The Best Exercises for the Gluteus Maximus and Gluteus Medius

In a past article, I discussed assessing and treating dysfunction of the gluteus medius.  I reviewed an article from the Journal of Strength and Conditioning Research and the author’s recommendations.  Taking this information one step further, a recent article in JOSPT has quantified electromyographic activity of the gluteus maximus and gluteus medius during common exercises.  This information is helpful when deciding which exercises to perform in your patients or clients.

Based on the results of this study, we can identify exercises that produce the highest amount of EMG activity.

Gluteus Medius

  • imageSide-lying hip abduction – 81%
  • Single limb squat – 64%
  • Lateral band walk – 61%
  • Single-limb deadlift – 58%

image  image  image

Gluteus Maximus

  • Single-limb squat – 59%
  • Single-limb deadlift – 59%
  • Sideways, front, and transverse lunges – 41-49%

In addition to the manuscript, there are good video demonstrations of the exercises and a PowerPoint presentation available at the JOSPT website.  When I accessed this, I believe they were all available for free.

Clinical Implications

Based on the results of this article, here are a few things that came to my mind

  • Side-lying hip abduction should be used in all people needing glut medius strengthening.  EMG activity was almost 20% higher than the next exercise.
  • The single limb squat and single-limb deadlift exercises activated high amounts of EMG activity for both muscles
  • image The clam exercises produced EMG activity between 34-40% for both muscles.  While this is low in comparison to other exercises, the authors did not use resistance during testing.  I would still use this, especially with a resistance band around the thighs, as am early-stage or activation exercise. The authors also compared clams at 30 degrees and 60 degrees of knee flexion and showed no different in gluteus medius activity.
  • The lunge exercises produce a moderate amount of EMG activity and are likely good early-stage exercises to progress to prior to the single-leg squat and deadlift exercises

DiStefano, L. (2009). Gluteal Muscle Activation During Common Therapeutic Exercises Journal of Orthopaedic and Sports Physical Therapy DOI: 10.2519/jospt.2009.2796

29 replies
  1. GD
    GD says:

    The side-lying hip abduction is clearly a good choice, but it is painful for me to do so. What does this indicate? Gluteal tendinopathy? GT bursitis?

  2. Laura
    Laura says:

    HI Mike, I came across your website while preparing for a department inservice and was so glad I did. I work in an inpatient rehab setting with geriatric, orthopedic, and neurological diagnoses so I definitely come with a different perspective vs. treating the sports medicine population. However, our common goal as physical therapists is ultimately the same: FUNCTION.

    As a certified PNF therapist I am very interested in movement/gait analysis and especially the role of the hip/pelvis in stability during gait and other functional activities. Although I feel confident in my abilities to provide hands on treatment to recruit hip and trunk stability, I would like to implement more exercises for patients to perform on their own to target this area.

    I was wondering if you had read the article, EMG Analysis of the 3 Subdivisions of the Gluteus Medius during Weight Bearing Exercises ( In this article it is stated that weight-bearing strengthening exercises have been shown to produce significantly higher GM activity in comparison to non-weight-bearing exercises. This seems like conflicting evidence as to the findings in your discussed article.

    The 3 exercises analyzed are the single leg wall squat, pelvic drop, and wall press. You can also read about these exercises at, where the author references the same article as you discussed. I have began utilizing the pelvic drop with some of my patients as it makes good sense to me – the patient holds and slowly lowers/raises the pelvis while standing with opposite leg on a 2 inch block. This coincides with gait as our pelvis drops 2 inches during the gait cycle, which is controlled by contralateral gluteus medius. This exercise is also the easiest out of the 3 for my patient population to perform with success, many who are in their 70 to 80’s. I am a big advocate of functional/movement reeducation exercise and activities whenever possible.

    Is this an exercise you commonly give patients? It would be nice to know how these 3 weight bearing exercises compared to non weight bearing exercises in terms of EMG activity, especially the SLHA. Unfortunately I do not have access it seems to many of the given references/journal publications to try to put the pieces together.

    I would love to know your thoughts on this. Thanks!

  3. Adam
    Adam says:

    Mike- with regard to the resistance band lateral walk: despite the high glute medius EMG, how do you feel about its potential to facilitate the TFL? Most people do them in a slight squat, toes dead forward. This seems like a position that might bring in more TFL than we’d like. Thoughts on performance of this exercise? It’s one we all have done…

    Also, how about cable resisted abduction with slight extension and external rotation? These seem to work for me, but do you think they’re similar/better/worse than the side-lying abduction?

  4. Carrie
    Carrie says:

    I took a course on patella femoral pain. They said the sidelying hip abduction exercise does recruit a lot of glut med, but that it also recruits too much TFL and ends up causing more PFP. What do you think?

  5. Mike Reinold
    Mike Reinold says:

    @ Greg – interesting point, I would have to say I agree. I wonder if there was a difference in MVIC testing that can be attributed? Anyway, nice site by the way, I will be reading. Best,

  6. Greg Lehman
    Greg Lehman says:

    Hi Mike,

    Love your blog. I just wrote a post about the article you discuss and it does not pass the smell test to me. The design looks fine but 82% MVC for sidelying hip abduction is a bit high. Two other groups before (Bolgla'05 and Ekstrom '07) show activity between 39 and 42%. In my lab we were showing even less.

    There is just something odd about these results and I would be a little skeptical.

    My full post is at


    Greg Lehman
    Physio, Chiro and reformed Biomechanics researcher

  7. Anonymous
    Anonymous says:

    Has anyone seen glut medius syndrome in young female athletes. signs/ symptoms: Trendelenburg gait/ weak glut medius/ pain in the supra-iliac crest/ low back area. Treatment? Progression?

  8. Ravi Lescher
    Ravi Lescher says:

    1) Can we surmise from this research that an appropriate gluteus medius progression might be (as an example): clamshells with Tband, crabwalk with Tband, ending with sidelying hip abduction (based on the MVIC%)?

    2) To answer Amy: it's always hard to answer a question like this without knowing exactly the treatment you've had in the past (what exercises, etc). However, I've found that it is very easy to develop compensation patterns using the TFL/ITB and deep hip external rotators, in essence, getting around strengthening the gluteus medius. It's important to know first exactly WHERE you should feel the muscle contraction as you're doing the exercise (biofeedback can help with this), stretching regularly, self-massage with tennis ball, and backing off strengthening temporarily if increased symptoms, and bodymechanics training (not allowing your knee to fall inward from you hip every time you stand up/down, do stairs, squats, or cycling. Hope this hellp. Good luck.

  9. Amy
    Amy says:

    I am writing as an injured athlete and former runner/triathlete. I suffered a tear in my gluetus medius 5 years ago. It was diagnosed by MRI and then 1 year later showed guteal thickening, but no longer a tear (a second MRI).

    My symptoms have been ongoing, intermittent chronic pain, that worsens with running and aggressive biking. I have not been able to run since the injury. I continue to cycle and do mild weight training but am unable to push myself physically, as the injury seems to worsen.

    I have sought many treatments including physical therapy, massage, stretching and strengthening, shockwave therapy, low-level laser therapy, and have consulted with a surgeon. The surgeon has strongly discouraged surgery. All of these therapies provided mild relief that was very short term (i.e., a few days).

    Recently, I read a few promising articles on percutaneous needle tenotomy. Is anyone familiar with this treatment for this injury? I live in Toronto, Ontario, Canada and am not able to find resources up here. Any suggestions? Any thoughts on further treatment?

  10. Carl Valle
    Carl Valle says:

    sprinting at top speed is vertical and the glute system acts as a postural supporter with humans. Vertical displacement is higher with faster sprinters.

    Also bridges have no eccentric action as plyos such as lateral heidens will do more for the medial glute than those JOPT exercises. Maybe for stroke patients but for a running back getting ready for D1 performance no way.

  11. Kory Zimney, PT
    Kory Zimney, PT says:

    Thanks Mike for the insight with EMG readings from a research prospective and translation into the clinic.

    Thanks for referring me back to your rotator cuff article about the change in the position changing EMG. I have always thought this just from "feeling" the exercises and thinking through biomechanics, but its always good to see research backing it up. I like adjusting exercises in this way to try and get the person through the movement to enhance some proprioceptive input but using other muscles to assist with the movement initially. As the muscle strength improves adjusting the exercise to tweak other muscles out to make the movement more isolated for the specific muscle weakness targeting.

    My thoughts behind the max contraction go back to thinking of body movement as an orchestra of movement. Sometimes some instruments need to play loud during certain portions of a song and some times they might need to play a little softer to make the music sound best. If all the instruments always play as loud as possible, doesn't mean the music will sound the best. Just something I keep in mind as I try to optimize my patients movement patterns.

  12. Dr, Wayne Button, BSc, DC
    Dr, Wayne Button, BSc, DC says:

    Anything by JOSPT usually turns out to be a good study. They even include videos online which is always great for new orthopedic test. I often use the "monster walk" or "crab walk" with my athlete patients. This can be challenged by tossing a weight ball back and fort. However, for the patient that lacks good kinesthic awareness I will often us the side lying hip abduction. Particularly for runners try exercises purposed by micheal fredickson. His methods of getting the gluts are highly functionally translatable to running purposes. Thanks for a good blog.

  13. Mike Reinold
    Mike Reinold says:


    Good questions. I'll try to answer some:

    – I would say that high EMG is a positive thing, I don't think that in the orthopedic population we are worried about tone and the negative consequences. Think of EMG this way, studies like this dont show the "best" exercise for anything, it just compares the exercises in the study. So that being said the high EMG articles in this article are more effective in eliciting contraction of the intended muscle. Important when considering between several different exercise. And as @ CHristie nicely wrote, sometimes the answer surprises us and what we consider to be "basic" exercises are actually very useful.

    – EMG does pick up different types of contractions (conc, ecc, iso). During the study, the researchers observe the EMG activity during the entire phase of the exercise and correlate the timing, so yes, we do know the type of contract by combining the EMG with kinematic data.

    – 100% MVIC is again a relative concept. All this means is that a certain exercises produces a % of the maximal isometric contraction that the muscle can perform. This is determined by using standard manual muscle testing position. It is not uncommon to have over 100% MVIC, especially if you use a suboptimal manual muscle test position. It just means that the exercise produces a certain amount of EMG in relationship to a max manual muscle test. Again, just a reference point to compare the rest of the exercises. I look at the MVIC's in relationship to the study. There could be results in the 20 and 30 % range in one study that is significant and not in others.

    – To answer you last question, yes, changing posture and other muscle recruitment will absolutely change the EMG of both the muscle you want to hit and the surrounding muscles. Think of my study comparing supra activity during full can and empty can. % MVIC of the supra were the same between the two but the full can was deemed the better exercise because the position produced less surrounding muscle activity.

    Hope that helps.

  14. Harrison Vaughan, PT, DPT
    Harrison Vaughan, PT, DPT says:

    I do enjoy these articles as they are helpful in putting evidence-based studies in treatment. Its hard to find good treatment articles to say what different exercises strengthen what we are aiming to do. Plus, I like numbers (such as percentage max contraction) as it makes sense to me and to colleagues/students.

    I know there are TONS of gluteus medius & maximus exercises out there but wonder how much more, less or same EMG activation is with pilates type exercises as side-lying hip circles and side-lying hip flexion/extension.

    I find these are another "step up" to strengthening these muscles as it is not good clinical decision to get my 65 y/o back pain pt doing single leg dead lifts! (which now I know is not any better EMG than S/L hip abd)

  15. Chad Ballard, PT
    Chad Ballard, PT says:

    Thanks for the info Mike. I like to add a diagonal reach to the single leg deadlifts to kick in the external rotators as well.

  16. Kory Zimney, PT
    Kory Zimney, PT says:

    I think these EMG muscles are helpful in determining better exercises to do with our patients to get the best results. But we need to keep in mind the "strongest" muscle does not always mean the most "functional" muscle. We know this because the strongest athlete is not always the best athlete. (The strongest golfer doesn't always hit the ball the furtherest or straightest for example) While being stronger is an important aspect, we need to remember motor control, coordination, timing and proprioceptive response issues play a large role in gaining the most "functional" muscle.

    I often wonder, maybe somebody out there can answer these for me, just because there is more EMG does this necessarily mean it is good EMG? Just because a muscle is firing does that mean it is good firing? We know in a stroke patient with high tone this is not good, we need to control that tone. So in my mind I don't always think of increase EMG as always positive EMG. Just want to know what others think? Also does an EMG pick up info the same whether the contraction is isometric, eccentric or concentric? Also for optimal function do we want to get 100% contraction? If not, why train it to get max contraction, does this confuse the body on how it might function in real life? Also when doing an exercise can I change the EMG in a muscle based on how I might use other muscles? If I do a lunge holding my hands over head, do I get as much EMG in the gluts as if I reach my hands down to my feet as I lunge?

    I know this post has more questions than any answers, but these are things we as therapist need to consider when setting up our exercises for our patients. While we need to use this EMG info of these studies, we still need to critically assess movement patterns and what exercises improve those movement patterns.

    And I would agree 100% with you Christie, we don't always need the crazy clever/"shock" factor to get the job done. We just need the "right" exercises…that a patient can and will do.

  17. Christie Downing, PT, DPT, Dip. MDT
    Christie Downing, PT, DPT, Dip. MDT says:

    I find it very interesting…that the most "basic" exercise is the one that results in the most EMG activity. I think as PTs we too often try to think of more "clever" ways to get something done. In the end, we wind up passing on some very good, basic exercises.

    We have to remember that the treatment should reflect the needs of the patient and not how "clever" it makes us look.

    …of course then we always have to transition it to functional activity…

  18. Jan
    Jan says:


    I think SLS en SLDL have more clinical effectiveness because it's more functional in daily life. The SLHA is more of an isolating exercise, which is nice to start with.

    This paper about glute training is solid stuff. I've read the article and his book. The only thing is, that he tested it on 4 people, but the rationale very reasonable. The most striking insight is the use of loading vectors. If you want to sprint, don't do squats. It's not a good glute exercise and the loading vector is axial, while sprinting is anterior/posterior. I can recommend it as a good training manual.

    Grtz Jan

Trackbacks & Pingbacks

  1. […] target the gluetus minimus, while lunges can strengthen the gluteus maximus. Whoops! Better start doing lunges, side-lying hip abduction and some clams instead of relying just on squats to give some shape. Depending on how my body handles T25, I will […]

  2. […] So they’re very important in standing up and are very involved in walking.  Here’s how to get your butt strong.  This article gives some good information but you have to be pretty functional to do some […]

  3. Core Strengthening Exercises For Women says:

    […] The Best Exercises For The Gluteus Maximus And Gluteus Medius …I find these are another "step up" to strengthening these muscles as it is not good clinical decision to get my 65 y/o back pain pt doing Best Core Exercises For Women – April 13, 2012 [] The Best Exercises For The Gluteus Maximus And Gluteus Medius …The Best Exercises for the Gluteus … Access Document […]

  4. Best Core Exercises For Women says:

    […] The Best Exercises For The Gluteus Maximus And Gluteus Medius …The Best Exercises for the Gluteus Maximus and Gluteus Medius. October 26, 2009 21 Comments Core Stability From the Inside Out; Assessing and Treating Dysfunction of the Gluteus Medius; The Best Exercises for the Gluteus Maximus and Gluteus Medius … Retrieve Content […]

  5. Best Back Exercises says:

    […] The Best Exercises For The Gluteus Maximus And Gluteus Medius …The Best Exercises for the Gluteus Maximus and Gluteus Medius. October 26, 2009 21 Comments have no eccentric action as plyos such as lateral heidens will do more for the medial glute than those JOPT exercises. Maybe for stroke patients but for a running back … Read Content […]

  6. […] to perform up to 30 repetitions of hip abduction (if able).  Isn’t Hip abduction linked as highest EMG exercise for the gluteus medius?  It would make sense to test it this way then, right?  I can get them to […]

  7. […] requests for Gluteus Medius exercises. The best, most concise, one-click site that I could find is here. Clarifications: the sidelying hip abduction (1st photo) can be done with an ankle weight. I'm a […]

Comments are closed.