Original Research
Correlation between muscle strength and throwing speed in adolescent baseball players Anita S. Clements, Karen A. Ginn and Elizabeth Henley
Anita S. Clements BAppSc (Phty) (Hons), Physiotherapist, Mater Misericordiae Hospital, Sydney, Australia Karen A. Ginn MHPEd BSc GDManipTher, Senior Lecturer (Anatomy), Faculty of Health Sciences, The University of Sydney, Sydney, Australia Elizabeth Henley MClSc BSc BPT, Senior Lecturer (Physiotherapy), Faculty of Health Sciences, The University of Sydney, Sydney, Australia Correspondence to: Elizabeth Henley, School of Physiotherapy, Faculty of Health Sciences, University of Sydney, PO Box 170, Lidcombe, NSW 1825, Australia. Tel: 61 2 9351 9268; Fax: 61 2 9351 9601; E-mail: E.
[email protected]. edu.au
One of the aims of baseball training is to increase the throwing speed of players. It has been assumed previously that increasing muscle strength will increase throwing speed. Studies involving adult baseball players have demonstrated that the strength of shoulder adductors, wrist extensors and elbow extensors have predicted throwing speed. Objective: The aim of this study was to determine if upper limb muscle strength correlates with throwing speed in adolescent baseball players. Setting: This study was carried out in the Faculty of Health Sciences, University of Sydney ± laboratory and ®eld-testing. Participants: Eighteen elite adolescent baseball players (aged 13±16 years). Outcome measures: Participants underwent throwing speed tests and shoulder and elbow isometric, concentric and eccentric strength tests on an isokinetic dynamometer. Results: Stepwise regression analysis indicated that 71% of the variation in throwing speed in these adolescent baseball players was explained by isometric internal rotation and concentric elbow extension torque-to-body weight in approximately equal proportions. Conclusions: These results demonstrate that both elbow extensor and shoulder internal rotation strength has a large in¯uence on throwing speed. Therefore, preferential strength training for elbow extension and shoulder internal rotation muscle groups in adolescents, without neglecting the antagonistic muscle groups, is recommended. Physiotherapists involved with adolescent baseball athletes need to be aware of the potential effect that strength increases and muscle imbalances in these muscle groups may have in maturing the musculoskeletal system. c 2000 Harcourt Publishers Ltd *
Introduction One aim of training baseball players is to increase the speed of the ball when it is pitched or thrown. As throwing requires speci®c muscle actions, alterations in muscle strength are considered necessary for optimal performance when pitching and throwing. It has been suggested, therefore, that increasing muscle strength in the upper limb will increase throwing speed (Bartlett et al. 1989; Pedegana et al. 1982; Toyoshima et al. 1974). Concentric shoulder adduction, wrist extension and elbow extension peak torque
c 2001 Harcourt Publishers Ltd *
have been shown to signi®cantly predict throwing speed in adult baseball players (Bartlett et al. 1989; Pedegana et al. 1982). However, Toyoshima and colleagues (1974) found that only 53.1% of throwing speed can be attributed to upper limb involvement. No studies to date have examined the correlation between muscle strength and throwing speed in adolescent baseball players. The purpose of this study was to examine the relationship between upper limb muscle strength and throwing speed in adolescent baseball players. In the current study all three modes of muscle contraction were tested, since
Physical Therapy In Sport
(2001) 2, 123±131
doi : 10.1054/ptsp.2000.0025, available online at http://www.idealibrary.com on
123
Physical Therapy in Sport
upper limb muscles work isometrically, concentrically and eccentrically during pitching and throwing to produce upper limb joint movement. The muscles work to accelerate and decelerate movements and to stabilise the upper limb joints, especially the shoulder joint (Jobe et al. 1983; Jobe et al. 1984).
Methods Eighteen elite, male, adolescent baseball players (age 13±16 years) in a selectively chosen team volunteered to participate in the study. These adolescents had played baseball (including teeball) for an average of 6.6 + 1.9 years. They had been identi®ed ( from district league teams) as displaying special talent and as such, had been selected for special coaching in pitching and throwing mechanics as well as baseball game strategies. All players were in contention for representative positions in State and National teams, with one player being a current member of the `under 18' National team. At the time of testing, the players were training at least twice weekly and playing each weekend. Informed consent was obtained from each subject and his parent. Data were recorded of each subject's age, height and weight, years of experience playing baseball and classi®cation of pitcher or ®elder. Muscle strength and throwing speed tests were performed on the throwing arm over two days. Throwing tests were carried out on one day, with isometric, concentric and
eccentric strength tests measured on a Biodex isokinetic dynamometer (Biodex Medical Systems Inc, NY) approximately two weeks later.
Throwing speed Average speed of throwing was calculated using videotape analysis. Subjects threw a baseball to a target 20.3 m away. The target size was 45 cm by 70 cm, the average size of a strike zone. The centre of the target was 70 cm from the ground. The video camera was set up 7.3 m from the line of the pitch and 2.5 m from the subject towards the target (Fig. 1). The video camera was focussed to allow detection of the moment of release of the ball and the ®rst ®ve frames of the throw, after release, to be viewed on the screen. The camera speed was set at 1/250 of a second. Two one metre rules were set up vertically and horizontally in front of the camera in the line of the pitch to allow for calibration of the displacement of the ball. Subjects were given as many warm-up throws as they felt necessary. The throw comprised a step from behind the throwing plate so that the point of release of the ball to the target also commenced at the pitching mound. After one minute's rest, the throws were video recorded. Standardized instructions to throw the ball `as fast as you can to hit the target' were given to each subject. Subjects threw the baseball three times. Only throws that target 0.7 m
subject 20.3 m
2.5 m 7.3 m
camera Fig. 1 Schematic diagram of the throwing speed set-up.
124
Physical Therapy in Sport
(2001) 2, 123±131
c 2001 Harcourt Publishers Ltd *
Correlation between muscle strength and throwing speed
hit the target were considered for analysis. The most accurate throw, de®ned as the one hitting closest to the centre of the target, was analysed. The spatial coordinates of the ball were realised using a manual digitizing tablet and a video-play-back unit in a con®guration similar to that described by Crosbie and colleagues (1994). Images from the video unit (National AG6200 freeze-frame recorder) and from an overhead camera which captured the active area of a digitizing tablet (SummaSketch II Professional Plus, Summagraphics Inc.) were mixed (National WJ-SIN) and displayed on a monitor. The video ®lm was played back at 50 frames per second. The digitizing tablet had a theoretical resolution of less than 0.1 mm. The coordinates realised were stored using commercial software (SigmaScan, Jandel Scienti®c Inc.) and subsequently analysed. Calibration was set using the length of the rule as 100 cm in both the horizontal and vertical planes. Five frames of the ball immediately following ball release were digitized frame-byframe. The centre of the ball was readily identi®ed from the video image and the speed of the ball at frames 2, 3 and 4 was calculated ( frames 1 and 5 being eliminated from analysis to reduce parallax error). The speeds at each frame were then averaged to obtain the average speed of the ball during the most accurate throw.
Muscle strength measurements Concentric, eccentric and isometric strength were tested with the Biodex Multi-Joint Testing and Exercising Dynamometer (Biodex Medical Systems Inc, NY) which was calibrated prior to testing. Good reliability of the Biodex in testing shoulder internal and external rotation for concentric (ICC 0.60±0.95) and eccentric (ICC 0.77±0.83) contractions at 1208/s and isometric (ICC 0.81±0.93) contractions has been reported (Frisiello et al. 1994; Malerba et al. 1993). The subject was ®rmly strapped into the chair and standardized instructions were given to the subject. Testing was randomized according to joint, mode of contraction and movement. The subject was placed in a standardized position for the movement being tested. The chosen test positions allowed the activated muscle to be placed at approximately
c 2001 Harcourt Publishers Ltd *
its mid-length. Shoulder internal and external rotations were tested with the shoulder abducted 908, elbow ¯exed 908 and forearm in neutral. Isometric testing was performed with the forearm parallel to the ground (Fig. 2). Elbow ¯exion and extension were tested with the shoulder ¯exed 908 in the scapular plane and the forearm in neutral. Isometric testing was performed with the elbow ¯exed at 908 (Fig. 3). Forearm pronation and supination were tested with the subject's arm resting by his side and elbow ¯exed to 908 (Fig. 4). Isometric testing was not performed for forearm pronation and supination, as the majority of the subjects were unable to produce enough torque to register on the Biodex. For concentric and eccentric contractions, subjects were given three sub-maximal warmup attempts, followed by three consecutive maximal contractions at 1208/s. It is acknowledged that isokinetic dynamometry is not able to test muscle strength anywhere near the large rotational velocities obtained by the arm during throwing, ranging from 6000± 70008/s (Dillman et al. 1993); therefore, slower velocities must be used. It also must be acknowledged that the performance of the throwing task involves isoinertial movement rather than isokinetic movement. However, recognizing these limitations, isokinetic dynamometry is a recognized tool for measuring through-range torques at varied velocities. Cook et al. (1987) stated that 3008/s was too fast to accurately record the peak torque produced during contractions as there appeared to be a subconscious attempt to decelerate the arm at the end of range. Therefore, in this study a velocity of 1208/s was chosen as, although not speci®c to normal throwing velocity, it was found to be a reliable velocity at which to measure isokinetic muscle strength (Frisiello et al. 1994; Malerba et al. 1993). All test movements were blocked 158 before the end of range. The largest peak torque value was recorded. Isometric testing involved one sub-maximal warm-up contraction, followed by three maximal contractions each held for 5 seconds with 15 seconds rest between contractions. The torque was calculated as the average torque over the 5 seconds. The largest torque produced was recorded.
Physical Therapy In Sport
(2001) 2, 123±131
doi : 10.1054/ptsp.2000.0025, available online at http://www.idealibrary.com on
125
Physical Therapy in Sport
Fig. 2 Position for shoulder strength testing.
Peak torque-to-body weight ratios for each test were calculated by the Biodex and recorded. Subjects had 2 minutes rest between each mode of contraction. The test was terminated if pain was produced and the measurement was excluded from analysis.
Statistical analysis Means and standard deviations were calculated for peak torque-to-body weight ratios measured using a Biodex isokinetic dynamometer and for throwing speed measurements. Two-tailed Pearson correlation coef®cients were used to determine correlations between throwing speed and peak torque-to-body weight ratios. This was followed by a stepwise regression of throwing speed against the strength measurements that correlated highly with
126
Physical Therapy in Sport
(2001) 2, 123±131
throwing speed (Tabachnick & Fidell 1989). A signi®cance level of P 5 0.05 was chosen.
Results The elite male adolescent baseball players in this study averaged 14.22 (SD 0.55) years of age, 169.2 cm (SD 10.58) in height and 58.04 kg (SD 12.69) in weight. Eleven were pitchers and seven were ®elders. The group had played baseball (including tee-ball) for an average of 6.6 + 1.9 years. The average throwing speed for the baseball players was 30.97 + 3.7 m/s. Peak torque-to-body weight ratios for concentric and eccentric muscle actions and average torque-to-body weight ratios for isometric contractions for the whole baseball group are given in Table 1. Pitchers tended to generate greater peak torque-to-body-weight ratios than ®elders for all movements tested in
c 2001 Harcourt Publishers Ltd *
Correlation between muscle strength and throwing speed
Fig. 3 Position for elbow strength testing.
all modes of contraction, however no tests of signi®cance were performed due to the small sample sizes involved. No tests were excluded from the analysis due to pain. However, some of the tests have data missing due to equipment failing to record the data on occasions and therefore, not all of the 18 baseball players are represented in all tests. The correlations between peak torque-tobody weight ratios and throwing speed are given in Table 2. Within this elite, adolescent baseball group isometric shoulder internal rotation and concentric elbow extension strength demonstrated signi®cant relationships with throwing speed (r 0.64 and r 0.58 respectively). As the isometric shoulder internal rotation torque-to-body weight ratio and concentric elbow extension torque-to-body weight ratio were most signi®cantly correlated with
c 2001 Harcourt Publishers Ltd *
throwing speed, a post-hoc stepwise regression was performed on throwing speed. These two factors and each of the factors with r values greater than 0.4 were used, so there were three factors in the regression analysis. When elbow extension strength was part of the analysis, 71% of the variation in throwing speed was explained by isometric shoulder internal rotation torque-to-body weight and concentric elbow extension torque-to-body weight in approximately equal proportions (F(2,11) 13.42, P 5 0.01) (Table 3). The residual unexplained variation in throwing speed was not signi®cant.
Discussion Isometric shoulder internal rotation and concentric elbow extension peak torque-tobody-weight ratios demonstrated signi®cant
Physical Therapy In Sport
(2001) 2, 123±131
doi : 10.1054/ptsp.2000.0025, available online at http://www.idealibrary.com on
127
Physical Therapy in Sport
Fig. 4 Position for forearm strength testing.
correlation with throwing speed in the elite early adolescent baseball players tested in this study. Concentric shoulder internal rotation peak torque-to-body weight also correlated moderately, but not signi®cantly, with throwing speed. In this study, isometric shoulder internal rotation and concentric elbow extension peak torque-to-body weight ratios together accounted for 71% of the variation in throwing speed. A previous study examining the relationship between peak torque in various upper limb muscle groups in adult professional baseball players found that concentric elbow extension together with concentric wrist extension peak torque accounted for the largest proportion of the variation (65%) in throwing speed (Pedegana et al. 1982). However, as they did not include shoulder internal rotator strength in their regression analyses and the
128
Physical Therapy in Sport
(2001) 2, 123±131
present study did not include wrist extensor strength, direct comparisons are dif®cult. In agreement with the present study, concentric elbow extension peak torque has been found previously to signi®cantly predict throwing speed in adult baseball pitchers, explaining 27% of the variation in throwing speed (Pedegana et al. 1982). Additionally, Cohen et al. (1994) found a relationship between increased tennis serving speed and concentric elbow extension peak torque in elite adult tennis players. In contrast to the present study, concentric shoulder internal rotation peak torque has been reported previously to only contribute minimally (6±10%) to throwing speed in baseball players (Bartlett et al. 1989; Pedegana et al. 1982). Bartlett et al. (1989), however, reported that concentric shoulder adduction peak torque accounted for 55% of
c 2001 Harcourt Publishers Ltd *
Correlation between muscle strength and throwing speed
Table 1 Torque-to-body weight ratios (Nm/kg) of the entire group of baseball players measured using an isokinetic dynamometer* Movement
Mean
SD
Shoulder internal rotation Concentric (n 17) Eccentric (n 17) Isometric (n 17)
62.48 81.24 55.66
6.91 9.65 9.16
Shoulder external rotation Concentric (n 17) Eccentric (n 17) Isometric (n 17)
45.35 60.29 50.64
7.87 11.95 9.42
Elbow ¯exion Concentric (n 17) Eccentric (n 18) Isometric (n 18)
50.37 89.42 75.92
12.34 14.17 17.74
Elbow extension Concentric (n 17) Eccentric (n 16) Isometric (n 17)
56.28 96.23 67.01
9.07 15.76 13.24
Forearm pronation Concentric (n 17) Eccentric (n 16)
18.85 23.27
5.03 7.11
Forearm supination Concentric (n 17) Eccentric (n 17)
15.34 19.00
5.77 7.82
*Concentric and eccentric data represent peak torque values. Isometric data represent average torque values.
Table 2 Correlation between torque-to-body weight ratios of upper limb muscle groups and throwing speed Movement
r
P
Shoulder internal rotation Concentric (n 17) Eccentric (n 17) Isometric (n 17)
0.4336 0.2941 0.6369
0.082 0.236 0.006**
Shoulder external rotation Concentric (n 17) Eccentric (n 17) Isometric (n 17)
0.3363 0.3426 0.4228
0.187 0.178 0.091
Elbow ¯exion Concentric (n 17) Eccentric (n 18) Isometric (n 18)
0.2094 0.2941 0.2533
0.420 0.236 0.311
Elbow extension Concentric (n 17) Eccentric (n 16) Isometric (n 17)
0.5841 0.1797 0.2519
0.014* 0.506 0.313
Forearm pronation Concentric (n 17) Eccentric (n 16)
0.3505 ÿ0.1621
0.168 0.549
Forearm supination Concentric (n 17) Eccentric (n 17)
0.1143 ÿ0.0019
0.662 0.994
*P 5 0.05, **P 5 0.01.
c 2001 Harcourt Publishers Ltd *
Table 3 Stepwise regression of throwing speed against torque-to-body weight ratios Variable entered
R2
B
b
P
Isometric shoulder internal rotation Concentric elbow extension Constant ÿ0.6192
0.46
0.2793
0.5564
0.004**
0.71
0.2892
0.5158
0.007** 0.92
**P 5 0.01.
throwing speed. Although shoulder adduction torque was not tested in this study, one would expect there to be a relationship between shoulder internal rotation and shoulder adduction torque in their ability to predict throwing speed. Many of the internal rotators, such as pectoralis major, latissimus dorsi and teres major, are also shoulder adductors. By using only Pearson correlation coef®cients to assess muscle torque contribution to throwing speed, Bartlett et al. (1989) did not account for the overlapping of the contribution of different muscle groups. Without performing a stepwise regression analysis the individual contribution of each muscle group could not be determined and the contribution of the shoulder adductors to throwing speed in their study is likely to have been overestimated. The results of this study would support a signi®cant role for the shoulder internal rotator and elbow extensor muscles in producing the force on the ball which determines its speed during the acceleration phase of throwing in adolescent baseball players. A baseball training program for baseball which included exercises designed to preferentially increase the isometric shoulder internal rotation strength and concentric elbow extension strength would seem appropriate. Microtrauma commonly occurs in the upper limb decelerators (shoulder external rotators and elbow ¯exors) of baseball players (Blackburn 1986). Therefore, as part of a strength training program designed to increase throwing speed involving shoulder internal rotator and elbow extensor muscles, shoulder external rotator and elbow ¯exor muscles should also be strengthened eccentrically. This will enable the thrower to control deceleration of the upper limb during the follow through phase of throwing and to maintain normal
Physical Therapy In Sport
(2001) 2, 123±131
doi : 10.1054/ptsp.2000.0025, available online at http://www.idealibrary.com on
129
Physical Therapy in Sport
muscle strength balance around the upper limb joints, thus minimizing the risk of joint injury (Fleisig et al. 1995). Throwing speed cannot be entirely dependent on upper limb muscle strength. A recent study found no differences in isometric strength between the early adolescent baseball players in this study and matched control subjects for shoulder internal rotation and elbow extension (Clements 1996), although the adolescent baseball subjects threw the ball at 50% faster speeds than the control subjects. Coordination among the upper limb muscles and with the trunk and lower limb is essential in order to throw properly and ef®ciently. Dillman and colleagues (1993) state that a short delay between the onset of elbow extension and shoulder internal rotation during acceleration is crucial. When the elbow is extended, the internal rotation force required to move the arm is less. Therefore, the torque generated at the shoulder is able to cause greater acceleration of the arm and thus the ball (Dillman et al. 1993). To increase throwing speed without exposing the baseball player to increased risk of injury, players must learn to coordinate muscle activity within the throwing arm. They must also coordinate the motion of the throwing arm with the rest of the body (Toyoshima et al. 1974), as has been learned by the group of elite adolescent baseball players in this study. This is particularly important in adolescent baseball players who, because of their immature musculoskeletal systems, are at risk of not only joint and muscle injuries but also injury to their growing bones, particularly avulsion fractures at their epiphysial plates (Ireland & Hutchinson 1995; Jobe et al. 1993; Pappas 1982).
Conclusion The results of this study on early adolescent, elite baseball players reinforce the data from studies on adults that elbow extensor strength has a large in¯uence on throwing speed but imply a much greater in¯uence from shoulder internal rotators than adult studies indicate. Preferential strength training for elbow extension and shoulder internal rotation muscle groups in baseball training programs for adolescents is recommended. However, one
130
Physical Therapy in Sport
(2001) 2, 123±131
must not forget the contribution of antagonist muscle groups in decelerating the limb during throwing, as well as the signi®cant role of coordination between the upper limb with the trunk and lower limbs in throwing accurately and fast. Hence, when undergoing strength training of elbow extensors and shoulder internal rotators, care should be taken to minimize or eliminate excessive strength increases and strength imbalances in adolescent athletes which may increase predisposition to injury in the developing musculoskeletal system.
Acknowledgement The authors wish to thank the baseball players and coaches of the Touring Indians Baseball Team, particularly Mr Fred Rouse, who took part in this study. This work was conducted at the Faculty of Health Sciences, University of Sydney, Australia. It was partially supported by a Faculty Seed Grant from the Faculty of Health Sciences.
References Bartlett L R, Storey M D, Simons B D 1989 Measurement of upper extremity torque and its relationship to throwing speed in the competitive athlete. American Journal of Sports Medicine 17: 89±91 Blackburn T A 1986 Throwing injuries to the shoulder. In: Donatelli R (ed). Physical therapy of the shoulder, 2nd edn. Churchill Livingstone, New York, 239±270 Clements A S 1996 Upper limb musculoskeletal function and throwing performance in adolescent baseball players. Honours thesis, The University of Sydney, School of Physiotherapy Cohen D B, Mont M A, Campbell K R, Vogelstein B N, Loiwy J W 1994 Upper extremity physical factors affecting tennis serve velocity. American Journal of Sports Medicine 22: 746±750 Cook E, Gray V, Savinar-Nogue E, Medieros J 1987 Shoulder antagonist strength ratios: a comparison between college-level baseball pitchers and new pitchers. Journal of Orthopaedic & Sports Physical Therapy 8: 451±461 Crosbie J, Flynn W, Rutter L 1994 Effect of side load carriage on the kinematics of gait. Gait and Posture 2: 103±108 Dillman C J, Fleisig G S, Andrews J R 1993 Biomechanics of pitching with emphasis upon shoulder kinematics. Journal of Orthopaedic & Sports Physical Therapy 18: 402±408 Fleisig G S, Andrews J R, Dillman C J, Escamilla R F 1995 Kinetics of baseball pitching with implications about injury mechanisms. American Journal of Sports Medicine 23: 233±239
c 2001 Harcourt Publishers Ltd *
Correlation between muscle strength and throwing speed
Frisiello S, Gazaille A, O'Halloran J, Palmer M, Waugh D 1994 Test-reliability of eccentric peak torque values for shoulder medial and lateral rotation using the Biodex isokinetic dynamometer. Journal of Orthopedic & Sports Physical Therapy 19: 341±344 Ireland M L, Hutchinson M R 1995 Upper extremity injuries in young athletes. Clinics in Sports Medicine 14: 533±569 Jobe F W, Moynes D R, Tibone J E, Perry J 1984 An EMG analysis of the shoulder in pitching: a second report. American Journal of Sports Medicine 12: 218±220 Jobe F W, Tibone J E, Perry J, Moynes D 1983 An EMG analysis of the shoulder in throwing and pitching: a preliminary report. American Journal of Sports Medicine 11: 3±5 Malerba J, Adam M, Harris B, Krebs D 1993 Reliability of dynamic and isometric testing of shoulder external and
c 2001 Harcourt Publishers Ltd *
internal rotators. Journal of Orthopaedic & Sports Physical Therapy 18: 543±552 Pappas A M 1982 Elbow problems associated with baseball during childhood and adolescence. Clinical Orthopaedics and Related Research 164: 30±41 Pedegana L R, Elsner R C, Roberts D, Lang J, Farewell V 1982 The relationship of upper extremity strength to throwing speed. American Journal of Sports Medicine 10: 352±354 Tabachnick B G, Fidell L S 1989 Using multivariate statistics, 2nd edn. Harper Collins, New York, ch. 5 pp 123±191 Toyoshima S, Hoshikawa T, Miyashita M, Oguri T 1974 Contribution of the body parts to throwing performance. In: Nelson R C, Morehouse C A (eds). Biomechanics IV University Park Press, Baltimore, 169±174
Physical Therapy In Sport
(2001) 2, 123±131
doi : 10.1054/ptsp.2000.0025, available online at http://www.idealibrary.com on
131
