The ability to throw with a high degree of speed and accuracy is critical to performance in overhead throwing sports. Shoulder proprioception, that is, the ability to sense joint movement and position (17) has not only been implicated in injury risk but it has also been suggested to influence performance within throwing populations.
An enhanced ability to sense the movement and position of the body and, more specifically, the throwing arm, has long been thought to be advantageous to throwing athletes. High levels of proprioception are thought to allow for more accurate movements and facilitate real-time adjustments in response to movement error leading to more accurate throwing. Conversely, poor proprioception has been implicated in a reduced ability to accurately position the body segments in space, leading not only to suboptimal performance but also to increased injury risk via abnormal loading patterns (28).
Interestingly however, despite the widely held nature of this view, the evidence for such a relationship remains poor. Previous investigations provide only moderate levels of indirect evidence to support this link. Although throwing accuracy and shoulder proprioception have been shown to be reduced concomitantly after cryotherapy treatment (31), general exercise (4,16), and throwing specific exercise (28), no study to date has sought to demonstrate a specific relationship between shoulder proprioception and throwing performance in terms of speed or accuracy.
Additionally, shoulder proprioception has also been indirectly linked with throwing speed. Shoulder proprioception and humeral torsion, that is, the amount of twist about the long axis of the humerus (15) is significantly correlated in throwing athletes (33). Further, a link between humeral torsion and throwing speed has been proposed, with increased shoulder external rotational range (2,20) suggested to lead to greater internal rotation velocity and, consequently, ball speed (32). Given that humeral torsion is related to shoulder proprioception and throwing speed, one might also anticipate a relationship to exist between throwing speed and shoulder proprioception. Interestingly again however, no direct link between shoulder proprioception and throwing speed has been explored.
The establishment of a significant relationship between shoulder proprioception and throwing performance (speed or accuracy) would have significant implications for clinicians working with throwing athletes. It would provide a noninvasive screening tool that could be used to predict performance and provide stimulus for the development of novel training methods to improve performance. Specifically, clinicians could introduce exercises that target the enhancement of shoulder proprioception to improve throwing performance. Further, such a relationship would allow throwing performance to act as a proxy for shoulder proprioception during game situations, thereby providing clinicians with a monitoring tool for the prediction of injury risk.
The primary aim of this study, therefore, was to determine the relationship between the amount of active proprioception acuity in the early cocking phase of the throwing motion and throwing performance measured in terms of maximal throwing speed (MTS) and accuracy during maximal and submaximal speed throwing. We hypothesized that a significant correlation would be present between shoulder proprioception and performance in terms of both speed and accuracy.
Experimental Approach to the Problem
To determine the relationship between the shoulder proprioception acuity and throwing performance measured in terms of MTS and accuracy, a descriptive, cross-sectional laboratory study was conducted with healthy participants. Participants were assessed for shoulder proprioception and for MTS and throwing accuracy. The hypothesis was tested using bivariate correlations between the independent (proprioception) and dependent (speed and accuracy) variables to determine their relationships, if any.
Twenty-two adolescent male athletes (age, 19.6 ± 2.6 years; height, 180.5 ± 1.4 cm; weight, 75.2 ± 2.1 kg) currently representing New South Wales in the sport of Baseball volunteered to take part. Appropriate consent was obtained pursuant to law as all participants were informed of the experimental procedures involved before providing written consent to participate in the study. Parental or guardian consent was also obtained in the event the participant was younger than 18 years. The participants also completed a questionnaire regarding injury and playing history. All research was carried out in compliance with the ethical guidelines for human research laid down by the Australian National Health and Medical Research Council, and institutional board approval was provided by the University of Sydney Human Research Ethics Committee. The participants attended an indoor testing facility on a single occasion during the preseason phase of competition. The participants were instructed to be well rested, to maintain their regular diet, and to be adequately hydrated upon arrival.
Active Shoulder Proprioceptive Acuity
A measure of active proprioception was made bilaterally using the active movement extent discriminating apparatus (AMEDA) configured for shoulder testing (22). The measurement of the nondominant arm provided a control measure in throwing populations. The psychophysical method used is similar to that used by Magill and Parks (19) to test active proprioception at the elbow. The AMEDA consists of a computer-controlled stepper motor (RS Components Pty. Ltd., Silverwater, Australia) mounted on an adjustable hoist driving a geared shaft with a circular wooden disc at its end. The software driving the stepper motor can move the shaft to any 1 of 5 different positions, each approximately 1.5 cm apart. The participants were positioned with their upper arm supported on a height-adjustable padded rest so that the shoulder is at 90° of abduction. With the elbow flexed to 90° and the shoulder externally rotated to 90° (measured in situ with an inclinometer), the dorsum of the hand made contact with the disc in position 0. When the contact disc was moved anteriorly to positions 1–5, the amount of shoulder external rotation required for the subjects hand to make contact was decreased progressively from 85° to 74° (Figure 1).
The order of arm testing for active proprioception was randomized using randomization.com, and information regarding arm dominance was not obtained until all examination was complete. The participants were first instructed as to the nature of the testing and then given practice trials at the positioning task during which they were first informed of the AMEDA positions (1–5). In these practice sessions, as in the actual testing, the participants were instructed to direct their gaze straight ahead while the motor repositioned the contact plate. The participants stood with their arm supported and their elbow in full extension. Once the motor stopped moving, the participants first flexed their elbow to 90° then externally rotated at the shoulder until they made contact with the disc. The software for driving the motor included additional random forward and backward movements during the repositioning so the participant could neither determine how far the shaft had moved nor in which direction. The examiner was seated behind the participants to ensure correct movement form of each trial. During the practice session, all 5 of the test positions were sampled 3 times, and the terminal positions were sampled a further 3 times each for a total of 21 practice trials. Subsequently, the experiment proper commenced for the same arm with a randomized sequence of 50 test positions comprising 10 trials of each of the 5 test positions. During these trials, the participant was asked to estimate the test position (from 1 to 5) as soon as possible after their hand made contact with the wooden disc. The examiner entered this response data into the software driving the motor without knowledge of the actual position. The entire procedure was then repeated for the other arm. After data collection, arm dominance and other demographic information were obtained. The entire testing procedure took 30 minutes.
The participants then performed an assessment of throwing performance within an indoor laboratory. Before testing, the participants performed a general warm-up routine, initially consisting of 5–10 minutes of moderate intensity running, followed by 5–10 minutes of general stretching of the major muscle groups, before engaging in 10–15 minutes of throwing, which progressed gradually from low to high intensity.
During the testing session, the participants performed a total of 20 throws toward a specially designed target located 20 m away from the participants' back feet. The target consisted of a white circle measuring 7.0 cm in diameter, painted onto a black rubber mat that was suspended on a specifically designed metal frame (The University of Sydney, Lidcombe, Australia). The center of the target circle was 0.70 m above the ground. Using regulation size (diameter ∼6.7 cm, weight 142 g), baseballs (Rawlings Sporting Goods Company Inc., Maryville, St. Louis, MO, USA), 10 throws were performed at 100% of maximal effort, and 10 throws were performed at 80% of MTS, with a Cordless Speed Radar Gun (Jugs Corporation, Tualatin, OR, USA) positioned behind the target to ensure that the correct speed was maintained. Maximal speed throwing was selected to closely simulate in-game demands, whereas the submaximal condition was selected to observe whether or not a slower arm speed allowed for greater proprioceptive influence. Throws were performed in randomized blocks of 5 throws. A video camera (Canon MV800i; Canon Inc., Canon, Sydney, Australia, Australia) recording at 50 Hz placed directly behind the thrower (21.0 m from the target) was used to determine the point of ball contact relative to the target.
The participants were required to throw with an overhead technique and were permitted 1 stride forward with the front leg while maintaining the front foot behind the line until ball release, commonly referred to as throwing “from the set position.”
Data Processing—Proprioception Data
For each subject, receiver operator characteristic (ROC) graphs were generated for the dominant and nondominant arms. The ROC curve plots the sensitivity and 1-specificity, and the obtained area under the curve (AUC) gives the accuracy measure (27). An AUC of 1.0 would represent perfect accuracy, whereas a score of 0.5 would be because of chance alone. Macmillan and Creelman (18) define total discrimination ability as being obtained by comparing the extreme positions 1 and 5. Accordingly, it was from these positions that the AUC was calculated.
Data Processing—Throwing Accuracy Data
Video footage captured the point of contact between the ball and the target, which was subsequently converted to a still frame image. The number of pixels between the center of the ball and the target center was then determined, so that the pixel value could be converted to a distance using a calibration frame located on the target. Throwing accuracy was determined by assessment of specific parameters based on previous work (12) such as total error (E), absolute constant error (ACE), variable error (VE), vertical error (y), and horizontal error (x). Total error was calculated as the average distance between each throw and the target. The ACE was calculated as the distance between the typical throw and the target, whereas VE was calculated as the average distance between each throw and the typical throw. Horizontal and vertical errors were determined as the average distance in the horizontal and vertical directions, respectively, between each throw and the target.
The hypothesis was tested using a bivariate correlation analysis, performed on the proprioception (2 variables), accuracy (10 variables), and speed (1 variable) data. A 1-way analysis of variance (ANOVA) was performed to test for arm effects on proprioception. Because of the highly correlated nature of the dependent variables, multiple 1-way ANOVAs were performed on the accuracy data to determine speed effects with the Holm–Bonferroni adjustment applied. Paired samples t-tests were conducted to determine error direction and error type differences. The r value reported was Pearson's r. SPSS Version 20.0 was used for all analyses, and the significance was set a priori at p ≤ 0.05.
The proprioceptive acuity in the dominant and nondominant arms was similar. For throwing accuracy, total error (E), vertical error (Y), and ACE were significantly greater at 100% MTS compared with at 80% MTS. Vertical error (Y) was significantly greater than the horizontal error (X); however, no significant difference was observed between VE and ACE (Table 1).
Bivariate Correlation Analysis
Dominant and nondominant arm proprioceptions were strongly correlated; however, no significant correlation was found between proprioception and performance (speed or accuracy) for either arm at either speed (80 or 100% of MTS; Table 2). During maximal speed throwing, the total error was related to the vertical error, horizontal error, VE, and ACE. The ACE was related to vertical and horizontal error, whereas VE was related to vertical error only. Horizontal and vertical errors were related to each other. During submaximal speed throwing, total error was again related to vertical error, horizontal error, VE, and ACE. The ACE was related to horizontal error only, and the VE was related to vertical error only (Table 2). Horizontal error during submaximal speed throwing was significantly related to total error, horizontal error, and ACE during maximal speed throwing, whereas the ACE during submaximal speed throwing was related to horizontal error and ACE during maximal speed throwing (Table 2).
In this study, no significant relationship was evident between shoulder proprioception acuity and throwing performance (speed or accuracy) at either maximal or submaximal (80%) speed in a group of elite, adolescent male baseball players. Consequently, we reject the presented hypothesis.
Throwing accuracy was independent of shoulder proprioception such that high levels of throwing accuracy were achieved without a high degree of shoulder proprioceptive acuity, as measured by the AMEDA procedure. This indicates that shoulder joint proprioception was not a significant determinant of throwing accuracy, and suggests that factors beyond joint position acuity in this joint alone, contribute to one's ability to throw with accuracy during the overhead throw. A significant relationship between shoulder proprioception and throwing speed was also lacking. This highlights the need for clinicians to consider proprioception acuity throughout the entire kinetic chain when exploring determinants of performance in throwing athletes. Given the complex nature of proprioception and inherent limitations of its measurement however, the role of shoulder joint proprioception in throwing performance ought not to be categorically ruled out. Specifically, although we believe the proprioception measure implemented in this study holds great specificity for throwing athletes because it tests proprioceptive acuity in a highly functional position for throwers, and measures obtained rather than imposed proprioceptive acuity or reproduction ability, the measure is completed within a very specific range (∼80 to 90°) of humeral external rotation at a very moderate speed (∼45° per second, self-selected by the participants), which is a range much smaller, and speed much slower than typical 3-dimensional, unconstrained throwing requires. Although we believe that the evidence presented here is the strongest evidence to date regarding the relationship (or lack there of) between shoulder proprioceptive acuity and throwing performance, we also believe that further developments in the measurement methods of shoulder proprioception are indicated.
Despite this limitation however, this study provides no evidence to suggest that the measurement of active shoulder proprioception, even using this more advanced method with a higher degree of context specificity for throwing athletes, provides relevant information regarding the performance of throwing athletes of significant benefit to clinicians or coaching staff. Consequently, we believe that the role of proprioception in throwing performance is best understood in the context of the entire kinetic chain. As a result, we recommend that shoulder joint proprioception not be measured or targeted for training in isolation but only addressed in conjunction with the proprioceptive acuity of other joints within the kinetic chain. Although the measurement of proprioception within the shoulder joint alone seems to be of value from an injury risk perspective, this practice does not seem to be of value from a performance perspective.
Interestingly, shoulder proprioception was similar between the dominant and nondominant arms as indicated by the lack of a significant arm effect (Table 1), and a significant correlation between dominant and nondominant arm proprioceptions (Table 2). These findings are in agreement with those of previous investigations using the AMEDA proprioception measurement method (33). Although not significant, there was a trend toward reduced proprioception in the dominant arm compared with that in the nondominant arm, in agreement with the results of a body of work that has consistently shown decreased proprioception in the dominant arm of throwing athletes (1,3,6,24,28). Previous authors have attributed this decrement to cumulative microtrauma resulting from repeated throwing, likely from reduced function of the suprascapular nerve (5,7,11,14,21,23,25,26,34). It is likely (yet unverified) that a similar mechanism is responsible for the proprioception deficit observed in this study, yet we speculate that this did not reach significance, because the participants in this study were adolescents (age, 19.6 ± 2.6 years) and may not have experienced the same degree of throwing-induced microtrauma as did older participants.
Although not related to the primary outcome of the study, it is interesting to note that a significant speed accuracy trade-off was evident, such that total error was significantly higher during throws made at 100% MTS compared with those at 80% MTS. This resulted from significantly higher vertical error and ACE during maximal speed throwing and is in agreement with commonly held beliefs among baseball coaches that vertical accuracy is more difficult to attain as throwing speed increases. The evident speed accuracy trade-off is also in agreement with the result of work involving throwing in undergraduate students (13); cricket players of different sexes, ages, and skill levels (8,9); and European handball players at the novice (10,30) and elite levels (29). The specific biomechanical mechanisms for this phenomenon remain unclear. Of interest to note, however, is that the horizontal error during 100% MTS was significantly related to the horizontal error during 80% MTS. The same relationship was evident for ACE. That is, the participants who demonstrated large horizontal error and ACE during submaximal speed throwing also demonstrated large errors during maximal speed throwing. That is, the ability to throw with accuracy with respect to the horizontal direction seems independent of speed, whereas the attainment of vertical accuracy seems to be speed dependent. This is a stimulus for future investigations regarding the nature of error during throws made at different speeds.
Future investigations should continue to develop more sophisticated and context-relevant measures of shoulder proprioception for throwing athletes and apply them to a diverse range of throwing athletes. The protective effect of high levels of proprioception in relation to throwing related injury should also be explored.
Throwing speed and accuracy were not related to shoulder proprioceptive acuity. Consequently, clinicians are not advised to measure shoulder proprioceptive acuity in isolation as an indicator of performance, nor are they advised to develop and implement training strategies that aim to improve shoulder proprioception specifically. Rather, clinicians should understand the importance of proprioception throughout the entire kinetic chain, and implement training exercises that address the chain in its entirety rather than strategies that are joint specific. The development of more task-specific assessments of proprioceptive acuity should also be developed and validated involving movement at higher speeds to better reflect the demands of overhead throwing.
The results of this study do not constitute endorsement of the product by the authors of the National Strength and Conditioning Association. No external funding was received for this study. There are no conflicts of interest to declare.
1. Allegrucc M, Whitney SL, Lephart SM, Irrgang JJ, Fu FH. Shoulder kinesthesia in healthy unilateral athletes participating in upper extremity sports. J Orthop Sports Phys Ther 21: 220–226, 1995.
2. Atwater AE. Biomechanics of overarm throwing movements and of throwing injuries. Exerc Sport Sci Rev 7: 43–85, 1979.
3. Brindle TJ, Nyland J, Shapiro R, Caborn DNM, Stine R. Shoulder proprioception: Latent muscle reaction times. Med Sci Sports Exerc 31: 1394–1398, 1999.
4. Carpenter JE, Blasier RB, Pellizzon GG. The effects of muscle fatigue on shoulder joint position sense. Am J Sports Med 26: 262–265, 1998.
5. Cummins CA, Bowen M, Anderson K, Messer T. Suprascapular nerve entrapment at the spinoglenoid notch in a professional baseball pitcher. Am J Sports Med 27: 810–812, 1999.
6. Dover GC, Kaminski TW, Meister K, Powers ME, Horodyski M. Assessment of shoulder proprioception in the female softball athlete. Am J Sports Med 31: 431–437, 2003.
7. Ferretti A, De Carli A, Fontana M. Injury of the suprascapular nerve at the spinoglenoid notch. The natural history of infraspinatus atrophy in volleyball players. Am J Sports Med 26: 759–763, 1998.
8. Freeston J, Ferdinands R, Rooney K. Throwing velocity and accuracy in elite and sub-elite cricket players: A descriptive study. Eur J Sport Sci 7: 231–237, 2007.
9. Freeston J, Rooney K. Progressive velocity throwing training increases velocity without detriment to accuracy in sub-elite cricket players: A randomized controlled trial. Eur J Sport Sci 8: 373–378, 2008.
10. Garcia JA, Sabido R, Barbado D, Moreno FJ. Analysis of the relation between throwing speed and throwing accuracy in team-handball according to instruction. Eur J Sport Sci, 1–6, 2011.
11. Glennon TP. Isolated injury of the infraspinatus branch of the suprascapular nerve. Arch Phys Med Rehabil 73: 201–202, 1992.
12. Hancock GR, Butler MS, Fischman MG. On the problem of two-dimensional error scores: Measures and analyses of accuracy, bias and consistency. J Mot Behav 27: 241–250, 1995.
13. Indermill C, Husak WS. Relationship between speed and accuracy in an overarm throw. Percept Mot Skills 59: 219–222, 1984.
14. Jackson DL, Farrage J, Hynninen BC, Caborn DN. Suprascapular neuropathy in athletes: Case reports. Clin J Sport Med 5: 134–137, 1995.
15. Krahl VE. Humeral torsion in man. [Experimental research report]. Am J Phys Anthropol 3: 229–253, 1945.
16. Lee HM, Liau JJ, Cheng CK, Tan CM, Shih JT. Evaluation of shoulder proprioception following muscle fatigue. Clin Biomech 18: 843–847, 2003.
17. Lephart SM, Fu FH. Proprioception and Neuromuscular Control in Joint Stability. Champaign, IL: Human Kinetics, 2000.
18. Macmillan NA, Creelman CD. Detection Theory: A User's Guide. 2nd ed. Mahwah, NJ: Lawrence Erlbaum, 2005.
19. Magill RA, Parks PF. The psychophysics of kinesthesis for positioning responses: The physical stimulus–psychological response relationship. Res Q Exerc Sport 54: 346–351, 1983.
20. Matsuo T, Escamilla RF, Fleisig GS, Barrentine SW, Andrews JR. Comparison of kinematic and temporal parameters between different pitch velocity groups. J Appl Biomech 17: 1–13, 2001.
21. Montagna P, Colonna S. Suprascapular neuropathy restricted to the infraspinatus muscle in volleyball players. Acta Neurol Scand 87: 248–250, 1993.
22. Naughton J, Adams R, Maher C. Discriminating overhead points of contact after arm raising. Percept Mot Skills 95: 1187–1195, 2002.
23. Ringel SP, Treihaft M, Carry M, Fisher R, Jacobs P. Suprascapular neuropathy in pitchers. Am J Sports Med 18: 80–86, 1990.
24. Safran MR, Borsa PA, Lephart SM, Fu FH, Warner JJ. Shoulder proprioception in baseball pitchers. J Shoulder Elbow Surg 10: 438–444, 2001.
25. Sandow MJ, Ilic J. Suprascapular nerve rotator cuff compression syndrome in volleyball players. J Shoulder Elbow Surg 7: 516–521, 1998.
26. Smith AN. Suprascapular neuropathy in a collegiate pitcher. J Athl Train 30: 43–46, 1995.
27. Swets JA. Measuring the accuracy of diagnostic systems. Science 240: 1285–1293, 1988.
28. Tripp BL, Yochem EM, Uhl TL. Functional fatigue and upper extremity sensorimotor system acuity in baseball athletes. J Athl Train 42: 90–98, 2007.
29. Van Den Tillaar R, Ettema G. Instructions emphasizing velocity, accuracy, or both in performance and kinematics of overarm throwing by experienced team handball players. Percept Mot Skills 97: 731–742, 2003.
30. Van Den Tillaar R, Ettema G. A comparison between novices and experts of the velocity-accuracy trade-off in overarm throwing. Percept Mot Skills 103: 503–514, 2006.
31. Wassinger CA, Myers JB, Gatti JM, Conley KM, Lephart SM. Proprioception and throwing accuracy in the dominant shoulder after cryotherapy. J Athl Train 42: 84–89, 2007.
32. Whiteley R. Baseball throwing mechanics as they relate to pathology and performance—A review. [Review Article]. J Sports Sci Med 6: 1–20, 2007.
33. Whiteley R, Adams R, Ginn K, Nicholson L. Shoulder proprioception is associated with humeral torsion in adolescent baseball players. Phys Ther Sport 9: 177–184, 2008.
34. Witvrouw E, Cools A, Lysens R, Cambier D, Vanderstraeten G, Victor J, Sneyers C, Walravens M. Suprascapular neuropathy in volleyball players. Br J Sports Med 34: 174–180, 2000.