Biomechanics of Pitching with Emphasis upon Shoulder Kinematics

Biomechanics of Pitching with Emphasis upon Shoulder Kinematics Charles I. Dillman, PhD1 Glenn S. Fleisig, MS2 lames R. Andrews, MD" The American Sports Medicine Insfitute conducts research to increase understanding of mechanisms involved in upper extremity injuries to throwing athletes. This paper presents a qualitative overview of pitching and a detailed quantitative description of arm motion about the shoulder during this highly dynamic activity. Data on kinematics of arm motions about the shoulder are presented for 29 elite throwers. The major motion about the shoulder is externallinternal rotation. Scapulothoracic and glenohumeral flexibility permit the arm to reach an externally rotated position of 175". Approximately 30 msec before release, the arm internally rotates 8O0, reaching peak angular velocities near 7,00O0/sec. In rehabilitation of injured throwers, there is a need to appreciate the Charles J. Dillman Glenn S. Fleisig highly dynamic nature of this skill and to attempt to simulate these dynamic motions and loads as part of the final phase of treatment before the athlete returns to competition. 0 ver the last several years at the American Sports Medicine Institute, a series of studies has been initiated to g;tin understanding of the niechanisms of upper extremity in-juries to the throwing athlete. This research initially concentrated on pitching and included: Iniproving qualitative understanding of the pitching motion through high-speed videograph y, Developing a clinical procedure for evaluating in.jured pitchers. Conducting quantitative three-dimensional descriptions of upper extremity kinematics, Key Words: shoulder kinematics, biomechanics, pitching ' Executive Director, The Steadman Sports Medicine Foundation, 181 W. Meadow Dr., Suite 420, Vail, CO 81657; lormerly, Director ol Research, American Sports Medicine Institute (ASMI), Birmingham, A1 Director of Research, ASMI Biomechanics laboratory, Birmingham, A1 I Medical Director, Chairman ol the Board, ASMI, Alabama Sports Medicine and Orthopaedic Center, Birmingham, A1 4) Analysis of the resultant joint forces and torques created in pitching, and 5) Conducting appropriate cadaveric studies to assess the effects of these external loads upon the internal structures of the upper extremity. This paper includes the results from Studies 1 (qualitative analysis of pitching) and 3 (arm motions about the'shoulder during pitching). Publications of the other aspects of our work are listed in the bibliographv (3.4,9,10,17). PREVIOUS RESEARCH Most published research on throwing (pitching) has been descriptive studies limited to two-dimensional analysis of a few subjects (l,2, 5,l 1-16). The exception to these studies is the work by Feltner and Dapena (6-8), who conducted a three-dimensional kinematic and kinetic analysis of six college pitchers. The purpose of their research was to describe, for the first time, the real threedimensional motions and forces created in throwing a baseball. In their main article ('i), the results of one subject were presented. The major motion abou; the shoulder in th;owing was external/internal rotation. The external rotation reached a maximum of 170 (80" in their coordinate system), and the peak internal rotational velocity achieved a phenomenal 6,100 /sec. Our purpose was to replicate some of Feltner and Dapena's analyses as one part of a series of studies to improve understanding of upper extremity injuries (anterior Volume 18 Number 2 August 1993 0JOSPT

subluxation, impingement, rotator cuff, etc.) to throwing athletes. QUALITATIVE ANALYSIS OF PITCHING Compared to all throwing activities, pitching a baseball is perhaps the most dynamic. This skill can be divided into six phases: windup, stride, arm cocking, arm acceleration, arm deceleration, and followthrough. The following qualitative description is for a right-handed pitcher. Y IAI (Dl (El Windup Pitching has evolved into a downhill throwing skill from a mound with a vertical height of 10 in. From a standing position facing the batter, the pitcher initiates the throw by stepping backward with what will become the stride foot/leg. With the body weight momentarily supported by the stride foot, the sup porting foot is placed laterally in front of the rubber (Figure la). When the weight is shifted back from the stride foot to the supporting foot, the windup is initiated. This shifting of weight from the stride foot to the supporting foot sets the rhythm for the delivery of the pitch. As the windup is initiated, the body rotates 90, and the striding leg is elevated and flexed so that the left side of the body is now facing the batter (Figure 1 b). It is important for the pitcher to achieve a good balanced position when the knee of the stride leg has reached its maximum height. From this position, the delivery of the ball to the catcher is initiated. FIGURE 1. Sequence of motion in pitching (A-K). key element is to keep the trunk back as much as possible to retain its potential for contributing to the velocity of the pitch (Figures lc-e). As the striding leg moves downward and toward the catcher, the hand/ball breaks from the glove and moves in a down/upward motion in rhythm with the body. Removal of the ball from the glove when the stride is initiated and the down/upward motion of the arm ensure that the throwing arm will be properlv synchronized with the body. This coordination is one of the most critical aspects of throwing (Figures 2a-c). If the throwing arm and striding leg are coordinated proper1 y, the arm will be up in a semicocked position when the stride foot contacts the ground (Figure 2d). The stride should be long enough for the pitcher to stretch out the body but not so long that the Stride After the windup, the supporting leg is flexed, lowering the body, and the left foot/leg is moved toward the plate. Normally the stride is directed toward the catcher. The FIGURE 2. Proper motion of handlball during stride. A) Lateral motion toward plate. B) Down-up motion o/ hand. C) Hand on top of ball. D) Arm up in semicocked position. JOSPT Volume I8 Number 2 August 1993

athlete cannot rotate his legs and hips properly. For most pitchers, the stride length from the rubber should be slightly less than the pitcher's height (Figure 3). Perhaps more important than the stride length is the location of the front foot. The stride foot should land almost directly in front of the back foot, with the toes pointing slightly in (Figure 3). If the foot is placed too much toward the pitcher's right, the pitcher may end up "throwing across his body," which means that the hips will not be able to rotate and the athlete will end up throwing without much energy contributed by the lower body. Conversely, if the foot is placed too much toward the left, the pitcher is "too open," which will cause the hips to rotate and face the batter too early. As a result of such improper timing, energy from the hips will be applied to the trunk too soon and will not help the upper trunk rotate. A. Mean = 75% of height (Standard deviat~on = 4%) B. Mean = 87% of hetght (Standard devtatton =5%) C. Mean = +0.4 crn (Standard devtation = 8.3crn 0 Mean = 150 (Standard deviation = lo0) FIGURE 3. The stride. Arm Cocking (Figures 1 f-h). Once the stride toward the plate is completed, the trunk moves laterally toward the catcher and hip rotation is initiated. Trunk rotation follows the hip but, in highly skilled pitchers, hyperextension of the upper trunk occurs as it is rotated around to face the plate. As the trunk is undergoing rotation and extension, the upper arm is flexed at the elbow, and the shoulder undergoes external rotation (cocking of the arm). As the trunk faces the batter, the shoulder achieves maximum external rotation, and the arm cocking phase is completed (Figure 1 h). It is better to refer to this phase as "arm cocking" Compared to all thro wing activities, pitching a baseball is perhaps the most dynamic. rather than simple "cocking." Clearly, only the arm is cocked by the end of this phase, while the thrower's legs, hips, and trunk have already accelerated. Arm Acceleration (Figures I h, i). The arm acceleration phase starts when the humerus begins to internally rotate about the shoulder. To pitch properly and efficiently, a short delay between the onset of elbow extension and shoulder internal rotation is crucial. By extending the arm at the elbow, the pitcher can reduce the inertia that must be rotated at the shoulder. With less inertia, the internal rotation torque generated at the shoulder can accelerate the arm to a greater angular velocity. When the ball is released, the trunk is flexed, the arm is almost in a fully extended position at the elbow, and the shoulder is undergoing internal rotation. At release, the pitcher's trunk should be tilted forward and the lead knee should be extending. The arm acceleration phase ends with the release of the ball (Figure 1 i). Arm Deceleration (Figures 1 i, j). After ball release, the arm continues to extend at the elbow and internally rotate at the shoulder. These two motions may help settle a controversy that has existed in baseball since slow-motion video was first introduced. On many pitches, such as the curveball, a pitcher may insist that his forearm is not pronating after release, while a coach or researcher with slow-motion video is certain that he sees excessive pronation. In truth, the "forearm pronation" observed may actually be the combined effect of extension at the elbow and internal rotation at the shoulder. In the arm deceleration phase. shoulder internal rotation angular velocity decreases to zero from its maximum value observed near the time of ball release. Arm deceleration ends when the arm has reached an internal rotation position of a p proximately 0" (Figure lj). Follow-Through (Figures Ij, k). The importance of a good follow-through is often overlooked. Although a good followthrough cannot directly improve the throw, it is critical in minimizing the risk of injury. Follow-through is completed with extension of the stride leg, continued hip flexion, shoulder adduction, horizontal adduction, elbow flexion, and forearm supination. The highly dynamic nature of this skill is perhaps best indicated by the fact that the average time from 404 Volume 18 Number 2 August 1993 JOSPT

foot contact of the stride leg until ball release is 0.145 seconds. During this brief interval, the ball is typically accelerated from 4 to 85 mph, and the motions of hip rotation, trunk rotation, upper trunk extension, elbow flexion, shoulder external rotation, elbow extension, hip flexion, upper trunk flexion, shoulder internal rotation, and pronation of the forearm are performed in sequence. There is no doubt that pitching is one of the most dynamic movements in all sports! METHODS Since our series of studies had a dual purpose-clinical and research-an automated high-speed video digitizing system (Motion Analysis, Inc., Santa Rosa, CA) was selected to record the three-diniensional throwing patterns of motion. Each subject was marked with retroreflective, 1-in diameter balls on all of the body's major joints since clinical evaluation of throwing mechanics required a total body analysis. The reflections of these markers were tracked individually by four electronic cameras at 200 Hz. The data from each camera was merged mathematically to accurately reconstruct the three-dimensional motion of pitching. From the basic data, various parameters could be calculated to describe this complex throwing skill. To study arm motions about the shoulder, the body markers were used to mathematically construct a system of local (segmental) three-dimensional coordinate systems, which were used to calculate the motion of the arm in anatomical reference planes (Figure 4). This three-dimensional modeling technique required estimation of two coordinate axes and a translation from surface markers to joint centers. Although this technique may have less accuracy than the method of fixing three rigid markers for each arm segment, it was necessary since the resolution of JOSPT * Volume 18 * Number 2 * August 1993 FIGURE 4. Segmental coordinate systems. the system did not allow both three markers per segment and total body analysis. Based on independent measures and theoretical calculations, the procedure enabled the angular position of a segment to be determined within an accuracy of 5". To identify proper mechanics, data from healthy, successful adult pitchers were summarized. Healthy pitchers were defined as those who were not currently injured or recovering from an injury at the time of their pitching evaluation and who had not had surgery for at least 12 months. Pitchers who had undergone surgery were not considered healthy if they felt that they had not returned to " 100% form." Successful adult pitchers were defined as those who were currently competing at the college or professional level and whose average fastball analvzed during testing was at least 83 mph. Because of the large number of pitchers seen, we were able to include only the highest quality pitchers in the data base and still have a significantly large sample number. Twenty-nine pitchers met the criteria and are referred to as the "elite" pitchers. These subjects had an average height of 1.88 m (SD = 0.05 m) and an average weight of 860 N (SD = 82 N). The average radar gun speed was 38 m/sec (85 mph) with a standard deviation of 1 m/sec (2 mph). Using a sequential moving average method, parameter estimates Msd shoulder seemed to stabilize at approximately three pitches per subject. The three fastest thrown pitches that hit within the strike zone ribbon were therefore digitized and averaged for each pitcher. Feltner and Dapena (7) used only one trial per subject, since they found "little variability among the fastball pitches of any given player." Although Pappas et at analyzed 10 pitches per subject, they concluded that "an individual pitcher is reniarkablv consistent in his delivery." ARM MOTIONS ABOUT THE SHOULDER The primary movements about the shoulder during pitching are shoulder abduction, horizontal adduction, and external/internal rotation. In order to assemble average data for the 29 elite pitchers, the data for all pitchers were timematched at the time of foot contact and at the time of ball release. The time interval from foot contact to release averaged 0.1 45 seconds, with a standard deviation of 0.0 15 seconds. Figure 5 illustrates the angular coordinate systems used to define the arm motions about the shoulder. Figure 6 depicts shoulder abduction and horizontal adduction/abduction during the pitching motion. Each bold line on Figures 6-8 represents the mean value for the elite pitchers and the corresponding light, dashed lines show the standard deviation. The time of foot contact (FC), maxi-

moves backward in a horizontally abducted direction in response to rapid internal rotation of the humerus about the shoulder. At release, the arm is positioned at 0" of horizontal abduction. After the ball is released, the arm continues to move in a horizontally adducted pattern. IAI External/lnternal Rotation FIGURE 5. Angular coordinate systems. mum external rotation (MER), ball release (REL), and maximum internal rotation (MIR) are provided on each graph. The arm cocking (Cocking), arm acceleration (Acc), arm deceleration (Dec), and follow-through (F-T) phases are labeled on each graph as well. Shoulder Abduction In most throwing activities, the upper arm is placed in an abducted position about the shoulder. Figure 5a depicts the angular reference system used to define shoulder abduction; Figure 6 illustrates the average angular displacement pattern for the elite pitchers while throwing fastballs. The pattern of shoulder abduction displacement is fairly constant from foot contact to release. The arm tends to remain in an abducted position of about 100" until just before release, when the arm is slightly adducted to a position of approximately 95 ". After release, the arm rapidly abducts about the shoulder. (Note the sharp upward inflection of the graph in the arm deceleration phase.) Atwater (1) illustrated that in most throwing and striking skills, the shoulder abduction angle remains fairly constant. Higher and lower ball release points are achieved by tilting the trunk and not increasing or decreasing the shoulder abduction angle with respect to the trunk. These findings indicate that the 90-1 10" abducted position must be a very strong, dvnamic position for the arm and shoulder. Angular deviations greater than 10" outside this range (<SO0 or >I 20") during the foot contact to ball release period of throwing might suggest abnormal positioning of the arm. Horizontal Abduction/Adduction Figure 5b depicts the coordinate system used to define horizontal abduction/adduction of the arm. Figure 6 shows the average horizontal abduction/adduction pattern for the highly skilled pitchers. Initially, the arm is placed in a horizontally abducted position of 50". During trunk rotation, the arm moves forward with respect to the trunk to a position of 14". At this point, just before the ball is released. the arm FIGURE 6. Shoulder abduction and horizontal adduction. FC = time of foot contact, MER = maximum external rotation, RE1 = ball release, MIR = maximum internal rotation, Cocking = arm cocking, Acc = arm acceleration, Dec = arm deceleration, F-T = follow-through. There is no doubt that one of the most dvnamic movements in the human body is external and internal rotation of the arm about the shoulder in throwing. The reference system for defining this movement is illustrated in Figure 5c; the average movement pattern for the 29 subjects is depicted in Figure 7. The previous discussion has shown that at foot contact, the arm is elevated in a semicocked position. At the time of foot contact, the arm is undergoing external rotation, having reached a position of approximately 55". After foot contact of the striding leg, the arm continues external rotation, reaching a maximum average position of 1 78". (The exact contribution to total external rotation by each of the shoulder components of glenohumeral, scapulothoracic, and trunk hyperextension was not quantified in this study.) This FC MER RE1 MIR FIGURE 7. External/internal rotation. FC = time of foot contact, MER = maximum external rotation, REL = ball release, MIR = maximum internal rotation, Cocking = arm cocking, Acc = arm acceleration, Dec = arm deceleration, F-T = follow-through. Volume 18 Number 2 August 1993 *JOSPT

12.5" of external rotation (.?%I 78") occurs during the first 80% of the foot contact to release phase (armcocking). During the last 20% of this interval (approximately 0.029 seconds), the arm rotates internally from 178 to 105 " at release. This is one of the fastest human movements observed in any physical skill, reaching a mean maximum angular velocitv of 6.940 "/set (f 1080 "/set) in this study (Figure 8). At release, the arm is still externally rotated at a position of 1 10". When the throwing motion is viewed from the side, the arm appears to have reached the 90" position at release because it has an absolute vertical orientation. Actually, the arm is 10-15" behind the trunk line at this point because the trunk is flexed forward at release. After the ball is released, the shoulder continues this rapid internal rotation as the forearm and hand undergo pronation to follow this unique and dynamic shoulder movement. SUMMARY OF SHOULDER KINEMATICS Shoulder Abduction The arm position relative to the trunk remains fairly constant-be- FIGURE 8. External/internal angular velocity. FC = time oi foot contact, MER = maximum external rotation, REL = ball release, MIR = maximum internal rotation, Cocking = arm cocking, ACC = arm acceleration, Dec = arm deceleration, F-T = follow-through. tween 90 and 1 10" during the footcontact-release period. Immediately after release, the shoulder abducts. Horizontal Adduction The arm rotates relative to the trunk from an abducted position of SO" to 14" of adduction during the initial 80% of the foot-contact-release phase. During the final period of this phase, when internal rotation about the shoulder is performed, the arm seems to horizontally rotate backward (horizontally abduct) to the 0" position. During followthrough, the arm continues in horizontal adduction. In the arm deceleration phase, shoulder internal rotation angular velocity decreases to zero from its maximum value observed near the time of ball release. External/lnternal Rotation The shoulder externally rotates to 175" during the initial 80% of the foot-contact-release period and, subsequently, undergoes rapid internal rotation, continuing through release and arm deceleration. The arm reaches an externally rotated position of 100-1 10" at release. CONCLUSION Based upon the qualitative and quantitative information presented on pitching, it can be concluded that throwing a baseball at maximum velocity is one of the most highly dynamic skills in all sports. These results also illustrate the dynamic capabilities of the shoulder complex. Because of the highly mobile nature of this joint, great care and physical preparation of the shoulder are required before and during participation in throwing events that require maximum velocity. To improve upon present physical preparation and rehabilitation of throwers, greater understanding of the specific "loads" and muscular responses that occur during these types of highly dynamic activities will be required. IOSPT ACKNOWLEDGMENTS The biomechanics of baseball pitching has been a major study at the American Sports Medicine Institute for three years, and a number of professionals have made substantial contributions for this project. The authors would like to acknowledge the scientific contributions of Bill McLeod, PhD; Deric Wisleder, MD; Rafael Escamilla, MS. CSCS; Clifford Schob, MD; and Andy DeMonia; the medical contributions of Seth Kupferman, MD; and the statistical help of Tom Woolley, PhD. We would also like to thank Bill Thurston of Amherst College and Dewey Robinson of the Chicago White Sox for their experience and contributions in defining proper mechanics, and Mike Keirns, PT, ATC, CSCS, for his help in the classification of pathology. REFERENCES Atwater AE: Biomechanics of overarm throwing movements and of throwing injuries. Exerc Sport Sci Rev 71:43-85, 1980 DiGiovine NM, lobe FW, Pink M, Perry I: Electromyography of upper extremity in pitching. 1 Shoulder Elbow Surg 1: 15-25, 1992 Dillman CI: Proper mechanics of pitching. Sports Med Update, HealthSouth Sports Medicine Network Spring: 15-18, 1990 Dillman CI, Fleisig GS, Werner SL, Andrews IR: Biomechanics of the shoulder in sports: Throwing activities. Postgraduate Studies in Sports Physical JOSPT Volume 18 Number 2 August 1993

Therapy, Berryville, VA: Forum Medicum Inc., 1991 5. Elliott B, Grove /R, Cison 6, Thurston 6: A three-dimensional cinematographic analysis of the fastball and curveball pitches in baseball. Int I Sport Biomech 2:20-28, 1986 6. Feltner ME: Three-dimensional interactions in a two-segment kinetic chain. Part 11: Application to the throwing arm in baseball pitching. Int I Sport Biomech 5:420-450, 1989 7. Feltner M, Dapena I: Dynamics of the shoulder and elbow joints of the throwing arm during the baseball pitch. Int I Sport Biornech 2:235-259, 1986 8. Feltner ME, Dapena I: Three-dimensional interactions in a two-segment kinetic chain. Part I: General model. Int Sport Biomech 5:403-4 19, 1989 9. Fleisig CS, Dillman CI, Andrews /R: Proper mechanics for baseball pitching. Clin Sports Med 1: 15 1-170, 1989 10. Fleisig CS, Dillman C/, Andrews /R: A biornechanical description of the shoulder joint during pitching. Sports Med Update FalllWinter: 10-15, 1 99 1 1 1. Cainor Bl, Piotrowski C, Puhl 1, Allen WC, Hagen R: The throw: Biomechanics and acute injury. Am / Sports Med 8(2):114-118, 1980 12. lobe FW, Moynes DR, Tibone /E, Perry 1: An EMC analysis of the shoulder in pitching: A second report. Am I Sports Med 1 2(3):2 18-220, 1984 13. lobe FW, Tibone /E, Perry 1, Moynes D: An EMC analysis of the shoulder in throwing and pitching: A preliminary report. Am 1 Sports Med 11(1):3-5, 1983 14. Pappas AM, Azwacki RM, Sullivan TI: Biomechanics of baseball pitching. Am 1 Sports Med 13(4):2 16-222, 1985 15. Sisto Dl, lobe FW, Moynes DR, Antonelli Dl: An electromyographic analysis of the elbow in pitching. Am / Sports Med 15(3):260-263, 1987 16. Tullos HS, KinglW: Throwing mechanics in sports. Orthop Clin North Am 4(3):709-720, 1973 17. Wisleder D, Fleisig CS, Dillman C/, Schob CI, Andrews /R: Biornechanics-Development of a biomechanical analysis of throwing with clinical applications for pitchers. Sports Med Update 4(2):28-3 1, 1989 Volume 18 Number 2 Aupst 1993 *JOSPT