Radio-Ulnar Dynamics and Emergence in the Pitching Delivery: Revisiting “Active” Pronation

by | Jul 26, 2016 | Coaches | 4 comments

We recently reviewed a study published earlier this year in Sports Biomechanics by Matsuo et al., which demonstrated that the proximal radioulnar joint (i.e., the elbow) supinates (although remains “pronated”, relatively speaking) just prior to ball release. As we discussed in our review, this study casts some doubt on the concept of pronation training, which has long been suggested in baseball pitching circles to improve throwing velocity and mitigate injury risk. Of note, we drew on Dr. Gabriele Wulf’s constrained-action hypothesis to explain why pronation-based training may lead to faulty self-organization of the motor system during pitching.

The purpose of this article is to expand on this idea; to further the discussion of the paper by Matsuo et al. from a biomechanical perspective, and provide an alternate but aligning view of how the central nervous system (CNS) self-organizes movement to meet the demands of pitching. In addition, we’ll look at some clinical evidence which demonstrates a lack of appropriate radioulnar motion in many of the injured throwers we work with.

The study of baseball pitching has become synonymous with “mechanics”, and the never ending chase to perfect the mechanical contributions to pitching efficiently, effectively, and with the highest velocity possible. Along the same lines, it is unfortunate that pitchers who perform poorly, or develop a throwing-related injury are quickly painted with the “poor mechanics” brush.

At the Baseball Performance Group, we have outlined numerous issues with this widespread focus on mechanics, drawing on concepts from motor control, dynamical systems theory, and the understanding of complex systems. In addition, to fully and completely analyze the mechanical parameters of the pitching motion it is very difficult to do so on a macroscale. A macroscale view of the pitching motion allows us to make fairly good general estimations of what actually happens at each joint involved in the action and for the most part this can generate a somewhat accurate representation of what occurs, however it can also lead to faulty assumptions and poorly applied logic. This is one of the main reasons why it is necessary to delve into the microscale and understand the small, subtle details of gross movement.

One of the areas where this macroscale view has predominated is at the elbow. This has led to a very general understanding that from the moment of arm acceleration (i.e., the phase of the pitching delivery following “lay back”) the forearm starts to pronate, and continues to pronate into ball release and follow through. This observed pronation has been associated with having “good” mechanics, and the ability to generate elite ball velocity. This macroscale understanding of upper limb biomechanics has perpetuated many different forms of training, including the heavy training of pronation using various overweight implements, and the explicit cuing of so-called “active” pronation to improve arm action and ball velocity. Likewise, in the midst of the so-called “epidemic” of elbow injuries in baseball, it has also led to the rather simplistic view of muscle function in that, because the forearm pronates, we should strengthen the muscles that pronate the forearm (e.g., pronator teres) to protect the ulnar collateral ligament and mitigate injury risk. However, from the perspective of motor coordination, this assumption is misguided, as this is not how motor commands are generated. (For more on the Tommy John “epidemic”, check out our interview with Los Angeles Dodgers’ Physical Therapist, Steve Smith.)

One of the most intriguing takeaways from the recent study by Matsuo et al. is that late forearm supination prior to the instant of ball release may in fact occur to adjust the position of the wrist to account for subtle pitch-to-pitch variability in shoulder kinematics, which may contribute positively to throwing accuracy and velocity. Clearly the authors are describing an emergent phenomenon generated by the central nervous system to meet the demands of the pitching task (e.g., speed and accuracy), which fits nicely into Karl Newell’s Model of Constraints. As we discussed here, constraints can be broadly characterized as any element that will lead to a change in motor outcome, and can be subdivided into three major categories:

  1. Organismic constraints.
  2. Environmental constraints.
  3. Task constraints.

During goal-oriented activity, such as throwing a fastball as fast and as accurate as possible, the interaction of all constraints acting on the athlete, and how the CNS interprets them leads to the motor coordination profile that becomes optimized with practice and experience.

In the paper, the authors stayed away from describing their findings from a truly descriptive biomechanical perspective (refreshing!), and actually discussed the fact that the forearm supination observed just prior to ball release was not simply a result of passive force, but may in fact have been a requirement of the task to maintain the hand oriented towards the target. Supination, in this case, is an emergent phenomenon.

But why does this happen?

To answer this, it is necessary to delve into the underlying biomechanics, and to consider how the CNS organizes the many degrees of freedom (DOF) to improve the efficiency of movement to maximize performance outcomes while at the same time minimizing energy consumption and adverse effects on the structures and tissues involved in the movement. A major question that needs to be addressed, then, is:

How does the CNS exploit the biomechanical properties of the joints for movement organization, production and coordination?

One of the most obvious biomechanical properties of human movement is that the involved limbs are linkages of several segments, usually described from the perspective of the joints. This is what we casually refer to as the “kinetic chain”, which is how, during pitching, force generation from the lower body can be translated to the upper body to produce velocity. Due to this chain effect and the multi-joint structure of the limbs, the limbs will undergo a considerable amount of force during movement. Dr. Glenn Fleisig, and the research team at ASMI has provided us with ample data showing just how large these forces can be during the pitching motion. As a result of these forces, every joint within the chain will undergo what is termed “torque”. Torque application at one joint instantaneously creates motion at all joints of the chain, a phenomenon that is often referred to as “dynamic coupling”. Strictly speaking, torque is a measure of how force acting on a joint causes the joint to rotate and thus is related to the concept of angular momentum, whereby increases in angular momentum are caused by a net unbalancing of the torque across a joint. At any joint at any point in time there are two major causes of torque:

  1. Muscle force.
  2. Gravity.

For the purposes of our discussion we are going to focus on a third component of torque, which is created passively as an “interaction torque” that is exerted from one joint to the next. Interaction torque makes motions at all joints of the limb involved in a movement highly interdependent. In essence, interaction torques arise due to the serial linking of limb segments. With increased movement speed interaction torque increases, therefore during a high speed movement such as pitching it would be expected that the interaction torque would be high. Nikolai Bernstein was the first to hypothesize that the CNS develops a strategy of exploiting the interaction torque for improving movement production and efficiency. Consequently, to produce a desired arm path and hand trajectory during the pitching motion, the human arm, which is in and of itself a multi-joint serial chain, the torque at any joint must be composed by taking into account the movement of all joints involved.

From a motor control perspective the biomechanical concept of interaction torque can be explained using the leading joint hypothesis as it allows for an apparent categorization of the joints in the chain based on their mechanical subordination in the linkage. There is usually one joint that is labelled as the leading joint as it creates the movement dynamics for the rest of the limb. The interaction torque generated by the leading joint thus produces a very large and powerful effect on the subsequent (subordinate) joints. In order to meet task requirements the role of each subordinate joint and its surrounding musculature is to monitor the effect of the interaction torque and utilize it to create the necessary pattern of muscle activation and joint position that results in coordinated limb motion that is required by the task, including such things as movement direction, plane, velocity and accuracy. This biomechanical concept fits nicely into the concept of muscle synergies, which is how the CNS organizes movement into a pattern of activation of a grouping of muscles that cross or act on a particular joint as a result of a motor task or goal. Now, suffice to say, all of this biomechanical information is not altogether stored in our brain as a set of instructions to be run each time the pitching motion is undertaken. Although there is likely a general representation in the cortex, this process is driven by the periphery, as the ongoing monitoring of our motor output is based on the interaction between sensory information, the environmental context, and the desired performance outcome. Control resides within the triad of couplings between the nervous system and the body, the nervous system and the task, and the nervous system and the environment.

How does this work in pitching? From the task viewpoint it is important as a pitcher that the ball is thrown with as much velocity as can be generated and at the same time with as much accuracy as possible. Our nervous system “understands” this and adopts a particular motor control strategy that allows for the utilization of interaction torque to create higher levels of joint angular velocity while doing so safely. From this perspective, the interaction torque can be thought of as the “actuator” that the CNS positively produces to create an emergent pattern of movement. Interaction torque stems from the muscle torque at higher order or more proximal joints. It has been shown that muscle torque is high at the trunk and shoulder which increases the angular velocity at each specific joint and in addition increases the angular velocity of the elbow and wrist through the interaction torque. It is interesting to note that during long axis rotation of the arm, which is the parameter of the throwing motion whereby the elbow starts to extend before the shoulder starts to internally rotate, the interaction torque switches from a proximal-to-distal (shoulder-to-elbow), to a distal-to-proximal (elbow-to-shoulder) direction to drive the internal rotation of the arm which allows the elite thrower to continue to generate wrist velocity. As ball release is approached, the interaction torque creates supination of the forearm to maximize the accuracy of the hand path and finger position to allow the ball to come out in the position that best allows for the dual task of throwing accurately with maximal throwing velocity. In addition, this can be interpreted as an attempt of the CNS to mitigate large amounts of elbow forces at this juncture of the pitching motion, as a more supinated proximal radioulnar joint position resembles a better position to absorb and transfer forces.

It seems to be clear that during motor performance the continual interplay between biomechanics and the CNS allow for features of behaviour to emerge that may not be clear when looked at on the macroscale. The Matsuo et al. article really allowed us to understand the biomechanical interactions of the proximal radioulnar joint at the microscale and how it relates to the gross movement of pitching.  The next obvious question would be how does that relate to the physical management (and training?) of pitchers?

When examining elite pitchers, we at the BPG take a very thorough approach and correlate our “on table” findings to the analysis of movement. Each specific joint involved in the pitching motion is taken through a systematic active and passive examination, as well as a movement-based examination looking for findings indicative of altered kinematics and poor motor control.

It is interesting to note that one of the biggest limitations we see clinically during assessment of elite injured pitchers is a lack of appropriate supination at the forearm, while still maintaining the ability to achieve full and necessary ranges of pronation. In the videos below, you can see two college pitchers showing a lack of radio-ulnar dissociation, or an inability to fully supinate, both actively and passively, on physical assessment.

This finding leads us to conclude many things about the role of the forearm in throwing mechanics but more importantly this has implications for motor learning and the control of movement as this shows an inability to dissociate the radius and ulna from the humerus in critical degrees of elbow extension where it has been shown that there are large amounts of valgus force.  To us testing dissociation gives us a glimpse into the microscale as it is the ability to move one joint relative to another and is paramount for optimal and efficient movement at the macroscale because it gives us an indication of the movement potential of the joint, its ability to generate feedback to the CNS, how it will potentially utilize interaction torque and lastly how that will lead to complexity of movement. Quite simply, if the pitcher cannot dissociate any joint involved in the movement it can become a source of impaired movement efficiency and limits the options available for appropriate self organization as the CNS must reorganize its hierarchical control of the movement based on the limited feedback it receives.  This obviously has performance and injury implications in that if the central control of the movement leads to pattern restructuring each subordinate joint in the linkage will fail in the fine regulation and utilization of the interaction torque, leading to changes in outcome variables such as arm path, release point, and pitch velocity/accuracy but it will also lead to less economical movement, and increases in accepted joint force due to inefficient muscular and connective tissue buffering.

It is our opinion based on the data and how the data fit into the model of the self-organization of movement and how the CNS governs the biomechanical interactions of this movement, that it is necessary to limit the emphasis on pronation based training at the forearm as it has little to do with increasing pitching performance.



  1. Masaya Hirashima, Kazutoshi Kudo, Koji Watarai, Tatsuyuki Ohtsuki. Control of 3D Limb Dynamics in Unconstrained Overarm Throws of Different Speeds Performed by Skilled Baseball Players. Journal of Neurophysiology. 97: 680-691. 2007
  2. Natalia Dounskaia. Control of Human Limb Movements: The Leading Joint Hypothesis and its Practical Applications. Exercise and Sport Science Reviews. 38: 201-208. 2010
  3. Tomoyuki Matsuo, Tsutomu Jinji, Daisuku Hirayama, Daiki Nasu, Hiroki Ozaki. Radioulnar Joint Supinates Around Ball Release During Baseball Fastball Pitching. Sports Biomechanics. 2016
  4. Steven Barrentine, Tomoyuki Matsuo, Rafeal Escamilla, Glenn Fleisig, James Andrews. Kinematic Analysis of the Wrist and Forearm During Baseball Pitching. Journal of Applied Biomechanics. 14:24-39. 1998