corrective exercise bell curve

The Corrective Exercise Bell Curve

I’m going to admit something that may come as a surprise to you.  Corrective exercises don’t always work for me.  There, I said it, I feel liberated now!

Corrective exercises are one of those things that have seen a recent rage in popularity, in both the rehab world but probably even more so in the personal trainer world.  Everyone is now assessing biomechanics and movement patterns and trying to prescribe corrective exercises to address what they see.  This is fantastic.

I recently co-authored an article with Jon Goodman from the Personal Trainer Development Center on how physical therapy and personal training can collaborate more effectively.  We discussed this concept a little bit.  Jon took more of a hard stance against personal trainers performing assessments, for several reasons that he discussed.  I don’t feel as strongly Jon on the subject and welcome the development of systems like the Functional Movement Screen that all of us can use to look at movement patterns and communicate better between professions.  Anything we can do to individualize someone’s programming is awesome in my mind.  But there is a caveat…

[quote]There is a dirty little secret that I don’t hear a lot of people talking about – corrective exercises don’t always work.[/quote]

This has almost become like the story of the Emperor’s New Clothes, where people are a little afraid to admit that corrective exercises don’t always work.  Perhaps they think they aren’t skilled or intelligent enough to make the corrective exercises work!  Well, I am here to make you feel better.  Corrective exercises don’t always work for me, either, and understanding why they “don’t work” is just as important to understanding why the “do work.”


The Corrective Exercise Bell Curve

To better illustrate the spectrum of corrective exercise efficacy, I have developed the corrective exercise bell curve.  The corrective exercise bell curve explains why some people don’t respond to corrective exercises.  I am not 100% certain of the exact percentages, this is just a model, but a starting point for discussion at least.

corrective exercise bell curve

In this diagram, you can see that there is a certain percentage of people who are going to respond very favorably (and often rapidly) to corrective exercise.  These are the all-stars that we all love to work with, call them the rapid responders!

Conversely, there is a certain percentage of people who just aren’t going to respond to corrective exercises at all.  For these people, something is not allowing the correctives to work.  Perhaps its pain, pathology, malalignment, biomechanical, structural abnormalities, or even neurophysiological.  These people essentially need more than corrective exercises.

Then there is everyone else in the middle.  These are the people who may respond to corrective exercises, but it probably isn’t going to be a quick fix.  These people are going to take some time.


Applying the Corrective Exercise Bell Curve

This is all important to understand so you can begin to classify the people you screen.  If you are a personal trainer that just performed a movement screen, programmed some corrective exercise, and was able to clean up some poor movement patterns, congratulations!  That is awesome, you did a great job for your client and maybe even saved our healthcare system some money in the future!

For people who do not respond to corrective exercises, this is where I really see the benefit of personal trainers and physical therapists collaborating.  We can do much greater things together than alone!

For the smaller percentage that is never going to respond to corrective exercises alone, they need a full physical therapy evaluation and will need a combination of treatments including things like manual therapy techniques, neuromuscular motor planning techniques, and eventually corrective exercises.

But here is the really cool group to work with – everyone else!  This is the gray area that we could really collaborate well on to help people achieve their goals.  Notice in the middle group, I stated that corrective exercises MAY work or may take LONGER to work.  In this situation, if physical therapists and personal trainers collaborated more, we could really make a difference in a lot of people.

Physical therapy can work in tandem with personal training to help people achieve their goals faster by combining things like manual therapy with their workouts.

I’m lucky, throughout my career I have worked side by side with some of the best strength coaches and personal trainers, people who have made me better at why I do.  This is by far my ideal work environment and why I always try to team up with a multidiscipline group of people.



To better illustrate, let me come up with a a couple of examples.  Perhaps you notice someone has pretty poor squat mechanics.  Corrective exercises aren’t working.  This is a perfect person to collaborate with a physical therapists.  Perhaps their hip capsule is tight or their hips are not aligned well (just a couple of examples, it could be several things).  All they may need is just a kick start in the right direction with specific manual therapy and they may be good to go.

Or how about you screen someone with really poor shoulder mobility on one side.  They don’t really have many symptoms other than a general ache in their shoulder from time to time, but the corrective exercises don’t seem to be working.  I was actually referred a patient just like this last week from one of the personal trainers I work with frequently.  That specific patient had very specific glenohumeral capsular tightness that wasn’t going to get better with corrective exercises alone.

I do this all the time and can often times clean up someone’s patterns in a few sessions, sometimes even less, by using the right manual therapy techniques to address their concerns while they continue to train.  To me, that is fun.  Helping people achieve their goals faster through collaboration.

So next time you feel like your corrective exercises are not working, don’t feel bad about it.  You are not alone.  If you are a physical therapist or personal trainer, find a respective partner to collaborate with and remember the corrective exercise bell curve.





  • Kenny

    I have a few questions in regards to those who don’t benefit from corrective exercise due to hypothesized capsular tightness.

    1. How does one measure “capsular tightness” in a way that is reliable and valid?

    2. If capsular tightness was present, are we able to make lasting changes in the connective tissue that composes a joint capsule?

  • Kenny, I’m not sure your question is on topic with the article.

    But I will take your bait…

    I’ll throw it back to you, lets say someone walks into you clinic with shoulder pain. They have limited passive ER at 90 degrees abduction and limited elevation, also has a moderate decrease in inferior glenohumeral glide when compared to the other side (it’s obvious on exam)? MRI demonstrates a 6mm increase in thickening of the inferior capsular tissue and synovium?

    What do you do with this person if they are your patient? What do you tell them?

  • Kenny

    Hi Mike,

    Thanks for your reply, it is much appreciated. I don’t see how my questions are not relevant given that you suggest corrective exercises may not work in those with capsular tightness. I’m curious if (and how) we can identify these things and the framework that defines our assessments and subsequent treatments. Wouldn’t having a better understanding of this help us better identify who would and would not benefit from corrective ex?

    Anyway, knowing what we know about imaging findings and symptoms, can we say that the 6mm thickening is the cause of the pain and limited elevation? Are joint play assessments reliable and valid? Can we make a change in that thickening with the interventions and techniques performed in the clinic? I’m not sure we can say yes to any of those questions with any sort of confidence.

    I’m not disagreeing with your idea that some people might need a thorough PT evaluation and a subsequent combination of treatments (that you list above) before beginning corrective exercise. However, I do believe that we can benefit from using a deeper model when thinking about our assessments (in this case, capsular tightness and who benefits from corrective ex) and treatments.

    • Kenny, I never said your questions were not relevant, they were just off topic. I openly state in the post that the cause of corrective exercises not working is multifactorial and related to many systems of the body. Your questions are very relevant, but dissecting a very loose example provided to illustrate a point is well beyond the topic of a 1000 word blog post.

      While all your questions are important, I will repeat my questions to you. In your clinical experience:

      What do you do with this person if they are your patient? What do you tell them? What deeper model of assessment and treatment would you provide?

      • Kenny

        Hi Mike,

        Thanks for taking the time to reply. I do not disagree with the treatment ideas you suggest in your post (manual therapy, motor control, exercise) and think they are great directions to take with your case example as they all demonstrate improvements in helping people getting back to deficient, pain free movement. Understanding the patients goals and context of the injury/pain/limitations would also provide valuable information on where to go with your hypothetical patient. As for what I would tell them, that goes back to my original questions. Would I tell them they have capsular tightness or decreased inferior glenohumeral and this is why they have pain/aren’t responding to exercise/can’t elevate their shoulder? No, because I don’t feel I’m able to reliably and validly assess or make changes to that thickness with the interventions performed in a PT clinic. I’m open to being wrong about this (I am still just a student, after all!) and would love to hear your take on this, which is why I asked. It was ever my intention to “bait” you, I hope it didn’t seem that way. As for the deeper model I refer to, I think we can all benefit from thinking critically about what is occurring when we put our hands on patients, I’m sure we are in agreement with this. For further explanation of what I meant– I’ll defer to Diane Jacobs and her explanation of operator v interactor and shallow/deep models of rehab (

        • Kenny

          Also, apologies for the typing errors. Deficient should be efficient. The perils of typing on an iPad.

        • Kenny,
          Good job explaining your thoughts. That is what I was looking for! The article you references is fantastic, as is all of Diane’s thoughts and work. I couldn’t agree more that we need to have a deeper model. I am not a fan of the whole “you are out of alignment and I’m the only one that can help” etc mentality, our profession(s) need to take a step forward.

          So lets keep chasing down this road. So if you wouldnt tell the patient they have decrease inferior glenohumeral mobility because you don’t feel like you have a valid and reliable way of assessing clinically, and you wouldn’t treat the patient with joint mobilizations because you don’t feel like you have a valid and reliable way of making changes, what would you do with this patient that is 100% valid and reliable?

          Do you perform the manual therapy but explain the rationale and mechanism differently to the patient or do you do something else?

          To answer your question re: the hypothetical person, he/she person wants to raise their arm overhead with full mobility and without a pinch at the top of the shoulder – that is their goal.

          Keep thinking and explaining your thoughts, this is fun

          • Kenny

            Hi Mike,

            I appreciate the discussion (especially with a fellow NU Husky, no less)! I completely agree with you in regards to the “alignment” mentality. I think our profession is on the cusp of a paradigm shift (that will hopefully help us best meet the needs of our patients), and that is very exciting.

            I do think joint mobilizations would be a good choice for this patient, however I would use a different explanatory model. I don’t think the forces achieved in manual therapy are enough to induce mechanical deformation (see Threlkeld 2002, although I imagine you’re familiar with the article). With this information in mind, I tend to subscribe to the non-specific and neurophysiological mechanisms (those described by Bialosky and his group among others). I think they offer a great way to provide a novel stimulus to the PNS and CNS that may lead to some descending inhibition of pain.

            This can be useful in creating “buy in” on behalf of the patient and improve tolerance to further interventions (whether it be motor control exercises with verbal/visual/tactile cueing, Butler’s neurodynamic techniques, corrective exercise, skin stretch/DNM [nod to Diane Jacobs] etc.) and maximize the person’s function. Will this improve the hypothetical person’s range of motion? Maybe.. depends if the limitations in range are a defense mechanism from the pain and (thus improved to an extent when pain is reduced). Or is it a defect (such as a capsular thickening, contracture, etc.) that may not be possible to change with MT? I also do wonder about the long term effects of passive manual therapy, but that’s a completely different discussion.

            Ultimately, I think it is most important to empower the patient, restore the internal locus of control, and give them the power to self-manage.

            • Great answer. So you would actually perform manual therapy (joint mobs) but just change your rationale and discussion with the patient?

              Do the interventions you reference (“motor control exercises with verbal/visual/tactile cueing, Butler’s neurodynamic techniques, corrective exercise, skin stretch/DNM [nod to Diane Jacobs] etc.”) have 100% validity and reliability? Are you still comfortable using these interventions if they are not completely valid and reliable?

              Kenny, your thoughts are awesome and well beyond a student. Kudos to that. Being inquisitive is an awesome attribute. I call it the power of “why?” Always ask “why” and have a rationale for everything you do.

              I am eager for you to get out in the real world and start gaining skill and experience, neither of which come in school unfortunately. You will continue to learn and adapt your philosophies over time. I can almost guarantee that whatever you feel strongly about now, will likely change with time and experience. This happens to us all. In fact, I’d be disappointed if you didn’t adapt and change as we all continue to learn.

              Realistically, we don’t know and understand as much as we think we do. I’ll be the first to admit it. My clinical style and thoughts have changed greatly over ny career and continue to do so. I always go back to thought that 500 years from now our ancestors may be laughing at us for thinking the world is round.

              I’m not saying I agree or disagree with your comments, I think they are great and encourage you to continue in this direction. I’d recommend you keep an open mind as you enter the profession. Don’t limit your education due to a predisposed bias.

              We are a bit off topic, but just trying to help.

              • Garrett Pfeiffer

                Kenny and Mike (but mostly Kenny),

                These are the types of conversations that fuel our profession. Kenny, I will direct you to the body mechanic, Greg Lehman, if you are not already familiar with him. Sometimes I think he doesn’t do anything to anyone because nothing can be explained, but I know he is only using it as a discussion board for professional growth. I am in my second year out of school and like to think that I am developing a thought/clinical reasoning process that takes into account the information presented by Bialosky et al, David Butler, Peter O’Sullivan, etc as it relates to the real mechanisms of our interventions. We don’t always know why what we do does/doesn’t work, and a lot of it is likely related to the CNS, but let’s face it, we will NEVER figure that out as Mike alluded to, we can only attempt to reign it in as much as we can. Having said that, I caution you to not completely eliminate things from your vocabulary just because you can’t reliably or validly explain what’s happening. Between rehab professionals, that’s fine and fuels our world of progress. However,some, not all, patients can’t or don’t want to hear that you are interacting with their nervous system or that you, in effect, don’t really know why things work. Your words are extremely powerful, so you have to be careful. You will find that you spend a lot of time calming people down after they saw a surgeon because of that power of words. Anyway, they may want a simple answer like “we stretched your hip flexors and made you fire your butt muscles better so your back doesn’t hurt as much”. It’s a simple answer they want even if your thought process and eventual interventions were much deeper than that. Sometimes things are very simple, so we should explain them simply. Sometimes they are complex, but we still need to be simple and give the pt something concrete to grasp. If they can/want a more complex explanation, give it them, I do it all the time and think it really helps. I could go on for hours but clearly you are already ahead of the game and I probably don’t need to. However, I think Mike’s points and perspectives are, in part, related to some of my comments. Please forgive me if not Mike.



                • Kenny

                  Hi Mike and Garrett!

                  First, thank you for the kind words, Mike. It is much appreciated. In regards to your post:

                  “Do the interventions you reference (“motor control exercises with verbal/visual/tactile cueing, Butler’s neurodynamic techniques, corrective exercise, skin stretch/DNM [nod to Diane Jacobs] etc.”) have 100% validity and reliability? Are you still comfortable using these interventions if they are not completely valid and reliable?”

                  My initial post questioned the reliability and validity (construct validity, to be specific) of assessing capsular tightness. I was never questioning the effectiveness or efficacy of joint mobilizations as a treatment. As for the interventions I posted, I feel they are accordant with human physiology and have scientifically plausible explanatory models. Therefore, I would be comfortable using and explaining these interventions to a patient. I don’t feel mechanical deformation meets those two criteria above, hence why I side with the more parsimonious explanation of joint mobilizations. I have not completely ruled out mechanical deformation, I am open to being wrong about this, but have yet to see any evidence of it occurring through manual therapy intervention.


                  Thanks for your reply and the recommendation, it’s always great to hear another student’s perspective. I am familiar with Greg Lehman’s writing, he certainly has been a valuable resource to me as a student. Leveraging social media and blogs has had a huge influence on me, and it sounds like you might be in the same boat.

                  I agree that simplicity is extremely valuable in a therapeutic alliance and agree with much of what you said. However, I don’t feel that we have to sacrifice plausibility when being simplistic. I think we can do much better when being simple than saying something like “we stretched your hip flexors and made you fire your butt muscles better so your back doesn’t hurt as much”, because this is likely not the reason why the pain was ameliorated (I recognize this is just an example, and may not reflect your actual beliefs). I think Butler and Moseley’s book “Explain Pain” is a great starting point for being simple and plausible, if you have not read it, I highly recommend it.

  • Mike,
    Great post Mike, I could not agree more. Not sure about the Bell Curve figures because I think it depends on the population and region you where you work. In Florida, where I am located we have a higher percentage of senior citizens. In many cases the acquired structural deformities in these clients will limit the effectiveness of corrective exercises. In fact I would argue that the screening or assessments performed on these clients is sometimes more beneficial for teaching us what not avoid with the client versus what needs to be corrected. I think you are making this point as well. For example , if a client has knee replacement they will probably fail a overhead deep squat test. So should I work on squat corrections? Maybe, but I might just decide to modify their fitness plan to work within the limitations of the client versus focusing on corrections. Thanks for posting this info Mike.

    • Scott, excellent points, you are absolutely correct! thanks for sharing!

  • For teaching us what (to) avoid with the client (correction)

  • I completely agree with Mike on this!
    “if physical therapists and personal trainers collaborated more, we could really make a difference in a lot of people.”
    Truer words have never been spoken. Wouldn’t it be nice to see both therapist and trainers working side by side in treating clients?
    Two head are definitely better than one. It makes me cringe to see trainers and therapists trying the same routines to bring about change. When one has hit a wall shouldn’t it be time to collaborate with someone else to help make a difference?

    • Ethan Kreiswirth

      Great read, Mike and others! I was surprised that no one has asked; “what does the research say?” It seems the blog has discussed the tx options per patient and that may work for that case report, but, still low level evidence if correctives, and or joint mobs have the best outcomes.

      I will have to research this myself, but a quick PubMed search through clinical queries did not reveal much. I used [corrective exercise AND patient outcomes]. I also tried, exercise AND patient outcomes]. There should be better boolean terms with a search, maybe someone can aid.

      Although single patient outcomes are great, as practitioners, research should drive the treatment choice, yes?. The question is, stability, mobility, and or tissue limitations. I would presume, all three! 🙂

      • Ethan, going to be tough to research as “corrective exercise” is too broad of a term. There are studies on things like the effect of thoracic spine mobilization or manipulation on shoulder pain and function, just as an example.

        Exercise itself is always going to be hard to determine efficacy. Very hard to design a study with proper control.

        But you are right you have to presume everything until you start to weed it out.

  • ethan

    Yes , agree! We can narrow are search to a specific tx vs sham or other. Point being, research should drive the tx choice. In the land of FMS, SFMA, PRI, TPI, and a million more, someone, or maybe myself, needs to start doing stronger cohort studies. 🙂

    Thanks, Mike, always great stuff! You know how to fire people up! 🙂

  • BJ Stockton, DPT

    Hey Mike, great post as always. I also enjoy everyone’s discussion. I think it’s interesting the way we sometimes quantify the effectiveness of our treatments. I think the bell curve above is a good example in how people can really pick apart something that was meant to be a generalization, not meant to actually fit every patient in every clinic. For instance, after reading this, hopefully we aren’t going to look at our next patient and say “there is 20% chance these exercises won’t work on you”. No, we take each patient into account, and using the best available evidence, experience, and theoretical knowledge to see where this patient is most likely to fit on this bell curve. It’s not a “one size fits all” type of situation. What’s more, who’s to say what is meant by “may work” or “work well”. Obviously here this is subjective and would be hard to measure. What if someone were to gain 5% improvement by doing corrective exercises (insert your own method of measure here), would we say these exercises are working? And over what period of time do we do these? I’m sure the actual study addresses some of these questions.

    My point is, most therapist are getting information about how they practice from great blogs like this one, Dr. E’s, Cressey, etc and we don’t have time to read the entire study. I’d prefer not to assume any study can give us metaphysical knowledge about how the human body work, rather we should start building ways of practicing and structuring our treatments based on the availability of information, experience, and theoretical knowledge combined. Remember, if we only did evaluative exam techniques, followed by treatments which only based on “evidence”, we would probably be spending the most of our time with our hands in our pockets.

    Mike, great post and great responses as always!

  • Cory


    Great post. As a personal trainer I feel this was a much needed follow-up to your piece with Jon Goodman. Honestly, I was a bit disappointed by its tone and I think the message got lost.

    I think this goes more into the “how” and “when” of collaboration and actually empowers trainers to be a part of the process rather than existing in deference and subservience.

    We are fortunate to have great relationships with a variety of rehab professionals so its easy for us to refer out in confidence and have open dialogue. I’m also quick to point out to clients that corrective exercise isn’t magic. Clients certainly appreciate the humility that comes with not knowing ALL of the answers (but having a number of rational answers won’t hurt either).

    Anyhow, for trainers out there without great working relationships with rehab professionals I think this Bell Curve model is helpful for guiding decisions on when to collaborate more fully.

    Thanks for all of the great content!

    • I agree Cory, thought this would be a good follow up. Thanks for commenting.

  • Laura Neuburger


    I had a discussion with a Strength and Conditioning coach not too long ago regarding this topic. I believe that other fitness professionals are more than qualified to assess movement in a client.

    I like your application of a bell curve, AND to take that one step further I believe that Physical Therapy and Personal/Strength training are 2 bell curves that are placed along a continuum of one another. When a personal trainer is doing >10 minutes of “corrective” exercise in a training session, that person will probably benefit from being treated by a PT simultaneously. This will allow the PT and trainer to do their jobs more effectively and allow the patients to get more value out of their training sessions. By the same token, when a patient (especially athlete) is nearing the end of their treatment with me, I set them up with the appropriate professional to help them implement a more solid strength program.

    As it seems that you have found in your experience, this team collaboration to patient/client care is paramount to either profession working independently of one another. Hopefully this will continue and eventually become the norm among clinical practice in sports medicine.

    • That is a good perspective Laura, thanks for sharing!

  • Mike,
    We are currently working on a throwing program that will cover from little league ages to college age pitchers. A lot of research has been put into preventative testing measures (e.g. IR/ER ratio shoulder strength, Hip ROM, Scapular function, etc…) Then there is FMS, then end all of test measures (jk). So, the team is split. A lot of people have bought into the FMS as the only realy needed tool, and some of us feel that it isn’t quite complete for assessing a pitcher. What’s your take on this idea? And thanks for your time

    • FMS is great but yes not ideal for a unilateral sport athlete with very unique characteristics.

  • Daniel Ros

    This is the fallacy of the appeal to authority. As a ‘personal trainer’ Chek Practitioner and Gary Gray student I certainly don’t need a physical therapist to help my clients. The very few who don’t respond to corrective exercise are not consistent or have deep seated emotional problems neither of us is ‘qualified’ or better yet ‘educated’ to do. Those I certainly do refer out. Yes physical therapist get an amazing education but I chose my own education and in 9 times out of ten it’s been SUPERIOR to physical therapists I’ve delt with. There is no line in the sand that limits my ability to learn. Any such idea is ego driven dribble. The true fact is that most personal trainers are deficient in education to even work with ANY population and they would certainly benefit from working with a physical therapist. However, if you are a personal trainer the only thing stopping you from becoming as knowledgeable as any pt is a library card.

    • I agree with many of your points, but there is a lot more to PT than reading a book. I’m not sure I would agree that anyone could read a book and become as knowledgable as a PT. If you feel that way, I think you have a lot more reading to do.

Thanks for sharing! Follow me online for even more great content!

Send this to friend