Kinematic and kinetic findings in high vs. low consistency professional baseball pitchers

Abstract

While the performance metric ball velocity has often been associated with increased kinetics at the upper extremity and risk of injury in baseball pitchers, it is unclear if the performance metric pitch location consistency has any positive/negative associations with pitching kinetics. Professional pitchers subdivided into high(Hcon) and low(Lcon) consistency groups were instructed to throw 8–12 fastballs while assessed with motion-capture technology(480 Hz). To further assess pitching consistency, 95% confidence ellipses with comparisons of major and minor radii were conducted with an external comparison to a cohort of high school pitchers. Lastly, kinematic and kinetic values were compared between Hcon and Lcon professional pitchers. Professional baseball pitchers(n = 338) had consistency in pitch location comparable to high school pitchers(n = 59) (22.0 ± 6.7 vs. 23.2 ± 7.5% grid width respectively, p-value = 0.21). Hcon professional pitchers(n = 91) compared to Lcon pitchers(n = 98) had a smaller major radius(15.2 ± 3.0 vs. 26.3 ± 5.9 respectively, p-value<0.001) and a smaller minor radius(9.4 ± 1.9 vs. 16.1 ± 4.4 respectively, p-value<0.001) in the 95% confidence ellipses. Hcon pitchers compared to Lcon pitchers had increased arm slot(59.7 ± 13.5 vs. 54.7 ± 12.4° respectfully, p-value = 0.009), trunk tilt(-33.4 ± 9.1 vs. −37.2 ± 8.9° respectfully, p-value = 0.004), and trunk lateral flexion(-27.1 ± 9.3 vs. −31.8 ± 9.0° respectfully, p-value<0.001) at ball release. These pitchers also had lower shoulder(112.4 ± 15.9 vs. 118.3 ± 15.1% BW respectfully, p-value = 0.001) and elbow distraction forces(110.5 ± 17 vs. 117.0 ± 15.2% BW respectfully, p-value = 0.006) during arm deceleration. Professional pitchers who approach a sidearm style of pitching, typically involving less contralateral trunk tilt, may achieve higher consistency in their throws while also experiencing diminished peak distractive forces at the elbow and shoulder.

1. Introduction

A baseball pitcher's goal is to not only throw high velocity pitches, but to also throw balls with consistency and accuracy in pitch location.^1,2 Many pitchers strive for consistent pitching movements in order to achieve the most accurate pitch.² Because of the relative difficulty in measuring pitch consistency, pitch velocity is a more ubiquitously cited metric when evaluating players.3, 4, 5, 6 However, accuracy and consistency remain vital factors in player scouting and professional success.^1,7,8

Consistency in pitcher movement has been established as an important metric.8, 9, 10 One study of 190 regular season MLB starting pitchers found that those with increased consistency of pitch release location had greater success (smaller fielding independent pitching), and recommended improving pitch release consistency for player development.⁸ Increased pitch kinematic consistency has also been shown to improve at more advanced playing levels.⁹ While the importance in repeatability of a pitcher's movements when throwing has been recognized, there has not been significant research into pitch location consistency, or repeatability in throwing to the same location, as well as the kinetic and kinematic variables associated with such.¹⁰

A few research groups have evaluated ways to improve pitch location accuracy.^1,11, 12, 13 Hore et al.^11,12 found that pitch accuracy was affected by the timing of finger opening at ball release. They also identified that asynchronously timed finger rotation, relative to the rotation of other nearby joints, plays a role in inaccurate pitches due to the impact on poor ball release timing.¹² Kawamura et al.¹ compared pitching accuracy parameters between 5 professional and 8 high school pitchers, and found that professional pitchers had greater pitch accuracy and smaller error rate. Kusafuka et al.¹³ assessed release parameters in 7 adult baseball players and their impact on pitch accuracy, however, these parameters were focused on ball release locations and pitch angles, and did not assess aspects of the players total body motions.

Studies on pitch consistency have been less common than those on pitch accuracy, focusing predominantly on sensory inputs as predictors for pitch location.^2,10,14 A study of 16 college baseball pitchers assessing the role of balance in pitching consistency found that decreased vestibular input utilization was associated with increased pitching errors.¹⁴ In a study of 12 semi-professional pitchers assessing consistent hand dynamics, researchers suggested that in order to achieve more consistent and accurate pitches, players need to both reduce the variability in joint angles and take advantage of existing joint covariability.²

Despite these investigations, no study to date has assessed the differences in kinetic and kinematic parameters in professional pitchers of varying consistency in pitch location.^7,10 Moreover, while metrics to quantify pitch accuracy have been made,^1,15 no such measurements have been created to validate pitch consistency. By better understanding the variables associated with increased pitch consistency, professional baseball coaches and performance trainers can potentially improve player pitch accuracy. Therefore, this study aims to identify the specific kinetic and kinematic differences between professional baseball pitchers with high and low pitch location consistency. We hypothesize kinematic measures of the shoulder such as external rotation and shoulder abduction, as demonstrated by the Glanzer et al.¹⁰ model for pitch consistency, will significantly differ between pitchers with high and low pitch consistency.

2. Methods

2.1. Pitcher recruitment

Pitchers were recruited from 24 professional baseball teams. All pitchers received clearance from their team physician to play for the season and had a physical exam. Inclusion criteria consisted of (1) pitchers were part of either a Major League Baseball team or Minor League Baseball team comprising all levels of play (Low-A, High-A, AA, AAA), and (2) pitchers had no serious injury within six months of the time of testing (requiring more than 2 weeks of rest/rehabilitation). Data was de-identified and qualified for exempt review under federal guidelines approved by the institutional review board at Hospital for Special Surgery.

2.2. Pitching assessment

Measurements recorded include patient demographics, arm dominance, height and weight, and history of injury. Pitching evaluations were conducted as previously described.¹⁶ The pitchers were given as much warm-up time as typical for their usual routine. When the pitchers were ready, 42 reflective markers were placed on the landmarks described by Luera et al.¹⁶ Pitching was recorded with a 8-camera Raptor-E motion analysis system at 480 hz (Motion Analysis Corp, Santa Rosa, CA, USA) with standard pre-analysis spatial calibration.

Pitchers pitched 8–12 fastballs with game-like effort to a catcher at regulation distance of 18.4 m from a mount. They pitched either the wind-up or stretch. Ball speed was collected with a radar gun from behind the pitcher (Stalker Sports Radar, Richardson, TX, USA). A member of the pitching coach staff recorded the vertical and horizontal coordinates of the ball across the home plate, with the same staff member recording pitch locations for all pitchers of the same team.

2.3. Intra-rater validation

Intra-observer reliability was validated by grading raters on the repeatability of pitch location recordings for all pitchers in a single day. Median absolute deviation of consistency (MAD) utilizing the normalized spread of each pitch was calculated for each day of pitching. MAD has previously been validated as a robust measure of variability.^17,18 It was expected that pitchers within the same cohort of a single day from the same team would not deviate by more than 20% measured among all days. The 20% threshold was set prior to collection of pitch location metrics.

2.4. Data processing

Kinetic and kinematic analysis was performed using MATLAB (The Mathworks, Natick, MA, USA) as described by Luera et al.¹⁶ Data from the 42 reflective markers was filtered by a fourth order, zero lag Butterworth filter with 13.4hz cutoff frequency.¹⁹ Pitch time was measured from maximum knee height ending at maximum shoulder internal rotation. The maximum knee height (MKH) was identified on the frames as when the stride leg knee had maximum vertical displacement. Shoulder maximum external rotation (MER) was defined as the frame where the maximum angle was created between the forearm and an anterior to posterior line in the z axis of the elbow joint. Shoulder maximum internal rotation (MIR) was defined as the frame where the throwing shoulder reaches maximum internal rotation angle following ball release. Ball release (BR) was noted as 0.01 s after the wrist passes the elbow in the forward direction.^19,20 Maximum elbow extension (EE) was defined as the frame when the elbow reached maximum extension angle before the foot contacted the ground. Foot contact (FC) was defined as the frame when the heel or lead toe reaches the minimum in the Z axis. Hand separation (HS) was defined as the time of maximum acceleration of motion from the center of one wrist joint to the other between MKH and foot contact.

Overall, 23 kinematic variables were collected, including stride length normalized by body height. The peak kinetic parameters were recorded during three phases of pitching: (1) arm cocking, (2) arm acceleration, and (3) arm deceleration. The following kinetic values were determined during the arm cocking phase: shoulder internal rotation torque, shoulder horizontal adduction torque, shoulder superior force, shoulder anterior force, elbow varus torque, and elbow medial force. The following kinetic values were determined during the arm acceleration phase: elbow anterior force and elbow flexion torque. During the arm deceleration phase kinetic variables collected included peak shoulder adduction torque, shoulder distractive force, elbow distractive force, and elbow anterior force. All force and torque values were normalized by player weight and player weight multiplied by player height.

2.5. Statistical analysis

Pitch consistency was measured as the location of each pitch from the mean location of all pitches made by that same pitcher. These values were normalized as a percentage of the length and width of the pitching chart. The mean and standard deviations (SDs) were calculated for each pitchers' spreads. Pitchers were then classified as either high consistency (spread less than 0.5 SD below the mean, Hcon) or low consistency (spread more than 0.5 SD above the mean, Lcon). To further assess pitching consistency between groups, the average pitch location of each pitcher within these groups was fitted to a bivariate normal distribution. 95% confidence ellipses were constructed in relation to the relative center of each pitcher (mean location of the pitches, which was then defined as the origin in a X–Y axis). Major radii and minor radii of the ellipses were obtained to further characterize and quantify the spatial distribution differences between pitching groups.¹⁵ Additionally, normalized pitch location trajectory (nPLT) was obtained for each pitcher using a modified version of a method proposed by Kawamura et al.¹ To calculate pitch location trajectory, the distance (as a percentage of the length/width of the pitching grid) of each pitch location between Pitch (n) and Pitch (n + 1) was summed from the first to final pitch. To account for differences in the number of pitches per pitcher in the dataset, PLT was then normalized by dividing by the ∑ n - 1 to obtain nPLT.

As a final means of comparison, consistency of the entire population of professional baseball pitchers was compared to a cohort of high school pitchers to validate these consistency measurements with a different population. This population of high school pitchers had similar methodology in data acquisition with the following exceptions: 1) 8 local high school teams were incorporated, 2) consent to take part in this study was acquired from the pitchers' parents, and 3) clearance to play was carried out by each pitcher's primary care physician rather than a team physician.

Independent two-tailed t-tests were then used to compare the high consistency vs low consistency professional cohorts for all kinetic, kinematic, and temporal variables of interest. For all statistical tests, α was set at 0.01 to adjust for family-wise error. Analysis was performed using Matlab version R2020a (The Mathworks Inc, Apple Hill, Massachusetts).

3. Results

3.1. Demographics

338 professional baseball players were included in this study. The Hcon cohort (n = 91) had no difference in age (22.2 ± 2.4 vs. 21.7 ± 2.0 years respectfully, p = 0.14), handedness (77% vs. 76% right handed) or height (189.1 ± 5.8 vs. 190.1 ± 5.4 cm respectfully, p = 0.24) compared to the Lcon cohort (n = 98). Hcon pitchers were 2.3 kg heavier on average than Lcon pitchers, though this difference was not statistically significant (p-value = 0.077). Professional pitchers were significantly older (avg. difference: 5.6 years), heavier (avg. difference: 20.5 kg), and taller (avg. difference: 9.7 cm) than high school pitchers (p-value<0.001 for all).

3.2. Rater reliability

Professional coaching staff demonstrated the following reliability results. Only 8.8% of observation dates had a consistency MAD >5% grid length while no testing dates had a MAD >15% grid length. These measurements were well within a reasonable amount of deviation from predicted, demonstrating high intra-rater reliability of the coaching staff for consistency. Results of reliability measures for high school coaching staff showed only 7.4% of observation dates had a consistency MAD >5% grid length while no test dates had a consistency MAD >10% grid length, again demonstrating high intra-rater reliability of the coaching staff for consistency.

3.3. Pitcher consistency validation

The Hcon cohort had a shorter average spread of pitch locations, as a percentage of the total pitching grid width, compared to the Lcon pitchers (13.8 ± 4.6 vs. 29.5 ± 3.2% grid width respectively, p-value<0.001). 95% confidence ellipses of pitch distribution of Hcon and Lcon pitchers are shown in Fig. 1. The Hcon major radius was significantly smaller than the Lcon major radius (15.2 ± 3.0 vs. 26.3 ± 5.9 respectively, p-value<0.001) and the Hcon minor radius was significantly smaller than the Lcon minor radius (9.4 ± 1.9 vs. 16.1 ± 4.4 respectively, p-value<0.001). The nPLT for Hcon pitchers was significantly smaller than the nPLT for Lcon pitchers (21.9 ± 10.2 vs. 44.0 ± 12.1 respectively, p-value<0.001).

Fig. 1.

Comparison of 95% confidence ellipse distribution for A) high consistency professional pitchers, and B) low consistency professional pitchers. Ellipses centered at the pitcher's average pitch location.

The pitch distributions for the total cohort of professional pitchers were then compared with the pitch distribution of high schoolers. The professional cohort had an equivalent average spread of pitch locations, as a percentage of the total pitching grid width, compared to the high school pitchers (22.0 ± 6.7 vs. 23.2 ± 7.5% grid width respectively, p-value = 0.21). Average confidence ellipses for the two populations are shown in Fig. 2. Though the professional cohorts 95% confidence ellipse appears smaller than the high school cohort, the professional cohorts’ average major radius and average minor radius were comparable to the high schoolers (major radii: 22.4 ± 5.8 vs. 22.7 ± 6.7, p-value = 0.78; minor radii: 13.8 ± 4.3 vs. 14.0 ± 5.4, p-value = 0.79). The nPLT for the professional cohort was also equivalent to the nPLT for the high school cohort (33.8 ± 13.4 vs. 36.9 ± 14.9, p-value = 0.11).

Fig. 2.

Average 95% confidence ellipses for pitch distributions between professional (n = 338) and high school pitchers (n = 59).

3.4. Biomechanical evaluations

Kinematic and kinetic comparisons between Hcon and Lcon professional pitchers are shown in Table 1. Hcon pitchers had significantly less peak elbow distractive force (110.5 ± 17 vs. 117.0 ± 15.2%BW respectfully, p-value = 0.006) and shoulder distractive force (112.4 ± 15.9 vs. 118.3 ± 15.1%BW respectfully, p-value = 0.001) than Lcon pitchers. Hcon pitchers had significantly greater arm slot at BR than Hcon pitchers (59.7 ± 13.5 vs. 54.7 ± 12.4° respectfully, p-value = 0.009). Given most statistically significant differences were noted with measures of the trunk, trunk kinematics were plotted over the entire sequence of the pitching cycle and compared between Hcon and Lcon pitchers (Fig. 3). Hcon pitchers had significantly less trunk flexion at BR (11.9 ± 10.0 vs. 15.9 ± 9.0° respectfully, p-value = 0.005) and MIR (22.0 ± 10.8 vs. 26.9 ± 9.6° respectfully, p-value = 0.001). Hcon pitchers had significantly greater trunk lateral flexion at MER (−22.2 ± 10.6 vs. −27.2 ± 9.5° respectfully, p-value<0.001), BR (−27.1 ± 9.3 vs. −31.8 ± 9.0° respectfully, p-value<0.001), and MIR (−27.6 ± 9.0 vs. −32.3 ± 9.7° respectfully, p-value<0.001). Lastly, the Hcon cohort had significantly greater trunk tilt at BR (−33.4 ± 9.1 vs. −37.2 ± 8.9° respectfully, p-value = 0.004) and MIR (−45.6 ± 10.6 vs. −50.6 ± 10.3° respectfully, p-value = 0.001).

Table 1.

Kinematic and Kinetic Comparisons in High Consistency (Hcon) and Low Consistency (Lcon) Pitchers. Data presented as mean ± standard deviation, statistical significance (p-value<0.01) denoted by *. Note: FC, foot contact; MER, maximum shoulder external rotation; BR, ball release; BW, body weight; BH, body height.

	Hcon Pitchers	Lcon Pitchers	p-value
Kinematics
Ball Velocity, m/s	37.9 ± 1.8	38.4 ± 1.9	0.055
Stride Length, %BH	77.8 ± 5.5	79.4 ± 5.3	0.048
Stride Width at FC, cm	0.1 ± 0.2	0.1 ± 0.2	0.635
Lead Foot Rotation at FC, °	17.8 ± 11.1	14.1 ± 12.4	0.030
Lead Knee Flexion at FC, °	46.5 ± 8.6	46.1 ± 8.1	0.735
Lead Hip Flexion at FC, °	59.4 ± 12.5	59.9 ± 13.9	0.770
Back Hip Flexion at FC, °	−1.5 ± 14.7	−2.1 ± 14.5	0.772
Lead Hip Internal Rotation at FC, °	−8.3 ± 14.4	−10.7 ± 15.7	0.285
Back Hip Internal Rotation at FC, °	13.4 ± 11.9	13.1 ± 11.7	0.849
Shoulder External Rotation at MER, °	165.0 ± 8.8	165.0 ± 10.9	0.970
Elbow Flexion at BR, °	32.1 ± 5.9	31.9 ± 6	0.820
Shoulder Abduction at BR, °	90.7 ± 8.3	92 ± 8.3	0.284
Shoulder Horizontal Adduction at BR, °	3 ± 9.1	2.1 ± 8.4	0.440
Pelvic Rotation at BR, °	−13.2 ± 9.8	−11.8 ± 9.5	0.350
Pelvic Tilt at BR, °	−22 ± 7.9	−22.1 ± 7.7	0.958
Pelvis Obliquity at BR, °	9.2 ± 8.1	9.7 ± 9.2	0.673
Trunk Flexion at BR, °	11.9 ± 10	15.9 ± 9	0.005*
Trunk Lateral Flexion at BR, °	−27.1 ± 9.3	−31.8 ± 9	<0.001*
Trunk Tilt at BR, °	−33.4 ± 9.1	−37.2 ± 8.9	0.004*
Trunk Rotation at BR, °	−16.8 ± 8.8	−16.5 ± 8.2	0.845
Trunk Obliquity at BR, °	−16.6 ± 11.8	−19.9 ± 11.3	0.056
Forearm Pronation at BR, °	4.2 ± 13.8	3.6 ± 15.1	0.765
Arm Slot at BR, °	59.7 ± 13.5	54.7 ± 12.4	0.009*
Peak Kinetics
Elbow Varus Torque, %BWxBH	4.9 ± 0.7	5.0 ± 0.9	0.267
Elbow Medial Force, %BW	40.0 ± 5.8	40.7 ± 6.4	0.419
Elbow Anterior Force, %BW	41.5 ± 5.4	41.8 ± 6.2	0.671
Elbow Flexion Torque, %BWxBH	4.0 ± 0.6	4.0 ± 0.7	0.499
Elbow Distractive Force, %BW	110.5 ± 17	117.0 ± 15.2	0.006*
Shoulder Internal Rotation Torque, %BWxBH	5.0 ± 0.7	5.0 ± 0.8	0.695
Shoulder Horizontal Adduction Torque, %BWxBH	5.8 ± 1	5.5 ± 1	0.070
Shoulder Superior Force, %BW	17.5 ± 8.8	18.8 ± 9.5	0.325
Shoulder Anterior Force, %BW	41.9 ± 7.7	42.5 ± 7.2	0.555
Shoulder Adduction Torque, %BWxBH	8.6 ± 2.3	8.0 ± 2.0	0.048
Shoulder Distractive Force, %BW	112.4 ± 15.9	118.3 ± 15.1	0.001*

Fig. 3.

Trunk Parameters Between High Consistency (Hcon) and Low Consistency (Lcon) Professional Pitchers. Trunk kinematics featured include: A) trunk flexion, B) trunk lateral flexion, C) trunk tilt, D) trunk obliquity, and E) trunk rotation. Statistical significance denoted as * for 0.01>p-value>0.001, while for p-value<0.001. Note: MKH, maximum knee height; HS, hand separation; EE, elbow extension; FC, foot contact; MER, maximum shoulder external rotation; BR, ball release; MIR, maximum shoulder internal rotation.

4. Discussion and implications

This study was able to differentiate professional pitchers based on high consistency verses low consistency in pitch location, validated by several metrics. Professional baseball pitchers were found as a whole, to have consistency in pitch location comparable to high school pitchers. Hcon professional pitchers compared to Lcon pitchers had: 1) higher arm slot at BR, 2) higher trunk tilt at BR and MIR, 3) greater trunk lateral flexion at MER, BR, and MIR, and 4) decreased trunk flexion at BR and MIR. These pitchers also had less shoulder and elbow distraction forces during the arm deceleration phase of the pitch.

There was no difference between high school and professional pitcher consistency in pitch location. This was a curious finding given the postulation that professionals likely have superior pitch location performance metrics compared to more novice players. It may be possible that even if both cohorts throw fastballs that land in similar positions, the professionals are reaching there with more velocity and likely more movement on the ball. This gives the interesting thought that performance metrics may go beyond ball speed or final pitch location, rather considering the motion of the ball as well as the degree of spin may increase the complexity of a professional pitchers pitch. Conducting future studies utilizing Rapsodo(Rapsodo Baseball System, Rapsodo Inc. Fishers, IN) or other pitch trajectory tracking systems to compare these more nuanced performance metrics may be warranted.²¹

Lcon pitchers had increased contralateral trunk tilt, shoulder distraction force, and elbow distraction force compared to Hcon pitchers. Prior studies demonstrated pitchers with excessive contralateral trunk tilt (ie. less positive trunk tilt) achieved increased shoulder internal rotation torque, shoulder distractive force, and elbow varus torque in collegiate and high school pitchers.22, 23, 24 This may suggest the increased shoulder and elbow distraction forces in Lcon pitchers is more directly a result of the excessive contralateral trunk tilt these pitchers experience at late stages of the pitching cycle. When the shoulder is abducted and outwardly rotated during the late arm-cocking phase, contralateral trunk tilt toward the nonthrowing shoulder generates greater acceleration of the throwing shoulder and elbow.^22,25,26 This increased acceleration in a posterior to anterior direction can create more distraction, defined as a shear force on the humeral head, at the shoulder joint.²⁷ Distraction at the shoulder joint specifically, has been linked with pathologic conditions of the rotator cuff and glenoid labrum with important clinical implications as a potential risk factor for injury.^28,29 Therefore, minimizing trunk tilt may be an effective means to not only improve consistency in pitch location, but also possibly decrease peak distractive forces on the shoulder and elbow.

An alternative explanation as to the kinetic differences observed can be based on pitcher ball velocity. Indeed, increased contralateral trunk tilt has been associated with ball velocity in several studies.22, 23, 24 Furthermore, to achieve higher ball velocity in pitches, it intuitively makes sense that consistency in pitch location (as well as accuracy for that matter) may in turn be conceded. Though a statistically significant difference was not observed between Lcon and Hcon pitchers (p-value = 0.55), Lcon pitchers did have 0.5 m/s (1.1MPH) higher ball velocity. This difference in ball velocity can explain the differences in distractive forces noted at the shoulder and elbow.²² With increasing segmental velocities, distractive forces experienced at both the shoulder and elbow are likely necessitated to oppose the centrifugal force generated at the glenohumeral and elbow joints.

Hcon differed from Lcon pitchers predominantly based on trunk kinematic parameters during late stages of the pitching cycle (Fig. 3). Trunk tilt has previously been shown to be a predictor of pitch location consistency in pitchers of various playing levels.¹⁰ Improving pitch location performance, therefore may be achieved by keeping a more neutral, consistent posture with limited lean away from the pitching arm.³⁰ Robb et al.³¹ demonstrated flexibility of the hip muscles influences pelvis and trunk kinematics during pitching, suggesting the incorporation of training programs that strengthen the abdominal core muscles may aid the trunk in retaining a constant posture throughout the pitching cycle. In addition, to achieve higher consistency, pitchers may consider making the following adjustments in their trunk motions based on our observations. Professional pitchers can be instructed to keep their trunk less flexed at the moment of ball release and to increase trunk tilt towards the side of the pitching arm while concomitantly increasing lateral flexion. This may help them utilize trunk motion in the transverse plane and become less dependent on trunk motion in the frontal plane.²²

Arm slot and trunk tilt are intricately related,22, 23, 24^,32 with both observed to be significantly higher in Hcon pitchers. Pitchers of varying arm slots demonstrate meaningful differences in kinematic and kinetic parameters.³³ Hcon pitchers had a greater angle between the elbow joint and ground at the moment of BR, suggesting these pitchers have greater control of their arm motion when the forearm is angled more distantly from the horizontal plane.

Limitations of this study are as follows. Only a handful of pitches were collected for each pitcher, which may not adequately represent gameday effort where player exertion and fatigue likely play an important role in performance outcomes.34, 35, 36, 37, 38, 39, 40 Pitching measurements were assessed only for fastballs and thus the applicability of these results to other pitch styles is unclear. Coaching personnel recorded pitch locations for each pitch live, subjecting the process of data collection to human error. Lastly, though kinetics is utilized in this study as a surrogate for ligament and soft-tissue loading, we do not directly quantify these measures and therefore no conclusions on injury risk profiles can be definitively made.

5. Conclusion

Professional baseball pitchers had consistency in pitch location comparable to high school pitchers. Professional pitchers with high consistency in pitch location differ from low consistency pitchers predominantly based on trunk kinematic parameters at late stages of the pitching cycle. These pitchers had increased arm slot, trunk tilt, and trunk lateral flexion as well as decreased trunk flexion at BR and MIR. These pitchers also had less shoulder and elbow distraction forces during arm deceleration. Professional pitchers who approach a sidearm style of pitching, typically involving less contralateral trunk tilt, may achieve higher consistency in their throws while also experiencing diminished peak distractive forces at the elbow and shoulder.