ± 1.22 (−0.50–5.77)0.96–1.300.55 ± 1.01 (−0.80–5.17)0.41–0.690.31E-10
  WPA0.29 ± 1.03 (−1.90–3.64)0.14–0.44−0.03 ± 0.85 (−2.25–3.26)−0.15–0.090.61E-04
  pLI1.07 ± 0.30 (0.31–1.95)1.03–1.11.96 ± 0.27 (0.18–2.05)0.92–1.000.26E-05
  Pace**22.18 ± 2.22 (17.70–30.70)21.69–22.6722.61 ± 2.21 (18.10–29.60)22.12–23.100.003

Pitching Power Statistics and Strikeouts

  K/97.20 ± 2.02 (2.20–12.58)6.91–7.497.02 ± 2.00 (1.74–14.22)6.74–7.300.177
  BB/93.75 ± 1.42 (1.38–11.67)3.55–3.953.84 ± 1.48 (1.16–10.80)3.63–4.050.438
  FB%*63.9 ± 9.7 (21.4–88.9)62.2–65.660.3 ± 9.8 (31.1–85.0)58.6–62.00.42E-07
  FB velocity*91.8 ± 2.7 (81.2–99.3)91.3–92.390.9 ± 2.9 (78.6–97.7)90.4–91.40.002
  Contact %*78.8 ± 4.6 (63.2–89.4)78.0–79.679.3 ± 4.6 (65.5–88.7)78.5–80.10.152
  Strike Zone %*49.2 ± 4.6 (39.9–63.5)48.4–50.047.4 ± 4.7 (33.3–60.2)46.6–48.20.32E-11

Post-UCL, Years 2 and 3 Paired Outcomes Data

Pre-Surgery (n=148)Post-Surgery (n=148)p-value
Mean ± St. Dev (Min–Max)Mean ± St. Dev (Min–Max)
Basic Pitching Statistics

  W4.63 ± 3.82 (0.00–15.67)4.41 ± 4.35 (0.00–19.50)0.663
  L4.65 ± 3.45 (0.00–13.00)4.32 ± 3.29 (0.00–13.00)0.410
  IP82.0 ± 55.3 (6.1–234.1)75.3 ± 57.7 (6.0–234.1)0.339
  Pitches*1283 ± 853 (141–3348)1253 ± 922 (24–3396)0.965
  ERA4.57 ± 1.39 (0.69–11.37)5.12 ± 3.62 (1.43–42.97)0.087
  WHIP1.46 ± 0.26 (0.92–2.67)1.50 ± 0.63 (0.87–8.25)0.516
  H/99.22 ± 1.78 (4.50–15.28)9.47 ± 3.55 (2.09–45.95)0.428
  HR/91.03 ± 0.45 (0.00–2.77)1.14 ± 0.67 (0.00–5.14)0.063
  Avg.258 ± .038 (.143–.385).260 ± .046 (.070–.521)0.690

Advanced Pitching Specific Statistics

  FIP4.42 ± 0.93 (1.52–6.89)4.70 ± 1.48 (1.97–12.73)0.053
  xFIP*4.37 ± 0.74 (2.66–6.29)4.56 ± 1.73 (2.22–20.96)0.895
  tERA*4.65 ± 1.04 (2.12–7.45)5.01 ± 1.91 (2.05–18.45)0.337
  WAR0.89 ± 1.07 (−0.50–5.43)0.71 ± 1.31 (−1.50–5.70)0.244
  WPA0.13 ± 0.96 (−1.90–3.45)0.00 ± 1.10 (−2.62–4.08)0.270
  pLI1.03 ± 0.32 (0.13–1.91)0.96 ± 0.31 (0.34–2.10)0.070
  Pace**21.86 ± 2.04 (17.70–26.83)22.43 ± 2.14 (17.50–28.70)0.009

Pitching Power Statistics and Strikeouts

  K/97.12 ± 2.04 (2.20–12.58)6.89 ± 2.00 (2.21–13.50)0.321
  BB/93.97 ± 1.56 (1.38–11.67)4.04 ± 2.58 (1.16–28.30)0.778
  FB%*63.7% ± 9.6% (21.4%–82.7%)61.3% ± 11.3% (27.6%–87.5%)0.033
  FB velocity*91.39 ± 2.61 (81.17–99.30)90.80 ± 2.80 (83.20–97.70)0.443
  Contact %*79.1% ± 4.4% (70.1%–89.4%)79.9% ± 4.9% (67.0%–88.5%)0.123
  Strike Zone %*49.2% ± 4.6% (39.5%–60.4%)48.0% ± 5.0% (32.4%–60.2%)0.000
Authors

The authors are from the Department of Orthopedic Surgery, Northwestern Memorial Hospital, Chicago, Illinois.

Drs Selley, Portney, Lawton, Shockley, and Christian have no relevant financial relationships to disclose. Dr Saltzman is a paid consultant for and receives royalties from Medacta and Wright Medical. Dr Hsu is a paid consultant for Medtronic, Inc, Stryker, Nuvasive, Wright Medical, Bioventus, Globus, Agnovos, Mirus, and Allosource and receives royalties from Stryker and Inion.

This study was supported by an Internal Resident Grant (450-5242000-40042113-01).

Correspondence should be addressed to: Ryan S. Selley, MD, Department of Orthopedic Surgery, Northwestern Memorial Hospital, 676 N Saint Clair St, Chicago, IL 60611 ( ryan.selley@northwestern.edu).

Received: September 09, 2018
Accepted: November 16, 2018
Posted Online: September 12, 2019

10.3928/01477447-20190906-07

Sign up to receive

Journal E-contents