YTWL Shoulder Exercises

Why I Do Not Like YTWL Shoulder Exercises

I remember when the YTWL shoulder exercises started to become popular.  Using the “YTWL” letters was is a pretty good description and easy way for people to remember the shoulder exercises.  I even joined the bandwagon and started training everyone bilaterally, even injured people rehabilitating from a shoulder injury.  Because the YTWL shoulder exercises are performed bilateral, you can get more work done in a shorter amount of time and work on symmetry. But I was never really pleased with the exercises, trying a bunch of difference variations.

First was standing and bent over – A good place to start, seems simple, right?  Well I quickly found out that most people don’t get into a good position to perform these exercises.  Most people do not perform this parallel to the ground, but more at a 45 degree angle to the ground.  I don’t like that, this increases deltoid involvement, which is the last thing I want when performing rotator cuff and scapular exercises

YTWL Shoulder ExercisesNext I tried lying prone on a physioball – what a great idea, right?  Train the shoulder and scapular muscles while stabilizing the core!  Well, not exactly.  Research on this topic has been conflicting, but in general studies have not shown that EMG of the shoulder or core muscles is increased consistently when performing exercises on a physioball, but one trend emerges from the research – force output is reduced.  We’ll assume this is similar to the difference between back squats and leg press, you can lift more weight on a leg press because you don’t have to stabilize as much.  OK, this might not be that bad, specifically for healthy people or athletes training for function.  But remember why we are performing the YTWL shoulder exercises – to enhance shoulder and scapular function, which does not seem to be the emphasis when performing YTWL exercises on an unstable surface.

I should also mention that I wasn’t in love with the positioning most people got into when doing these exercises on a physioball.  Again, most people were not parallel to the ground and most people can’t perform the exercises through full range of motion as their arms are longer than the physioball.  So again, more sacrifices for maximizing benefits of the shoulder and scapular muscles.  Plus, seemed to me people probably didn’t have the best core stability and were rocking back into hyperextension of the lumbar spine to complete the movement pattern.

Next, I tried taking away the unstable surface and just performing the YTWL exercises prone on a table.  Doing this requires that you pretty much have your head and shoulders out over the edge of the table or bench.  Not bad, doing the YTWL shoulder exercises in this position actually seemed to be decent – you can still use your normal weight and you actually had to stabilize the lumbar spine in neutral (if you cue the person to stay flat on the table and not hyperextend).  I finally reached the “body parallel to the ground position” I had been searching for with the previous positions.

upper trapeziusUh-oh, now that I got there, I don’t like it.  To have your head out off the edge of the table and perform bilateral shoulder exercises you really need a lot of upper trapezius and levator activity.  We all know how I feel about reducing upper trapezius activity and getting out of our upper trap dominant posture.  In addition, again, we are performing these exercises to enhance shoulder and scapular function.  Anything that enhances upper trapezius and deltoid activity is probably working against our main goal, especially when we know that the ratio of upper trapezius to lower trapezius activity is correlated to shoulder impingement.  So again, seems counterproductive to me.

Why I Don’t Like the YTWL Shoulder Exercises

As you can see, there are some limitations when performing the YTWL shoulder exercises.  To summarize, here are some of the limitations of the YTWL exercises that concern me:

  • If not performed parallel to the ground, it changes the muscle angle and recruits more deltoid
  • Easy to hyperextend the lumbar spine
  • Performing on an unstable surface potentially reduces force output and reduces the emphasis of the shoulder and scapular muscles
  • Performing on a physioball does not allow for full range of motion
  • If performed off the front of a table or bench, recruits too much upper trapezius and levator to help hold the head up.

What I Would Recommend When Using the YTWL Shoulder Exercises

To me, if the primary goal is to increase the strength of the rotator cuff and scapular muscles, I am not a fan of the YTWL exercises.  I will perform them all but I really think we need to simplify things and just perform them unilaterally on a stable and parallel to the ground surface (like a treatment table).  Yes, you have to turn your head and not stay neutral, but at least the neck muscles are relaxed.  You can still perform the shoulder W exercise bilaterally (click the link to see my past post and video demonstration of the shoulder W exercise technique), but I would perform the Y’s, T’s, and L’s unilateral.

YTWL Exercise

If you are not rehabilitating from a specific injury or surgery, or if your primary goal isn’t to maximize shoulder and scapular strength, then performing the YTWL shoulder exercises may be OK especially if your goals are to maximize symmetry or movement function.  Just realize that if you have specific deficits you are working on you may be better suited to just perform the exercise the plain old boring way.  Perhaps that is your starting point and then when strength is restored, your progress to these other positions.  For the rehabilitation and fitness specialists out there, you really need to coach and cue during the bilateral YTWL exercises to make sure some of the compensatory patterns discussed above are not present.

There is a time and place for bilateral YTWL shoulder exercises, but the majority of time I am trying to enhance shoulder and scapular strength and function.  I see the bilateral YTWL exercises as a progression once you have adequate strength and stability of the shoulder and scapula.  I think performing the YTWL shoulder exercises bilaterally may take away from that goal a little bit, what do you think?

54 replies
  1. Mateusz
    Mateusz says:

    Fair enough, but what about if I want to prescribe some Home Exercises, but a patient obviously doesn’t have a treatment table. Then what ? What kind of exercise should this patient do ?

    Regards !

  2. Adri de Vrieze
    Adri de Vrieze says:

    I can still think of other ways to support the head like support by an assistant, table extension or even a wall/exercise mat therefore reducing upper trap activity. Or using a back extension (like the one by David) and bend down a bit more (I do that pretty often and check if it’s more a shoulder or back exercise). Just food for thought on my own personal view, no critique but more of a dance: I like it when patients get a little range of freedom. Then if you instruct the start position and you roughly instruct what to do that they tend to get to the intended strategy by finding out themselves. Ofcourse sometimes you might need to help out a bit. By using restrictions like the table you restrict the range freedom and the motor programming will be less applicable to other situations I can imagine.

  3. Patrick Brown
    Patrick Brown says:

    I use these as a maintenance exercise once a week and not for rehab. I use the ball to get anti gravity, but I put my knees on the floor so the focus is strengthening the scapular stabilizers and not on stabilizing on an unstable surface. I am more focused on the mid to end range of the motion anyways so that 20 to 30 degrees I miss doesn’t bother me. I then follow that up with horizontal plane d1 and d2 flexion low weight cables stepping far away from the machine to avoid a steep vector and target serratus ant and lower trap appropriately. I then do shoulder shrugs to hit the upper trap. I feel you just can’t neglect that muscle because all together.

  4. Vincent
    Vincent says:

    What about performing them bilaterally in standing with bands or a cable column? It allows for full range and proper core stabilization will prevent lumbar hyperextension.

  5. Josh
    Josh says:


    Great article. I brought this idea to a colleague and he brought up the counter-argument that as you pitch your deltoid and upper trapezius muscles are active so training these bilateral YTWL exercises may be appropriate considering the deltoid and upper trapezius are active but considerably less active, when done correctly, to the middle and lower trap / rhomboids. So from a functional perspective using the exercises to balance the muscle firing of all the muscles involved with a throw is the goal. Really wanted to know your thoughts on this perspective.

    Thank you,


  6. Dino Pappas
    Dino Pappas says:


    I’m getting to this post a little late, but a lot of focus has been the open chain exercises. What about closed kinetic chain exercises for scapular activation? Perhaps a quadruped position with isometic ER Rotation holds while fixating the scapula. May I suggest endurance or time goals to the exercise (4-6 sets of 10-20 seconds). There are other closed kinetic chain exercises, but you get the point.

    • John Buttari
      John Buttari says:

      I recently joined the inner circle and am as well getting to this post late. Dino can you describe or refer me to a picture of how the quadraped exercise is executed?
      Thank you for your time

  7. Craig Liebenson
    Craig Liebenson says:

    The tendency to overactivate the upper traps & to hinge in the L-spine makes this exercise an ideal test incorporating 2 of Pr Janda’s most significant faulty movement patterns. I agree the exercise is usually performed w/ poor technique.

    Although, for those that can handle it, assuming they can create what Pavel Kolar terms a pointum fixum 360 degrees through the mid-core so they hinge i extension at T4-8 rather than the L-spine & facilitate more lower scap musculature than upper scap musculature that it is a powerful progression.

    I typically guide it upright standing facing a door where once the abds are turned on to stabilze the core, the cues are reach up/roll back & down/raise off.


  8. Ben
    Ben says:

    Great post, Mike, and great f/u post from Greg above. Much appreciated, both very helpful clinically. Many of my impingement pts can’t tolerate the Y (obviously), so it’s good to know the SL ER and prone W/L generate as much LT activity as they do without too much UT. The allure of the p-ball can be strong, to both pts and clinicians alike (nothing screams ‘workin’ hard!’ like doing something on a p-ball), so I appreciate the idea of stepping back and refocusing on what we’re trying to accomplish, even it means returning to some basics. Thanks again – both of your blogs are must-reads.

  9. Greg Lehman
    Greg Lehman says:

    Hi Mike,

    Your post inspired me to take a look at some of the research in this area again. I just wrote a pictorial blog post that illustrates the muscle activity of the rotator cuff and scapulothoracic musculature during these exercises (Unilaterality) and compares the YWTs with other common exercises. Much of the research is your own so no surprises for you.

    You can see it here:



  10. Jonathan, PT
    Jonathan, PT says:

    Hey Mike. How about performing the YTWL in a modified isometric deadlift position (slight knee bend to minimize adverse neural tension in the legs)? This way you can adjust the spine so that it’s horizontal(increasing the hip flexion angle)and keep the exercise “pure”. We obviously have to be aware of hyperextension of the L/S but otherwise a good quad/hip/trunk stabilization exercise as well.

  11. Michael Boyle
    Michael Boyle says:

    Mike- great piece. I couldn’t agree more and for the record I was shocked at how hard these are when performed one arm at a time. We will be switching to this technique.

  12. federico
    federico says:

    Hello Mike! Really great article. Now I have a question: I had right shoulder surgery back in 2007 and could never regain strength and size of my right arm due to a 20º flexion and abduction for eleven months after surgery. Even though I have gone to kinesiology and all kinds of doctors, I never regain my normal strength or size. I have been doing YTWL and didnt improve that much either (not only do i have pain, but also a strange sound that i think its the biceps tendon moving its normal place). So, what would you recomend for someone like me? I know that because of surgery and imobilty for that long time, I have scapular dysfunction
    Thanks a lot

  13. Lewis
    Lewis says:

    Just thought I shared this with you guys, I have tried attaching the Gray Cook correction band to a cable machine and getting my clients into talling kneeling postition cueing them to think tall while peforming the T,W and L so far it has work quite well.

  14. ben
    ben says:

    Great post Mike. For my case, i have chronic shoulder injuries so using weights especially int/ext rotation will cause inflammation. However, i find scarecrows (no weights) really good for shoulder stability, and i can also feel some lower traps activated… also helps my scapula stick to the wall and minimise protraction.

  15. Rees
    Rees says:

    Great post.

    I completely agree and have been on this little conundrum for a few weeks now. So this was good to read.

    I use the YTW often and lately I’ve been putting my athletes in prone position on the ground. I feel like it keeps them more accountable to their shoulder positioning. What do you think?

    I’ve also been using the Y and T unilaterally while holding a plank. Just a variation I guess.

  16. Christie D
    Christie D says:

    I’ve always had trouble with the concept of the “Y” exercise. What I learned years ago was that this exercise was used for lower trap activity. However, I don’t think I’ve ever seen a patient do the “Y” without primary UT activity…no matter the position. Most of my patients are quite kyphotic and/or have poor shoulder mechanics anyway. My feeling is that the “Y” exercise just leads to impingement…but I am speculating. I rarely give the Y exercise…what do you think?

    • Greg Lehman
      Greg Lehman says:

      Nice observation Christie,

      Ann Cools recommends three exercises with a low UT/LT ratio: side-lying external rotation, side-lying forward
      flexion, and prone horizontal abduction with external rotation.

      Side lying external rotation was an interesting finding in this paper because you get the obvious benefits of training the rotator cuff AND Upper Trap (UT) activity is around 5% of max while the Lower Trap activity is around 51% of max.

      I wrote an article that summarizes this paper as well as some others looking Serratus Anterior and Upper Traps ratio )

      Rafael Escamilla (2009) conducted a review and found the following muscle activation levels during the exercises (Y, W and T). All values are expressed as a percentage of maximum.

      1. The Y – Prone horizontal abduction at 135 abduction with ER (thumb up)

      UT = 79, MT = 101 and LT = 97

      2. The T – Prone horizontal abduction at 90 degrees abduction with ER (thumb up)
      UT = 66 MT = 87 and LT = 74

      3. The W – Prone external rotation at 90 abduction
      UT = 20 MT = 45 and LT = 79

      Looks like the W and the side lying external rotation might be the winner in terms of minimizing UT activity and maximizing LT activity. But I don’t know how this translates clinically. I have clinical success with the YTW but this might be in spite of myself. It might still be important to train the Upper Traps in conjuction with the lower traps in a manner that does not cause impingement or create pain since this is what we have to do in daily life.

      All the best,


      • Mike Reinold
        Mike Reinold says:

        Nice reply Greg, The Y definitely has upper trap activity but is great for the lower trap and cuff. It’s all a give and take. The Y is pretty important to me.

        • Andy Masis
          Andy Masis says:

          The purpose of these exercises is to train mid and low trap function, by definition, if you are seeing a lot of UT over activation, then the exercise is being performed incorrectly … My guess is that the scapular depression is the often overlooked component. A lot of patients forget to depress their shoulders, or even when they do so, their affinity for UT activity based on their posture and their daily activities leads them to engage UT as a means of retracting their scape. Now having said that, these exercises are not the end all be all. The stability of the scapula-thoracic complex has to be paired with good mobility of the shoulder (GH) and the thoracic spine. Someone with an upper crossed syndrome … forward head and rounded shoulders, will have a predictable tightness of their pecs and lats. The biomechanical positional changes this creates on the scapulae tends to place the Mid and low traps in functionally lengthened and inhibited positions, so if someone can not perform the Y exercise (which is one of the hardest to master correctly), my suggestion would be to look at the entire sequence of function. Ultimately, the biggest challenge and what I spend most of my time with patients on is integration. We can have perfect scapular function in a joint by joint approach, but as soon as we ask someone to perform a task that involves multiple joints to provide either stability or mobility, our perfect scapular mechanics can fall apart.

          Mike, I completely agree with most of what you posted. Very well written article, and I could not agree more with the idea of focusing on single arm activities as well. Great topic and you have generated great discussion. All of these posters have great points and have stimulated others to contribute. Thank you for providing a forum that cultivates critical thinking!

  17. Parker
    Parker says:

    I am a chiropractor and our tables seem to be perfect for this. Our adjusting tables are a little low to the ground for the most part and a little expensive to add to your clinic for this purpose. However, the tables I use for rehab exercises and therapies are perfect and only run around $700.00. They have an open adjustable head piece so you can lay the patient prone and allow them to rest their head. It also narrows a little at the shoulders allowing great ROM. They can also be height adjusted.

  18. Greg Lehman
    Greg Lehman says:

    Hi Mike,

    Great post, as usual. But…I couldn’t disagree I still like these exercises when performed on a ball. Mr reasons:

    1.With good cueing you can eliminate the form problems (as you mentioned).
    2. I think you can still get full scapular ROM and I don’t think there is any need to move the arms are great deal anyway…why do we need full arm ROM if the target is our scapulothoracic musculature? The W still has full ROM on the ball. I have never had patients really move their arms for the other exercises more than 4 inches so the height of a ball was never a problem.
    3. From my EMG work with the physio ball and my interpretation of the literature I would not be too concerned with the Potential (I am not sure it would occur) lack of force output. We did not see any decrease in muscle activity in the scap muscles with similar exercises on a swiss ball so there would still be a comparable training stimulus to a bench. As for training the core on the ball…these muscles hardly work on the ball. The ball is useless for for that.
    4. I like your ideas on the unilateral function. Can’t say boo about that.By accident, I sometimes have patients do these on the end of their beds with one arm when they don’t have Swiss ball.

    Keep up the insights.


    • Mike Reinold
      Mike Reinold says:

      Greg, don’t forget that these exercises are not just scapula exercises, they are shoulder exercises, so ROM on the swiss ball is limited in that sense.

  19. Barry Wrench, PT, DPT
    Barry Wrench, PT, DPT says:

    We use the Saunders total back machine/apparatus for these and any prone exercises. It’s not very wide so as not to limit shoulder motion, yet provides correct positioning that can be adjusted on the fly for a variety of activities. I know that personally I would not be able to live without mine.

  20. Danielle
    Danielle says:

    So I hate to admit this but I have never heard of the YTWL shoulder exercise however, I may have had patients perform these exercises. Would anyone want to share a brief description?

  21. Stephen Thomas, PhD, ATC
    Stephen Thomas, PhD, ATC says:

    Nice topic Mike. I’ve commonly performed these exercises bilaterally with a towel roll under the patients forehead. If the exercise is performed correctly with scapular retraction/depression prior to shoulder extension then by using the towel roll method their body will be in correct postural alignment (ears, AC jt, greater tuberosity, lateral malleolus). Performing these exercises bilaterally will also increase force output due to the bilateral transfer effect. You could try this if your have one of those spring loaded grip strength devices. Just squeeze as hard as you can and see how far you can squeeze then try it again but while you are also squeezing device squeeze your other hand into a fist. It is shocking how much further you can go. This same effect will occur when training the scapula bilaterally. Just my two cents.

  22. Dan
    Dan says:

    Why not use a towel roll under the forehead with the unilateral exercises instead of turning your head?

    • Mike Reinold
      Mike Reinold says:

      Dan, you certainly can. I find that the towel roll causes a bit too much cervical extension, sort of jamming it, for some people. This could cause apprehension or muscle activation, but if they can do it, absolutely!

      • John Buttari
        John Buttari says:

        I recently joined the inner circle and getting to this post late. First, agree great discussion! I really like the idea of shortening the neck muculature to address ms imbalances for the levator scapula and UT. But for clarificaiton do you not recommend to turn head away from exercising arm, secondary to levator scapula shorten & hyper active and UT long & inhibited. By turning head to the opposite side you lenghten the LS & in an effort to inhibit it and Simultaneously shorten UT & increase activity? Thanks for your time.

  23. Kevin Huey
    Kevin Huey says:

    I’ve played around adapting this exercises for use on the pool deck with the swim team I coach. Swimmers are prone on the deck or kickboards and cued to move through the various positions. Biggest issue has been the upper trap over-activation – Thanks for the idea of switching to unilateral performance. The only problem will be keeping a group of 20 kids focused long enough to work through all the positions. Any thoughts on what a minimally effective set/rep scheme would be for improving scapular/RC stability? Thanks

    • Mike Reinold
      Mike Reinold says:

      Good point Kevin and good adaptation to swimmers! Sometimes I will cycle through 1×10 of each exercise a couple of times, have found that helps with boredom of doing each exercises 2×10 and then moving to the next, for example.

  24. Derek
    Derek says:


    Could one not perform the “T” in a similar fashion as the “W” but obviously with a different set up i.e. more shoulder flexion/elbows extended? Great article!

    • Mike Reinold
      Mike Reinold says:

      Sure, you can, but every change in position changes the contributing muscles. The YTWL formations are all good at hitting their specific musculature.

  25. Walt
    Walt says:

    Having head and shoulders off a table implies the table is simply too wide. Why not put a bench up on two aerobic blocks? The bench is skinny enough so the person can train bilaterally. Plus, the aerobic blocks get the bench off the ground enough so even people with long arms can perform reps with a full range of motion without dragging their knuckles on the ground.

    • Mike Reinold
      Mike Reinold says:

      Walt, good thoughts. A bench itself isn’t the answer as it won’t be high enough to perform the exercise with full range of motion. But adding the aerobic blocks could solve that problem, for sure. This would probably work. I still have found in my experience that performing them bilaterally still causes a lot of upper trap activity as it is harder to stabilize your head. But maybe with proper coaching and cues you can do this well.

Trackbacks & Pingbacks

  1. […] Why i do not like ytwl shoulder exercises | rotator cuff […]

  2. […] Why I Do Not Like YTWL Shoulder Exercises ( […]

  3. […] Your Rotor Cuff5 Reasons Direct Rotator Cuff Exercises are Necessary in a Strength Training ProgramWhy I Do Not Like YTWL Shoulder Exercises div.socialicons{float:left;display:block;margin-right: 10px;}div.socialicons p{margin-bottom: […]

  4. […] Why I Do Not Like YTWL Shoulder Exercises ( […]

  5. Unilateral YTW Shoulder Exercises by Dosage (EMG amplitude) « says:

    […] post by Mike Reinold has raised some concerns about performing these exercises bilaterally (click here).  Mike suggests performing the exercises unilaterally as bilateral performance may lead to some […]

  6. […] Click here and tell me what you think. Blog this! Digg this post Share on Facebook share via Reddit Share with Stumblers Tum­blr it Tweet about it Sub­scribe to the com­ments on this post Print for later Tell a friend […]

  7. […] Why I Do Not Like YTWL Shoulder Exercises – Here’s a great post from Mike Reinold – and I agree with him 100%. […]

Comments are closed.