Understanding Shoulder Kinetics for Peak Performance (Part 2, The Importance of Muscle Function and Programme Design Considerations)

Now, what is the relevance of understanding what was discussed in Part One of this article? Why is it valuable to recognise what the glenohumeral joint has to endure during specific sporting tasks? If we were to attempt a baseball pitch, throw a ball, or crunch a tennis serve, how well prepared do we feel our structures and tissues are to execute those tasks repeatedly at maximal effort? Because, let’s be honest, everyone wants to perform these sorts of activities at 100%. No one likes having to hold back due to nagging pain and the lack of confidence that your shoulder is going to give out. On the whole, I think it is fair to say that most of us do not fully respect the forces and velocities the joint encounters. Prior to my own personal shoulder issues, I certainly did not! Many believe performing the same banded external rotations and dumbbell bench pressing and one-arm rowing the same 15kg weight is the golden ticket to robust, high-performing shoulders. This could not be further from the truth!

So, why should we focus on training the shoulder in isolation? Research by Ben Kibler and Jeff Chandler (1994) suggests that failing to train the shoulder in an isolated fashion severely limits our performance ceiling while increasing injury risk. Their work highlights how the shoulder must function as a conduit for force transfer from the lower limbs and trunk (the kinetic chain) to the upper limbs. Later, Chandler and colleagues (1998) reinforced the importance of targeting key areas prone to injury—especially those not sufficiently trained within the sport itself. This aligns with Steffan Jones’ principle: “You train off the field what you don’t cover on it.” These studies provide enough clarity and evidence to suggest that training the shoulder is worthwhile for injury prevention reasons. But what about performance enhancement? Does training the shoulder directly contribute to improved sporting performance? And if so, how? I will attempt to answer these questions now. 

Currently, research on this topic is somewhat contradictory, particularly regarding the contribution of the shoulder and arm to ball speed in cricket, baseball, javelin, and tennis. For example, Glazier and colleagues (2000) examined fast bowling in cricket and acknowledged the arm’s role in generating ball speed. However, they emphasised other factors as having a greater impact, such as optimising the kinetic chain, run-up speed, and torso rotation. This perspective is something I will revisit in my concluding thoughts. Conversely, Elliot and colleagues (1986) conducted a study on cricket fast bowling and found that 41%–50% of ball speed was attributed to arm speed, making it the single largest contributor. In a separate study on the tennis serve, Elliot and colleagues again highlighted the critical role of arm speed in generating ball velocity. Further supporting this notion, Morriss and Bartlett (1996) studied javelin throwers and found that 70% of release speed was generated within the final 0.1 seconds before release, emphasising the ability for rapid arm acceleration.

So in short, the shoulder in throwing and serving motions goes through the fastest motion in human movement during the acceleration phase and then has to decelerate up to 110% of bodyweight in distraction forces during the deceleration phase (Pappas et al., 1985; Post et al., 2015; Werner et al., 2001; & Fleisig et al., 1999). Additionally, Feltner and Dapena (1986) suggest that these high velocities and forces must often be produced in compromising, near-end-range joint positions, making the demands on the shoulder even greater. From a neurological standpoint, it is also crucial to consider that the brain will only accelerate what it knows it can safely decelerate due to its protective mechanisms. This has direct implications for programming: there’s no point in developing a powerful engine if you don’t have the brakes to stop it! Clearly, multiple factors must be considered when designing training programmes for throwers or athletes who rely heavily on high-velocity and high-force overhead movements.

Now that we understand the forces the shoulder encounters and why improving certain physical qualities may be beneficial, the next question is: how do we develop those qualities? And just as importantly, which specific qualities should we be targeting in which area? Only by answering these questions can we effectively match training methods to the desired adaptations. However, before diving deeper into programme considerations, it is crucial to first understand which muscles and muscle actions have the greatest transfer to throwing and similar sporting tasks. Stodden and colleagues (2005) found that the pectoralis major is highly active during the acceleration phase of throwing, functioning as a powerful internal rotator and horizontal adductor of the humerus. Their study also revealed that athletes who achieved higher ball velocities exhibited greater activation and utilisation of the pec major, suggesting its key role in force generation. Supporting this, Takahashi and colleagues (2003) found that greater horizontal abduction resulted in increased angular velocity during horizontal adduction. This is due to the creation of a larger moment arm, placing the pec major predominantly on a greater stretch to accelerate through within the stretch-shortening cycle (SSC). Consequently, this resulted in faster arm speeds, which we previously highlighted is a strong determinant in increasing ball velocities. 

Shifting focus back to cricket fast bowling, Portus and colleagues (2000) found that individuals with larger chest girth demonstrated higher ball velocities and emphasised the pec major’s role during the acceleration phase. This finding is particularly interesting, as one might assume that the pecs’ function at a biomechanical disadvantage in fast bowling due to the orientation of their muscle fibres and reduction in the moment arm. Since the pec major’s fibres run predominantly horizontally to their attachment on the humerus, and the bowling arm reaches near 180° of shoulder abduction, it would seem difficult for the pecs to play a major role in arm acceleration. However, when we analyse many of the sporting movements we’ve covered, we can see that lateral flexion of the spine and torso rotation reposition the shoulder into an orientation, enabling the pec major to contribute more effectively in accelerating the arm. Despite a lack of electromyography (EMG) studies evaluating muscle activity during certain overhead tasks, Gowan and colleagues (1987) and Kelly and colleagues (2002) both found that the pec major was more active during the acceleration phases in both baseball pitching and in an American football throw compared to the lats. So I think it is safe to assume that EMG results would be very similar in the images below!

So, it is reasonable to suggest that training the pec major and the movement of horizontal abduction and adduction of the humerus will have the largest transfer to many overhead sporting tasks. Ultimately, bench pressing, chest flys, and push ups, along with all their variations, are probably a good place to start. As Steffan Jones states in cricket, “bowlers have no business worrying about specific drills if they can’t incline bench 1.25 x bodyweight for 3 reps.” This underlines the importance of attaining maximal strength and peak force qualities within the muscles and patterns prior to exploring more advanced methods focusing on RFD qualities or messing around with more sport-specific exercises. The ability to express peak force is a vital ingredient to power, and consequently speed, therefore should be developed first. Moreover, Marchetti and Uchida’s (2011) EMG research found that the pec major was also more active during the pullover exercise compared to the lat, with the highest levels of activity present in the concentric phase of the movement. Therefore, the use of pullovers and all their variations could be effective in developing those peak force characteristics in slightly greater degrees of shoulder abduction for the pecs. 

Once peak force and maximal strength characteristics have been developed to a sufficient level through more traditional strength training means, we can explore more specialised and advanced methods. One example is oscillatory isometrics. Renowned strength coach Cal Dietz, through the initial work of Leonoid Matveyev, has utilised this method to improve athletes abilities to relax their muscles faster. Through Matveyev’s research and understanding of Sherrington’s Law (reciprocal inhibition), he found that it was not the ability for muscle to contract faster that made a difference, but the speed of relaxing the antagonist muscle that did. Improving this coordinative, reflexive quality within the primary movers during the acceleration phase of throwing can yield very positive results. Additionally, impulse work popularised by Brady Volmering has enabled individuals to replicate some of the high-forces and high-velocities we see during throwing motions. The ability to accept force rapidly can also be considered a form of pseudo-isometric (“sort of isometric”) developing more RFD qualities on position-specific tissues. However, I will stress that individuals will benefit far more by ensuring their maximal strength levels are sufficient prior to using such methods. Referring back to Feltner and Dapena’s (1986) research, it is vital we also consider exercises that enable us to produce large magnitudes of force in the internal rotators near MER. A cable internal rotation eccentric ensuring appropriate setup where the lever places the largest amount of external torque on the shoulder would suffice. 

Thomas Cortebeeck demonstrating an eccentric cable internal rotation.

Now that you have developed the engine, it would irresponsible to let it go out with no brakes. Developing robust external rotators is crucial to handle the large distraction forces when accelerating the arm forward, but also to ensure that ER/IR strength ratios remain within optimal ranges. To target these specific tissues we can perform standing (90° abduction) and side lying (0° abduction) external rotation variations with dumbbells utilising methods such as heavy isometrics, overcoming isometrics or supramaximal eccentrics (eccentric overload) to really develop peak force characteristics in the external rotators. Also, it should go without saying that endurance qualities must be developed if we are to take into account the large volumes and the durations that certain sporting tasks require. Endurance qualities can be developed simply by applying tension to the muscle, or muscles, for prolonged periods of time at submaximal loads. 

Standing heavy external rotation isometric hold.

I hope that up until this point, it has been clear that we’ve taken a very isolated view of the shoulder. But, as with any joint, the shoulder doesn’t work in a vacuum—it’s part of a much bigger system. That being said, this doesn’t mean we can’t build strong, resilient rotator cuffs and pecs! The key is understanding how the shoulder integrates into whole-body function while still acknowledging its specific demands. While we’ve discussed the role of the pec major, the forces acting on the glenohumeral joint, and the importance of high-velocity tolerance, we must zoom out and consider kinematic capacities like trunk rotation range of motion, kinetic parameters like lower-body strength and power (as highlighted by Kibler), and the role of motor control and technical proficiency in both performance and injury prevention. The reality is, high-performance movement isn’t just about how strong or powerful a muscle is—it’s also about how well we can coordinate, control, and sequence movement across the entire system. 

This brings us to a crucial question: Does the shoulder act as a conduit for force transfer, or is it a primary force producer itself? The answer is—it depends. Either way, what is undeniable is that the shoulder must be able to tolerate and manage high forces at high velocities. This makes it essential to train the shoulder in isolation using the methods discussed (and many more!), especially considering the research on arm speed’s contribution to ball speed in various sports. Glazier and colleagues (2000) hinted at the bigger picture, recognising that the shoulder operates within a broader kinetic system. This was further emphasised in a podcast discussion with Phil Scott (England Cricket Physical Preparation Coach), where he recalled attending a workshop with renowned shoulder expert Jo Gibson. In that session, Gibson had attendees perform a single-leg squat to demonstrate just how much lower-body function influences shoulder stability. This underscores the reality that shoulder function is never just about the shoulder—it’s interconnected with the trunk, lower limbs, and even neuromuscular control mechanisms.

One aspect we haven’t explored in detail is the role of the musculature responsible for stabilising the scapulothoracic joint through these motions—a critical component of shoulder health and performance. However, even without delving into that, one thing is clear: no matter how many other systems contribute to overhead motion, the research leaves no doubt about how much stress the shoulder must handle. The challenge isn’t just generating force—it’s about increasing the joint’s capacity to tolerate and absorb these forces repeatedly. Ultimately, while training the shoulder in isolation is only one piece of the puzzle, it’s still a piece that cannot be overlooked. 

References / Sources:

Brady Volemering - Founder of DAC Performance & Health

Calvin Dietz - Strength & Conditioning Coach & Author of ‘Triphasic Training’

Chandler, T. J., Ellenbecker, T. S., & Roetert, E. P. (1998). Sport-specific muscle strength imbalances in tennis. Strength & Conditioning Journal, 20(2), 7-10.

Elliott, B., Marsh, T., & Blanksby, B. (1986). A three-dimensional cinematographic analysis of the tennis serve. Journal of Applied Biomechanics, 2(4), 260-271.

Elliott, B.C., Foster, D.H., & Gray, S. (1986). Biomechanical and physical factors influencing fast bowling. Australian Journal of Science and Medicine in Sport, 18, 16-21.

Gowan, I. D., Jobe, F. W., Tibone, J. E., Perry, J., & Moynes, D. R. (1987). A comparative electromyographic analysis of the shoulder during pitching: professional versus amateur pitchers. The American journal of sports medicine, 15(6), 586-590.

Kelly, B. T., Backus, S. I., Warren, R. F., & Williams, R. J. (2002). Electromyographic analysis and phase definition of the overhead football throw. The American journal of sports medicine, 30(6), 837-844.

Marchetti, P. H., & Uchida, M. C. (2011). Effects of the pullover exercise on the pectoralis major and latissimus dorsi muscles as evaluated by EMG. Journal of applied biomechanics, 27(4), 380-384.

Morriss, C., & Bartlett, R. (1996). Biomechanical factors critical for performance in the men’s javelin throw. Sports Medicine, 21, 438-446.

R. Portus, M., Sinclair, P. J., Burke, S. T., Moore, D. J., & Farhart, P. J. (2000). Cricket fast bowling performance and technique and the influence of selected physical factors during an 8-over spell. Journal of sports sciences, 18(12), 999-1011.

S. Glazier, P., Paradisis, G. P., & Cooper, S. M. (2000). Anthropometric and kinematic influences on release speed in men’s fast-medium bowling. Journal of Sports Sciences, 18(12), 1013-1021.

Stodden, D. F., Fleisig, G. S., McLean, S. P., & Andrews, J. R. (2005). Relationship of biomechanical factors to baseball pitching velocity: within pitcher variation. Journal of applied biomechanics, 21(1), 44–56. 

Takahashi, K., Fujii, N., & Ae, M. (2003). Kinematic and kinetic comparison of different velocity baseball pitchers. In International Society of Biomechanics XIXth Congress.

Thomas Cortebeeck - Performance Optimisation Coach for Olympians & Pro Athletes