Abstract
Background:
After a pitching injury, players must go through a return to sports protocol to enable them to get back to competition. However, this should be done while reducing the risk of reinjury. In the early stages of the return to sports protocol, it is important to minimize kinetics while ideally working on pitching mechanics, which may be achieved through interval throwing progressions.
Purpose:
To assess the differences in kinetics and kinematics during reduced effort pitching.
Study Design:
Descriptive laboratory study.
Methods:
Collegiate-aged pitchers (n = 19) throw 5 fastballs at each effort level, including 60%, 70%, 80%, 90%, and 100%. Kinetics and kinematics were recorded. Variables of interest included elbow varus torque, shoulder rotation at maximum external rotation (MER), elbow flexion at MER, maximum resultant shoulder force, peak pelvis rotation velocity, peak trunk rotation velocity, peak shoulder internal rotation velocity, peak elbow extension velocity, shoulder abduction at MER, maximum shoulder horizontal abduction, and maximum hand velocity. Once the data were extracted from the Kinatrax database, separate analyses of covariance tests were performed on each set of data, followed by a Tukey Honest Significant Difference post hoc test when the analysis of variance test returned a statistically significant P value (P < .05).
Results:
Only elbow varus torque was found to have statistically significant differences between effort levels, and only a statistically significant difference between 100% and 60% effort levels was found. At 100% effort level, elbow varus torque showed a mean value of 92.5 N·m, while 60% effort level pitching showed a mean elbow varus torque of 73.2 (P = .017).
Conclusion/Clinical Relevance:
These results suggest that players can throw at close to half effort to reduce their elbow kinetics while maintaining kinematics that would be occurring at 100% effort pitching.
A total of 49,955 injuries were sustained by Major and Minor League Baseball athletes between 2011 and 2016, with upper extremity injuries accounting for 39% of the injuries. Close to 10% of these injuries were season-ending, with ulnar collateral ligament (UCL) injuries being the leading cause. 4 Shoulder and elbow injuries have the greatest injury prevalence in college pitchers. 3 In professional baseball, almost 40% of all elbow ligament injuries were found to have ended in surgery. 6 Pitchers who specifically undergo UCL reconstruction are more likely to have reduced pitching workloads after surgery and demonstrate decreased fastball velocity, reducing the likelihood of returning to the previous level of play. 20
After surgery, pitchers need to demonstrate a normalized range of motion, strength, and scapular mechanics before safely moving on to a throwing plan to avoid reinjury. 16 Return to throwing programs, in the form of interval throwing progressions, are often used for players who are returning to sports from a throwing-related injury. 17 Interval throwing progressions begin with gentle throws, gradually increasing in intensity and distance until the thrower reaches maximum effort. 17 During rehabilitation and early return to throwing phases, arm kinetics should be kept low to avoid reinjury. 18 When the medial elbow undergoes high torques, the UCL has the potential to be at an increased risk of injury. 2 In addition, shoulder distraction forces have been associated with rotator cuff and glenoid labrum injuries. 1 These kinetics should be limited in the early phases of rehabilitation.
Changing throwing distances are traditionally used to control arm kinetics during the rehabilitation process; however, utilizing distance-based throwing measures does not limit arm kinetics as previously hypothesized. 10 Throwing at reduced effort levels is another potential intervention that may be beneficial to reduce arm kinetics. In a pilot study of 10 healthy high school pitchers, throwing at 50%, 75%, and maximum effort, demonstrated reductions in elbow varus torque, but not at a proportional rate. 7
In recent years, biomechanical analyses of the throwing athlete have identified key kinematics that may predispose an athlete to increased arm kinetics and injury risk.5,13 These kinematics include shoulder and elbow angles, as well as pelvis, trunk, and arm rotational velocities. Identifying poor mechanics could be instrumental to a safe return to throwing after an arm injury. However, 3-dimensional (3D) motion analysis evaluations are normally not conducted until a player is throwing at 100% effort, as it is thought that unrestricted effort throwing represents an athlete’s true mechanics. This assumption is unproven, as there is no literature comparing kinematics at different effort levels. During a return to throwing program, once a player is throwing at 100% effort, it may be more difficult to make mechanical adjustments, and players may have already incurred increased kinetics to their injured limb. If kinematics are similar at lower effort levels, biomechanical analyses could be conducted earlier in a player’s return to play program.
It is unknown whether arm kinetics are reduced or whether there are similar kinematics between varying submaximal effort throws that mimic interval throwing progressions. While preliminary pilot work has evaluated the association between 3 basic effort levels and elbow kinetics, larger samples are required that evaluate multiple kinetic and kinematic parameters. This study aimed to assess potential differences between various effort levels of fastball pitches on 11 different arm kinetic and kinematic variables. Observing these potential differences could help create safer protocols for pitchers to return to throwing after elbow or shoulder surgery.
Methods
Study Design
A cross-sectional study was performed on National Collegiate Athletic Association (NCAA) Division 1 baseball pitchers who were participating in spring preseason training. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) 21 reporting guidelines and was approved by the university’s institutional review board. Participants signed informed consent before data collection.
Participants
The inclusion criteria for this study included baseball players who (1) had a pitcher listed as one of their positions; (2) were a division 1 collegiate pitcher, and (3) were fully involved in spring preseason practices. The exclusion criteria included players who (1) were not pitching in live at-bats as a part of their preseason practice; (2) were in a rehabilitation program because they were not at 100% throwing health; or (3) were not throwing because of an injury.
Outcome
Outcomes for this study included maximum elbow varus torque (MEVT), shoulder rotation at maximum external rotation (MER), elbow flexion at MER, maximum resultant shoulder force (MRSF), peak pelvis rotation velocity, peak trunk rotation velocity, peak shoulder internal rotation velocity, peak elbow extension velocity, peak shoulder abduction at MER, maximum shoulder horizontal abduction, and hand velocity, which was used as a surrogate for ball velocity. 15 Each of these variables was calculated using data processing methods and established algorithms developed by KinaTrax.
Explanatory Variable
Players threw fastballs at 5 self-determined effort levels 7 of 60%, 70%, 80%, 90%, and 100%. The 5 effort levels were the explanatory variables that influenced the throwing arm kinetic and kinematic observed outcomes.
Data Collection
For data collection, 3D motion capture data were collected on pitchers using an 8-camera marker-less motion analysis system (KinaTrax, Inc). Marker-less motion capture has been validated against marker-based motion capture systems 19 and has been utilized for collegiate baseball pitchers in the literature. 8 Motion capture data were recorded at 400 Hz. Each pitcher was instructed to go through a normal pregame warm-up period, which varied for each pitcher based on their preferences. Warmups ranged from 10 minutes to 30 minutes and included stretching, weighted ball throwing, and resistance band exercises. After their warmups, pitchers addressed the mound for data collection.
The data collection process was explained to the pitchers before throwing. They were told that they would be throwing 3 fastballs at various effort levels, including 60%, 70%, 80%, 90% and high-intensity effort levels. Before they stepped on the mound, they threw up to 5 warm-up pitches in front of the mound to reacclimate their arm to throwing. Once they were ready for data collection, they were instructed on which effort level to throw, starting with 60%, and progressively increasing in effort to 70%, 80%, 90%, and 100%.
Data Extraction
Once processing and calculations were complete, the 11 outcomes listed above were extracted from the KinaTrax Dugout database.
Data Quality Checks
Data were initially investigated to remove any potential misreads that would skew the data. Biomechanical data were assessed for outliers for each pitcher at each effort level. Outliers were assessed for biological plausibility and consistency. Misreads included all biomechanical data for 1 pitch, 3 maximum MRSF values, 1 MER at MER value, 1 maximum pelvis rotational velocity value, 1 maximum trunk rotational velocity value, and 7 maximum shoulder horizontal abduction values because of values that were out of a biological plausibility range. Each outcome for each pitcher at each effort level was then averaged.
Statistical Analyses
There was only 1 pitch out of 278 (<0.01%) categorized as missing data, and a complete case analysis was performed. Pitches were averaged per participant and effort level before analyses. Descriptive statistics were calculated, including mean, standard deviation, minimum, maximum, and median values, and were visualized in box and whisker plots. Four separate analyses of covariance tests were performed to assess the potential differences of various effort level fastballs on shoulder rotation at MER, elbow flexion at MER, MRSF, and MEVT. The Tukey Honest Significant Difference post hoc test was performed. All statistical analyses were performed in R 4.3.2 with emmeans, lsmeans, lme4, Matrix, and dplyr packages.
Results
Participants and Descriptive Data
Nineteen pitchers were included, with a majority being right-handed (63.16%) (Table 1), and 277 fastball pitches were thrown. A total of 56 pitches were analyzed for 60% and 80% effort levels, 57 pitches were analyzed for 70% and 90% effort levels, and 51 pitches were analyzed at the 100% effort level.
Participant Characteristics a
Continuous data are reported as mean (SD), and count data are reported as n (%). BMI, body mass index.
Differences in Kinetics and Kinematics by Percent Effort
There were statistically significant differences in percent effort pitched for MEVT (N·m) (P = .021) (Figure 1). There were no statistically significant differences between percent effort levels for shoulder rotation at MER (deg) (P = .307), elbow flexion at MER (deg) (P = .994), and MRSF (N) (P = .155). In addition, there were no statistically significant differences between percent effort levels for peak pelvis rotation velocity (deg/sec) (P = .498), peak trunk rotation velocity (deg/sec) (P = .275), peak shoulder internal rotation velocity (deg/sec) (P = .211), peak elbow extension velocity (deg/sec) (P = .317), peak shoulder abduction at MER (deg) (P = .489), maximum shoulder horizontal abduction (deg) (P = .88), and hand velocity (m/sec) (P = .111). Pitch kinetics and kinematics at various effort levels are presented in Table 2.
Box and whisker plot for maximum elbow varus torque showing means, medians, upper and lower ranges, and outliers.
Pitch Kinetics and Kinematics at Various Effort Levels a
EEV, peek elbow extension velocity; EF, elbow flexion; peek elbow extension velocity; HV, hand velocity; MER, maximum external rotation; MEVT, maximum elbow varus torque; MRSF, maximum resultant shoulder force; PRV, peek pelvis rotation velocity; SA, shoulder abduction; SIRV, peek shoulder internal rotation velocity; SHA, shoulder horizontal abduction; SR, shoulder rotation.
Only 100% and 60% effort levels demonstrated statistical differences for EVT (P = .017) (Table 3).
The bold P value indicates a significant difference at P < .05. MD, mean difference.
Discussion
Minimal kinematic and kinetic differences were found when throwing at various effort levels. Only MEVT demonstrated notable differences between effort levels, with only the 100% effort level compared with the 60% effort level, showing a statistically significant difference of 19.35 N·m. Shoulder kinetics, hand speed, and other kinematics observed did not have significant differences between effort levels.
Only the lowest and highest fastball pitching effort levels of 100% and 60% exhibited differences in peak elbow varus torque. The highest and lowest effort levels showed a mean difference of 19.35 N·m, which equates to a change in varus torque of 20.9%. While other structures around the UCL complex absorb some of the forces generated from throwing a pitch, the tensile limit of the UCL itself was found to be between 30 and 40 N·m of the varus torque, 11 and 19 N·m represents around half of the tensile limit. After the repetitive nature of pitching, this adds up, leading to a high risk of injury to the UCL in the elbow. This large percentage of change shows the significant reduction in stress that the elbow experiences at a 60% effort level compared with a 100% effort level. This finding differs from previous literature that found pitching at 75% compared with 100% resulted in decreased elbow stress. 7 Fiegen et al 7 found that elbow stress significantly decreased with 25% reduced perceived effort. These disparate findings could be due to a couple of factors. The first being that Fiegen et al included only high school pitchers whose mechanics differ substantially from college pitchers, and these have an inconsistent relationship between elbow torque and ball velocity. 14 Furthermore, the present study featured only 19 participants, whereas Fiegen et al only had 10 participants; accordingly, these are small sample sizes that could lead to variability in the data.
Interestingly, hand velocity showed no statistically significant differences between effort levels. Using hand velocity as a surrogate for ball velocity, 14 these results are not consistent with previous literature that ball velocity decreases with decreased effort levels. Multiple studies have shown that ball velocity decreases from 100% to 75% reduced throwing effort, and then again from 75% to 50% effort throwing.7,12 Another study observed that when an athlete was asked to throw at reduced effort levels, the throwing metrics did not decrease proportionally with the perceived effort. 22 Considering these findings, it is possible that the pitchers in the present study were not throwing at the true effort levels that they were instructed to, which could result in similar hand velocities across all effort levels.
No significant differences were observed in shoulder distraction forces across effort levels. Because of the limited studies available on percent effort throwing, there have not been any other studies observing shoulder distraction forces across various effort levels. However, previous literature has shown that shoulder rotation at MER and ball velocity are high predictors of shoulder distraction force. 13 Because there was no statistically significant difference for shoulder rotation at MER and hand velocity, which is used as a surrogate for ball velocity, 15 it would also support the lack of statistically significant differences in shoulder distraction force.
There were no statistically significant differences in any of the kinematic variables across the various effort levels. Melugin et al 12 found similar results when observing shoulder rotation at MER and found no differences across various effort levels. 12 There have been limited studies on the other kinematic differences across effort levels. The pitchers in this study showed that they were able to maintain observed kinematics from 100% to 60% effort level pitching while reducing elbow kinetics. Collegiate-level pitchers are high-level pitchers and have advanced repeatable movement patterns, which enable them to maintain kinematics despite decreases in their perceived effort levels.
This study allows comparisons of future investigations that may aim to reduce throwing arm kinetics via reduced effort throwing. Second, although minimal kinetic and kinematic differences were observed, throwing at close to half effort showed decreased elbow varus torque, which was consistent with previous literature. 7 Throwing at 50% to 60% effort may be advised for pitchers as they begin interval throwing programs, particularly during recovery from elbow injury. Furthermore, no evidence suggests that lower efforts will protect shoulder-injured tissue during return to throwing programs at this time because there is no significant decrease in shoulder force at MER during reduced effort level pitching. Pitching mechanics are still maintained at low effort levels because no differences were determined in pitching kinematics. Thus, lower percent effort throwing will allow for continued proper throwing technique, preparing pitchers for higher effort throws as they progress through rehabilitation, and can be utilized to assess pitching mechanics at lower effort levels. Mechanics can be assessed with a 3D pitching evaluation early in the return to throwing plan, to ensure appropriate mechanics and reduce the risk of reinjury. 12
Limitations
As with any small-scale biomechanical cross-sectional study, certain limitations must be acknowledged. This study was only done on healthy NCAA Division 1 collegiate-level pitchers. Thus, the results may not translate to pitchers at other skill levels. In addition, the mechanics seen in a healthy pitcher may be different than a pitcher who is in the middle of a rehabilitation program. Perceived effort does not always equate to actual effort during throwing exercises, 12 and in the present study, hand velocity did not see a significant decrease between effort levels. Hand velocity was used as a surrogate for velocity, when ideally, ball velocity would have been measured and utilized for analysis. Effort level awareness and precision may be different by level of competition, varying significantly between professional and youth pitchers. While effort levels could not be confirmed, instructing an athlete to perform at an effort level is consistent with previous literature.7,9 This study only evaluated fastballs, and other pitch types—including breaking balls and changeups— may produce different arm kinetics and kinematics.7,9 Because of the logistics of data collection, percent effort had to be collected in a sequential order, which may have caused an ordered effect bias.
Conclusion
Pitchers throwing at effort levels between 60% to 100% demonstrated differences in MEVT. Only effort levels of 60% and 100% were significantly different, with a mean difference of 19.35 N·m. Shoulder kinetics, hand velocity, and other observed kinematics did not have significant differences across pitching effort levels of 60% to 100%. Additional studies are needed to observe the relationships between perceived effort levels and actual effort levels.
Footnotes
Acknowledgements
The authors acknowledge the Wake Forest Baseball pitchers and coaching staff for working with us to make this study happen.
Final revision submitted March 24, 2025; accepted March 31, 2025.
One or more of the authors has declared the following potential conflict of interest or source of funding: Funding for this study was provided by The American Orthopaedic Society for Sports Medicine and The Aircast Foundation. B.R.W. has received support for education from SouthTech Orthopedics, Arthrex, and Peerless Surgical; honoraria from Vericel and Musculoskeletal Transplant Foundation; consulting fees from FH Orthopedics, Vericel, DePuy Synthes Products, Medical Device Business Services, and Arthrex; and hospitality payments from Piedmont Plus Innovation; and is a paid speaker for Arthrex and Vericel. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from the Wake Forest School of Medicine Institutional Review Board (IRB) (IRB, 00100492).