A Self-Organization Model For Arm Pain Management

by Randy Sullivan

The following is an excerpt from Randy Sullivan, MPT. “Start WIth The Pain.” iBooks. 


“My name is Randy Sullivan. I am a physical therapist and a baseball coach. They call me “The Arm Pain Assassin.” I am one of the best arm pain specialists on the planet and if you ask me why your arm hurts I will tell you …

I don’t know. 

“Ok… so maybe that sounds a little arrogant, but you get the point. I know a LOT about throwing, a LOT about arm pain and a LOT about how to get rid of it. 

But, it’s complicated. You see, there are no “causes” of arm pain — only a multitude of contributors that combine in the perfect storm to create an environment for pain or injury to occur. The first step to managing arm pain is to stop looking for “diagnosis and cause” and start looking for dysfunction and contributors.”

“At the core of the problem is that there are thousands and thousands of different ways to throw. It’s what Dr. Nikolai Bernstein called “degrees of freedom”. In any human movement, there are multiple pathways to achieve the same goal. When executing a complicated movement like throwing, there is no such thing as “ideal technique” or a “repeatable delivery.” Every throw you make will present subtle deviations or errors that when combined with all the other possible permutations of variables, could lead to a painful throw.

When you throw a ball that doesn’t hurt, or when you throw a ball that hits a target spot-on, or when you throw a ball harder than you’ve ever thrown in your life: I may have some ideas about mechanical efficiency, disconnections, preparation, workload, etc… but if I’m truly honest with you I have to admit that I don’t really know.

But here’s the cool thing…

I don’t need to know exactly why your arm hurts.

Because your infinitely intelligent body already knows. And if we’re willing to observe and to listen, it will tell us how to bring you back.”


Researchers have identified, self-organization as one of the body’s preferred mechanisms for movement skill acquisition and performance improvement. Critics have dubbed it “the FIO (figure it out approach) approach, asserting that without the specific oversight of every repetition, the athlete will learn incorrect movements that either decrease his performance or put him at greater risk for injury.

According to Rob Gray of  The Perception and Action Podcast, “The main critique of self-organization is ‘if I just let the athlete do whatever they want they will develop bad habits and never find an effective solution.”  

As Gray notes, “If you use isolated, uncoupled, low variability drills, of course they will!  There is no reason to develop
a transferable solution in such conditions. A necessary requirement of self-organization is representative design.  Using a self-organization approach does not mean there are no wrong movements.  It means there is no one ideal movement you are going to prescribe.  Instead, there is an acceptable range you will let the athlete self-organize within to adapt to the own individual constraints.”  

Gray’s point:  If you break down training into sterile, repetitive drills that don’t present movement challenges within the context of game-like conditions, you could create an environment that encourages the development of “bad habits.”  For examples, if your practice involves robotic repetition of one movement, under one condition, training corruption could lead to “bad habits.”  Ideally, you need to create a richly variable training environment that encourages experimentation with different movement options that all have the same outcome goal, When the athlete if free to experiment with his body and presented in real time with  objective, measurable feedback (KR) on his performance, he is more likely to develop robust and resilient patterns that are resistant to perturbations and state shifts.


Two strategies for encouraging self-organization in skill acquisition are the Constraints-Led Approach (CLA) and Differential Learning.  

In the Constraints-Led Approach, coaches manipulate variables in the task, the environment or the athlete to encourage a player to choose more efficient movement patterns for accomplishing the prescribed goal, while discouraging him from choosing less optimal patterns.  

At the center of the differential learning debate is the understanding that there is no single optimal movement pattern for any task, and that it is impossible to repeat the same movement twice.  Instead of seeking a repeatable movement, a more intelligent goal might be to develop an adjustable movement pattern.  By creating a training experience rich in variability the coach allows the athlete to pre-rehearse the adjustments necessary for efficient movement :

One example of a CLA is the use of a connection ball between the biceps and the forearm when repatterning the movement of “forearm flyout.”  Forearm flyout occurs when at weight-bearing foot plant on the lead leg, the angle between the athlete’s humerus and forearm exceeds 90º.  According to researchers Aguinaldo and Chambers (2009), this places the UCL under significant stress.  Squeezing a connection ball between the forearm and humerus during the throwing act encourages the athlete to choose movement patterns that keep the elbow inside the of 90º flexion and discourages movement patterns that allow the elbow to drift outside of that range.  In doing so it nudges the athlete toward movement options that reduce stress on the elbow.  When the athlete is gradually weaned from the training aid, his body begins to move the new, less stressful pattern, to the forefront of his list of movement options.

An example of differential learning can be seen in weighted ball training.  Variable weighted plyo ball training assists in improving movement-patterns and refining arm action. The variable weights and sizes of the balls offer a wide array of coordinative experiences thus allowing an athlete to rehearse and refine subconscious adjustments.  These adjustment plans, known as “preflexes,” are predictive reflexes that allow for rapid adjustment of movement patterns in response to (indeed in anticipation of) the entire collection of perceptual information gathered by all available senses — visual, auditory, touch, proprioceptive and vestibular).  Real-time adjustments to movement patterns for coordination of athletic movements must occur under significant time pressure.  There isn’t enough time for all the information gathered to travel to the brain, undergo the process of interpretation, then have that information transferred to the movement center for an action plan, which then has to be delivered to the muscles.  In fact, the adjustments must take place in less time than a reflex (like a knee-jerk in the doctor’s office), which only has to travel to the spinal cord before being turned around into an action.  Therefore these responses are given the name “preflexes” (predictive reflexes).  Accuracy, timing, sequencing, and synergy of these preflexes are the essence of coordination and control.  Preflexes cannot be practiced through rigid, choreographed, repetition of a supposed “ideal” movement patterns.  However, preflexes can be trained through differential learning.  

My roots in human movement science are deeply grounded in the physical therapy universe.  I am the CEO and principal owner of a private practice physical therapy clinic.  Previously, under a misguided quest to achieve the mythical “ideal movement.” I and some of my colleagues have discouraged so-called “substitution patterns” or what we deem to be sub-optimal movement.   In doing so, we squelch our patient’s or athlete’s attempt at self-organization as he searches for movement solutions.  In the infinitely variable dynamic biological system that is presented by our athletes and patients, we will not find a single, “ideal” movement pattern that fits everyone.  When our injured athletes are coming back, we should first get them all the medical help we can.  After we have a clear picture of the presence or lack of structural damage, we begin the process of self-organization.  Of course, any movement that might put a patient or athlete at risk for injury should be immediately redirected, but as long as a movement is within the parameters of safety, we allow a modicum of variability.

The scope of the problem of a reductionist view of pain, injury and rehabilitation reaches far beyond the world of baseball. Compassionate doctors and therapists (myself included) of avid distance runners experiencing hip, knee, ankle, or foot pain have reflexively and routinely instructed their patients to stop running, yet when those athletes have resumed the activity, their injuries have persisted.

Early in my life as a physical therapist, It was still a common practice among physical therapy providers to inhibit or discourage what were perceived to be aberrant movement patterns or “compensations.”  For example, if a patient with a shoulder injury was asked to raise his arm he might substitute shoulder shrugging and lateral trunk tilt to accomplish the goal.  In most physical therapy clinics, this movement required immediate correction.  We feared that if we allowed inefficient movement patterns, we would cause the patient to develop“bad habits” that would be inalterably engrained. As my physical therapy practice and experience evolved, this fear proved to be unfounded.  Under the dynamical systems theory, these “substitution patterns” could be viewed as the body’s attempt to self-organize a movement that could eventually be nudged to greater efficiency and functionality.  As I released the reigns and allowed my patients the freedom to experiment within the parameters of safety, movement patterns improved.  Initially they weren’t what you might consider, pretty, but eventually, they normalized into more natural looking movements.

At The Florida Baseball Ranch, we use a combination of CLA and differential learning to create an environment that allows our athletes to self-organization away from arm pain.  In doing so, we seek to elicit movement patterns and connective tissue states that are more robust (i.e., less sensitive to perturbation) and more resilient (less likely to yield under stress).  We use a 3 tiered arm pain management process.  Our current director of player development, Corey Stump knows first hand the value of our approach and in our next blog, he’ll tell you how the FBR self-organization approach to pain management saved his career and gave him on last shining season in the sun.

Share this article

Leave a comment

Your email address will not be published. Required fields are marked *