Category Archives: Glute Training

Which variation produces greater glute activation: barbell hip thrusts, American hip thrusts, or band hip thrusts?


  1. The barbell hip thrust trumps the American hip thrust and band hip thrust in mean upper gluteus maximus activation (70% versus 57% and 49%)
  2. The barbell hip thrust trumps the American hip thrust and band hip thrust in peak upper gluteus maximus activation (172% versus 157% and 120%)
  3. The American hip thrust trumps the barbell hip thrust and band hip thrust in mean lower gluteus maximus activation (90% versus 87% and 79%)
  4. The barbell hip thrust trumps the American hip thrust and band hip thrust in peak lower gluteus maximus activation (216% versus 200% and 185%)
  5. The American hip thrust leads to slightly (non-significantly) higher hamstring activity than barbell and band hip thrusts and slightly (non-significantly) lower quadriceps activity than barbell and band hip thrusts
  6. Around 85% of subjects received the highest mean upper glute activation when performing the barbell hip thrust (8% for band hip thrust and 8% for American hip thrust)
  7. Around 77% of subjects received the highest peak upper glute activation when performing the barbell hip thrust (15% for band hip thrust and 8% for American hip thrust)
  8. Around 46% of subjects received the highest mean lower glute activation when performing the barbell hip thrust (15% for band hip thrust and 39% for American hip thrust)
  9. Around 39% of subjects received the highest peak lower glute activation when performing the barbell hip thrust (31% for band hip thrust and 31% for American hip thrust)
  10. The aforementioned variability helps explain how some lifters feel their glutes working to a greater degree in one particular variation compared to another, despite their similar average mean and peak values
  11. Due to the nature of incremental band loading, it was difficult to obtain true 10RMs for the band hip thrust, therefore the band hip thrust probably produces slightly higher EMG values than what was reported in the study
  12. Since muscle activation is similar between the different hip thrust variations, comfort, goals, and logistics should be taken into account when determining the optimal hip thrust variation to employ/prescribe

Barbell hip thrust

Similar Articles:

Click HERE to see how the hip thrust compares to the back squat

Click HERE to see how full, front, and parallel squats compare to each other

* Special thanks to Andrew Vigotsky, Brad Schoenfeld, Chris Beardsley, and John Cronin for their stellar contributions to this research


Which variation produces greater glute activation: barbell hip thrusts, American hip thrusts, or band hip thrusts?
By Bret Contreras

Just today, my team got THIS study published ahead of print in the Journal of Applied Biomechanics. We looked at mean and peak upper glute max, lower glute max, biceps femoris (hammies), and vastus lateralis (quads) during the barbell hip thrust, American hip thrust, and band hip thrust. In case you’re not aware of the different variations, check out this video below:

Here are the tables from the study:




Here are two graphs that didn’t make the study (reviewers tend to think having tables and charts is redundant).


Mean EMG Amplitudes (BBHT = barbell hip thrust, AHT = American hip thrust, BHT = band hip thrust, UGM = upper gluteus maximus, LGM = lower gluteus maximus, BF = biceps femoris, VL = vastus lateralis)


Peak EMG Amplitudes(BBHT = barbell hip thrust, AHT = American hip thrust, BHT = band hip thrust, UGM = upper gluteus maximus, LGM = lower gluteus maximus, BF = biceps femoris, VL = vastus lateralis)

To read the full paper, click HERE (this study is published ahead of print so it’s not formatted yet, hence the ugly appearance).

A Multiplanar Approach is Best for Developing the Glutes

Just over a year ago, I wrote one of my most popular articles titled, “Do More than Just Squat.” In the article, I outlined my belief that gluteus maximus hypertrophy cannot be maximized by any one exercise. For some strange reason, an alarming amount of coaches (might I add that these coaches never have any testimonials showing dramatic glute transformation pictures) will advise women seeking maximum glute development to “just squat.” I believe that this is the worst advice imaginable, which I already wrote about in the linked article. For the record, if you want maximum pec growth, you shouldn’t “just bench,” – you should also perform incline presses and single joint horizontal adduction movements, and if you want maximum back growth, you shouldn’t “just deadlift” – you should also perform chins, pulldowns, and rows. This is common sense just as is the case with glutes and squats – you should also perform hip thrusts and various glute isolation exercises.


The squat is a great glute exercise for many people. As is the deadlift. As is the lunge. As is the good morning. However, these are all vertically-based hip extension exercises. As a personal trainer, when I started implementing more horizontally-based hip extension exercises such as hip thrusts and back extensions in concert with vertically-based hip extension exercises, my clients started seeing markedly faster and greater results in the glute department. But the glutes aren’t just made for hip extension; the glutes do four distinct things:

  • Hip extension (in varying angles of hip abduction and hip external rotation)
  • Hip abduction (in varying angles of hip flexion)
  • Hip external rotation (in varying angles of hip flexion)
  • Pelvic posterior tilt (PPT is actually similar to hip extension as far as the head of the femur and the acetabulum are concerned)

For this reason, my training programs always include a variety of glute exercises. We make sure to progressively load up our hip thrusts (horizontally-based hip extension) and squats and deadlifts (vertically-based hip extension) for increasing mechanical tension. We employ single leg exercises (vertically-based hip extension) for muscle damage. We employ back extensions (horizontally-based hip extension) and various lateral band exercises (hip abduction and hip external rotation) for metabolic stress. We add some posterior pelvic tilting at the top of our horizontally-based hip extension exercises to increase mechanical tension at the range of motion that maximizes glute activation. We make sure to heavily work the upper glutes and the lower glutes alike. And we use the “same but different” concept by always focusing on the same movement patterns but employing just enough variety to prevent stagnation and keep the gains coming.

Is there any evidence that this approach is superior to the traditional approach of just squatting and possibly deadlifting? Well, I like to think that my extensive Testimonials page provides good evidence, along with my Twin Experiment. However, a brand new paper also sheds some light on this topic.


Just this past week, in HOMO – Journal of Comparative Human Biology, an article was published ahead of print titled, “The cross-sectional area of the gluteus maximus muscle varies according to habitual exercise loading: Implications for activity-related and evolutionary studies.”

The study examined the gluteus maximus size in 6 groups of people. Here’s how the authors described the subjects:

“Several different sports competitors were represented in the sample: 9 volleyball players, 10 high-jumpers, 9 soccer players, 10 squash players, 17 power-lifters, 18 endurance runners, and 18 swimmers. These sports were classified into five categories based on training history: high impact (volleyball players and high-jumpers), odd impact (soccer and squash players), high magnitude (power-lifters), repetitive impact (endurance runners), and repetitive non-impact (swimmers) loading. Definitions of these exercise loading types take into consideration both typical sport performance as well as the typical training regimen, which together establish the exercise loading type of the given sport. Playing volleyball and high-jumping include either maximal jumps and leaps or high impacts from a specific direction during typical sport performance and training, and these sports were considered to represent high-impact loading. Soccer and squash are sports that include rapidly accelerating and decelerating movements and quick turns of the body and the hip region to which they are not normally accustomed; these sports were considered to represent odd-impact loading. The difference between high impact and odd impact loadings was based on the idea of unusual loading direction. Powerlifting involves precise co-ordination of movement coupled with intense muscle force production, and this sport was considered to represent high-magnitude loading. Long-distance running, an endurance sport that includes a great number of repetitive weight-bearing impacts with the ground was considered to represent repetitive impact loading. Swimming is also an endurance sport with a great number of repetitive movements but lacking ground impact; this group was thus considered to represent repetitive non-impact loading”

Here is what the authors found:

Chart II

Chart III

As you can see, powerlifters indeed have markedly larger gluteus maximi than controls (controls still exercised 3 hours per week so they weren’t sedentary), and they had the highest isometric strength. But so did soccer/squash players, as did volleyball players/high jumpers (this group had the highest dynamic strength). Moreover, the gluteus maximi were pretty similar between these three groups, indicating that heavy loading via vertically-based hip extension exercises (powerlifting) isn’t the only way to build the glutes. Various sports have you accelerating and decelerating the body in the vertical, horizontal, lateral, and rotational vectors. These features likely account for the similar gluteus maximi sizes despite the markedly lower loading magnitudes.

Check out the endurance runners and swimmers – their gluteus maximi were not much higher than the controls. The swimming group was training 17 hours per week and still this was not enough to significantly enlarge the glutes. In other words, running and swimming do not build the glutes.

In my opinion, these results do not imply that you have to play sports to maximize glute size; they imply that you should work the glutes in multiple vectors and include a variety of hip extension, hip abduction, and hip external rotation exercises in your training. Sports, sprints, and plyometrics can be risky, but resistance exercises are controlled and predictable and contain a concentric and eccentric component. This study shows that simply activating the glutes to low degrees and performing mundane, repetitive hip extension tasks (running swimming) does not yield glute growth; progressive overload via heavy resistance training and explosive training are needed to significantly enlarge the glutes.

It should be pointed out that this study didn’t examine longitudinal changes in glute size through training…it is possible that individuals with naturally larger glutes gravitate toward sports.

Here are the authors’ conclusions:

“To conclude, muscle activation alone does not result in or require greater muscle size because large muscles are not a prerequisite for endurance type exercise. On the contrary, it seems that rapid and powerful movements favour larger gluteus maximus muscles. Therefore, according to our results, the greater size of M. gluteus maximus is likely to have evolved as a requirement for sprinting and/or jumping movements (for powerful hip extension and stabilizing the trunk against flexion) instead of endurance running or walking. The results of this study demonstrate that active female athletes possess better-developed cranial portions of M. gluteus maximus for high impact, odd impact or high magnitude loading than do a group of less active controls. Moreover, those athletes involved in sports requiring rapid lower limb movements and stabilization of an erect torso possess greater development than do those performing repetitive impact (endurance runners) and repetitive non-impact (swimmers) loadings. It seems likely that sprinting and climbing behaviours would facilitate food acquisition– hunting in addition to aboveground gathering of resources – and both would be implicated in fleeing from danger. Those sports requiring stabilization of an erect torso to provide a stable platform for useof the upper limbs, in this case squash in odd impact group, suggests an equally important evolutionary adaptation favouring development of the cranial portion of M. gluteus maximus. Conversely, endurance running and walking, which are often associated with hunting, do not seem to be sufficient in themselves to account for the development of the cranial portion of M. gluteus maximus.”

In summary, do more than just squat and deadlift for maximum gluteal development; make sure to incorporate glute exercises that have you extending the hips, abducting the hips, and externally rotating the hips. This is what we do in Strong Curves and Get Glutes, and it’s what my Glute Squad does out of my Glute Lab. It’s what my client Erin did to win her first overall in bikini. It works for me/them and it’ll work for you.



Squats Versus Hip Thrusts Part IV: Transfer to Performance

Quick Summary:

  • Hip thrusts are markedly more effective than front squats at improving sprinting acceleration, max isometric midthigh pull force, and max hip thrust strength in male teenage athletes. They also outperform the squat in horizontal jumping in male teenage athletes
  • Front squats are markedly more effective than hip thrusts at improving vertical jump and max front squat strength in male teenage athletes
  • Functional transfer from resistance training to performance is not based on what an exercise looks like but rather a variety of factors including the direction of the force vector, the torque angle curves and accentuated ranges of motion, levels of muscle activation elicited, and more
  • In teenage male athletes, hip thrusts build half as much strength as front squats at front squatting and front squats build half as much strength as hip thrusts at hip thrusting, indicating that various hip extension exercises transfer over to each other quite well
  • Future research is needed to compare the effects of squats versus hip thrusts (and versus deadlifts) in other populations. Different periodization protocols and training durations should be examined and a comprehensive battery of performance outcomes should be tested. Finally, hypertrophic adaptations should be examined in future research as well

Hi Fitness Friends! This is part IV of a 5-part series on squats versus hip thrusts. The data from this series comes from my doctoral thesis, which should hopefully be posted online for anyone to read next year (assuming I pass my defense in December…wouldn’t it be hilarious if I hyped this up and then failed my defense and PhD?). Parts I and III look at mechanistic data, namely what happens when you perform the two exercises while wearing electrodes or while on top of a force plate. Parts II and IV look at what actually happens following a 6-week training protocol. Part V will summarize the findings and point out limitations and directions for future research. I’ll post part V over the next week or so.

And now for part IV – the findings that should make a very large impact in strength & conditioning practices around the world. Speed is king in sports. Don’t believe me? Watch this rugby video:

There’s certainly much more to sports than just speed; Usain Bolt couldn’t simply dominate every ground sport imaginable without mastering other aspects of each individual sport. However, if you can run like the wind, you have a huge advantage in sports like football, soccer, rugby, baseball, lacrosse, and obviously track & field. Until now, the vast majority of coaches believed that the best exercise for improving sprinting speed is the squat. But the squat does not maximally activate the glutes, and it poorly activates the hamstrings, which are known to be the most important muscles in sprinting. Moreover, the squat works the muscles effectively way down deep in a flexed position, but not so much in an upright position which is characteristic of applying force into the ground while running.

I have been saying this for years (that hip thrusts seem better than squats for the purpose of improving acceleration and speed), as have other coaches. However, our speculation has been met with considerable hostility in the fields of S&C and T&F. In the past, all we had was theory, but now there is some evicence. Three months ago I alluded to this study HERE, and it’s the first to lend some support to the claim that the squat isn’t the best exercise for maximizing sprinting acceleration.

The study was 6 weeks in duration and consisted of two training sessions per week. If you think about it, 12 total training sessions is nothing. I was worried that it wouldn’t be long enough to elicit any significant improvements. Luckily, my fears were unfounded, at least for this population of male teenage athletes. One group performed just front squats for lower body for all 12 sessions whereas another group performed just hip thrusts for lower body for all 12 sessions in a periodized fashion that started out with sets of 12 reps and finished off with sets of 6 reps.

Pre and post testing measured 10m acceleration, 20m acceleration, vertical jump, horizontal jump, maximum isometric midthigh pull, 3RM front squat, and 3RM hip thrust. Here are the results of the study:


Click on the image to enlarge

As you can see, the “Force Vector Hypothesis” does seem to be legit in that front squats better improved vertical jump and hip thrusts better improved acceleration and horizontal jump. The twin experiment detailed in part III also lent support to the force vector theory.

One performance measure that didn’t adhere to the force vector hypothesis was the isometric mid-thigh pull (iMTP). The iMTP is a vertical task, but the front squat didn’t improve this task. In contrast, the hip thrust improved it very significantly. This is likely due to the ranges of motion stressed in the two lifts; front squats work the muscles down deep in hip and knee flexion, whereas hip thrusts work the hips more in greater hip extension.

This could infer that hip thrusts are better suited than front squats for improving deadlift lockout strength, and one thing I failed to mention in the twin study was that following the 6-week DUP protocol, Ashley (the hip thrust twin) was much stronger than Cassy (the squat twin) at deadlifting. I surmise that they were dead equal at deadlift strength prior to the training regimen and that it was the exercise performed that accounted for the improvements, but I didn’t measure deadlift strength pre and post training so this is pure speculation on my part. Future research should examine the transfer of squats versus hip thrusts to deadlift strength, and possibly even break it down into isometric deadlift strength at lift-off, just below or above the knees, and at lockout.

This image shows the differences in position between an iMTP and a deadlift lockout, courtesy of THIS study.

This image shows the differences in position between an iMTP and a deadlift lockout, courtesy of THIS study.

Many powerlifters lately seem to suggest that pure specificity is the only way to improve a lift, and they tend to ignore the potential transfer from one exercise to another. While it is obvious that performing the specific lift is the wisest and most efficient way to improve a lift, many coaches have indeed noted transfer from one lift to another and therefore employ assistance lifts to strengthen muscles and improve sticking regions.

This study showed that an athlete that performed the hip thrust gained half as much front squat strength as an athlete performing the front squat, and an athlete that performed the front squat gained half as much hip thrust strength as an athlete performing the hip thrust. This is very important as it could mean that maximizing powerlifting strength requires the utilization of assistance lifts, and it likely means that strength on one lift can be maintained to a greater during times of injury by performing other lifts. For example, an athlete with a low back or knee injury might be able to maintain much of his or her squat or deadlift strength by performing hip thrusts if well tolerated. This is why it’s so important to train around injuries and not through them.

I’ll expand on the limitations of this research in the final part of the series (part V), so please stay tuned for that in the next week.

Forward thinking strength & conditioning coaches should seriously consider 1) making the hip thrust a staple in their programming, if they haven’t already, 2) possibly centering their programming around hip thrusts and making it the first exercise performed during the day and week, 3) possibly periodizing training around maximizing vertical and horizontal force production performance, 4) consider having separate stations dedicated solely to hip thrusting so that benches and power racks aren’t being used up.

However, more cautious strength coaches could indeed make a mental note of these findings and wait for future research to duplicate the findings on athletes from other sports, or female athletes, or more highly trained individuals.

Please note that this research was carried out by New Zealand strength coaches; I was blinded from the training and the testing. Also please know that I wanted to test the back squat and not the front squat, but the head coach was more comfortable having his teenage athletes front squat, for which I was very understanding of. It’s important for researchers to be flexible when working with strength coaches during training studies; rigid individuals are unable to collaborate and get any research done. I intend on submitting this research to a sports science journal and getting it published, so hopefully you’ll have a more detailed report pertaining to this data at your disposal in the next 6 months or so.

Dani Shugart

Dani Shugart

Squats Versus Hip Thrusts Part III: Forcetime Data

Quick Summary:

  • Squats involve almost twice as much range of motion (ROM) in terms of barbell displacement than hip thrusts, in addition to more time under tension at identical rep ranges
  • The force output during the eccentric phase during squats is only around 10% lower than it is during the concentric phase, whereas the force output during hip thrusts is almost 3X greater during the concentric phase compared to during the eccentric phase
  • Individuals tend to let gravity lower the bar during hip thrusts, whereas much more muscular output is involved during the lower phase when squatting
  • The hip thrust generates greater concentric force output than the squat, whereas the squat generates greater eccentric and total force output than the hip thrust
  • Due to the greater ROM, TUT, and eccentric force output involved in squatting, the squat produces greater total work, total impulse, and total power outputs than the hip thrust
  • The squat is a large ROM concentric and eccentric exercise, whereas the hip thrust is a short ROM mostly concentric exercise
  • Future research should be undertaken to confirm these findings; to split each variable of interest into concentric and eccentric components; to test different styles of squatting and hip thrusting; to determine how anthropometry effects forcetime data, and to determine how these variables of interest impact neuromuscular adaptations

Hi Fitness Friends! This is part III of a 5-part series on squats versus hip thrusts. The data from this series comes from my doctoral thesis, which should hopefully be posted online for anyone to read next year (assuming I pass my defense in December…wouldn’t it be hilarious if I hyped this up and then failed my defense and PhD?). Parts I and III look at mechanistic data, namely what happens when you perform the two exercises while wearing electrodes or while on top of a force plate. Parts II and IV will look at what actually happens following a 6-week training protocol. Part V will summarize the findings and point out limitations and directions for future research. I posted part I three weeks ago and part II two weeks ago, and I’ll post parts IV andV over week or so.

This study portrays an awesome aspect of learning as a researcher as I ended up being wrong and changing my stance. My PhD supervisor John Cronin wanted me to carry out force plate research on the squat versus the hip thrust as part of my thesis (I didn’t want to at the time we planned it). Several years ago, I wrote an article about how silly I think forcetime data is with regards to optimal exercise selection (see HERE). Since then, my stance has softened, primarily due to the process involved in the present study I’m discussing – one can indeed glean valuable and practical information from performing exercises on a force plate and analyzing the data. However, longitudinal training studies are needed to test and validate any hypotheses that are generated from the mechanistic forcetime data.

Prior to the study, I was talking on the phone to my colleague Andrew Vigotsky. Here’s what I said to him. “Andrew, this study is so stupid. People are stronger in the hip thrust compared to the squat. The hip thrust is also the more explosive lift. Since force equals mass times acceleration (f = ma), the hip thrust will generate greater force. And since force output lays the foundation for work since work equals force times distance (w = fd), impulse since impulse equals force times time (i = ft), and power since power equals force times velocity (p = fv), the hip thrust is going to kick the shit out of the squat in every category. This study is merely a formality to show the obvious.”

Turns out I was quite wrong. Had we just looked at concentric data, I might indeed have been right (well, definitely for force and probably for power, but maybe still not right for work and impulse). However, what I failed to consider ahead of time was the eccentric phase and how this impacted total outputs that combined the concentric and eccentric phases. This isn’t the first time I’ve been wrong as a scientist and it sure won’t be the last time. What’s important is that I update my knowledge-base, inform my followers about the truth, and learn from the experience (which I’m doing). Now let’s talk specifics.

We had ten fairly strong dudes with an average of around 7 years of resistance training experience and a 10RM of around 216 lbs in the squat and 252 lbs in the hip thrust perform the two lifts on a force plate. Luckily, I spoke to badass UK biomechanics researcher Jason Lake prior to collecting my data – he was adamant about having the subjects hover silently on the force plate for a moment before starting their sets and performing their reps. This allows for the system load to be cancelled out when analyzing the data so that only the effects of muscular effort can be examined (we want to know what the muscles do, not what gravity + muscles do in concert with the body + barbell system).


Here are the results:


When I got the data, my initial instinct was like, “This is all off, there’s no way this is correct, something screwy occurred with the software.” At this point, I hadn’t looked at barbell displacement, time under tension, or concentric force, so the data didn’t make sense to me. I pondered what I knew about biomechanics and what factors could have impacted the data to come out the way they did. I decided to have my assistant check out barbell displacement, and I was shocked to find out how much greater ROM was used in the squat compared to the hip thrust.

I recall back when I invented my Skorcher, I tested barbell ROM in the hip thrust and it wasn’t much different compared to a parallel back squat or sumo deadlift. But the Skorcher involves more ROM than the traditional hip thrust because the feet are also elevated and the dorsiflexion that occurs throughout the concentric ROM translates into greater vertical bar displacement (not the case in the standard hip thrust off of a bench). In addition, we examined the full squat in this study, not the parallel squat. Hip ROM isn’t drastically different in a squat compared to a hip thrust depending on how the two lifts are performed and depending on the height and anatomy of the lifter, but since there’s not much knee ROM in a hip thrust, the total ROM in a squat is much higher. So this makes sense indeed.

Nevertheless, I went out to my garage and performed a set of barbell hip thrusts with a meter stick lined up with the barbell – turns out the forceplate/motion capture data was legit and my assumptions were off-based. I also had my assistant check out time under tension, and it wasn’t surprising to see that the squat took more time due to slower lowering speeds and more ROM.

Next, I asked my assistant to split up the concentric and eccentric force phases, and this is when I discovered how glaringly different the force outputs are during the lowering phases of the two lifts.

Between the differences in displacement, time under tension, and eccentric force, all of the data make perfect sense. Now, one could astutely point out that lifters can indeed lower the barbell slowly during a hip thrust if instructed to do so. What’s great about this study is that we examined what people naturally do during the two lifts. Future research can examine how the effects of hip thrusts with controlled eccentric phases impact forcetime data, or how higher box heights (this study used a box that was around 16″ high) impact forcetime data, or how anthropometry impacts squat versus hip thrust forcetime data, etc.

You’ll note that we didn’t separate all variables into concentric and eccentric phases. This is because my thesis deadline came to an end and we ran out of time. Had we had more time, I surmise that the hip thrust has greater concentric power outputs than the squat, but I’m not sure about concentric work and impulse. Everybody in S&C loves to talk about eccentrics, but it’s important to note that acceleration sprinting is mostly concentric in nature. This could impact training adaptations.

To wrap things up so far, Part I demonstrated a clear advantage of the hip thrust over the squat in terms of EMG activity. Part III (this article) demonstrated a clear advantage of the squat over the hip thrust in forcetime data (force, work, impulse, power) and displacement. Part II provided some clues as to how the two lifts differ in terms of actual training adaptations. In part IV, we’ll look at the effects of an actual training study in performance. These are vital in S&C as they show what DOES happen, not what SHOULD happen based on humans’ often limited and biased opinions pertaining to acute mechanisms, sensations, and transfer of training theories. In part V, I’ll wrap it all up and summarize my findings.