Posted on

Hip Thrusts vs Squats Revisited

Barbalho KO's Contreras

Since I am American, I believe in kicking a man when he’s down.  It’s just efficiency: he’s right by your foot.  So it’s time to kick Bret Contreras a bit more and revisit the issue of hip thrusts vs squats in terms of which is superior for glute growth (and no I am NOT bringing up the other stuff since I care only about the science).

Now you might be wondering what “new” information there is to examine in the last week and a half or so.  Actually it’s not new in a literal sense.  In fact, it’s  “old” even if it will be “new” to the majority of people.  This gibberish will make sense in a second.

Old “New” Information Hip Thrust vs Squats

So you ask, what is Lyle going to address in this article?  Well I tell you what I’m not going to address and that was Bret’s “article” in the AARR as a response to Barbalho.  It’s simply not worth my time.  Between the appeal to authority (ONLY trust PhD’s to assess science), anecdotal evidence (DATA FROM AN APP) and a pathetic ad hominem attack against Barbalho, well…

Barbalho-Ad-Hominem.png

Seriously, Bret?  Seriously?  This is just sad.  Also, where did Barbalho ever claim to be a POWERLIFTING coach. It’s not only a pathetic ad hominem attack but also, well…I’m not sure which logical fallacy you’re using here in using that caption but it’s bullshit no matter how you cut it.

 

But the whole thing is pathetic as hell.  If you’re gonna play the “I’m a PhD/only listen to study analysis from PhD’s” appeal to authority, stick with science. The above is the kind of thing every bro bodybuilder dipshit in the world uses against science “Bro, do you even lift?”.

It’s just sad.  But I am not going to address it any further than that.

So what then do I have to look at?   Well to paraphrase great philosopher J. McClain (Die Hard, 1988)

Ho, ho, ho, now I have a full copy of Bret’s PHD Dissertation/Thesis.

Bret Contreras' PHD Thesis

Before I continue, let me make it clear that I didn’t have to go through any shenanigans to get this.   It’s available to anyone at the AUT thesis repository.  Incidentally, you might note who the secondary supervisor was on the thesis: Brad Schoenfeld.  He of the shoddy statistical analysis (well not shoddy so much as ignored when they don’t suit his conclusions) and unblinded ultrasound measurement.

Now it’s not uncommon to have someone outside of the university do this.  However, it is interesting given the rather closed nature of a certain current seminar circuit that this was the case (both are also on the NSCA Board of Directors).  It makes me wonder just how early the circle jerk formed….  I mean, I’m not sayin’ nothin’, I’m just sayin’.

For ease, you can also download Bret’s entire thesis here, in case you want to read it yourself or make sure that I represented it honestly.  Unlike others who like to make unsubstantiated claims (James, I’m still waiting for that reference showing BF10  <3 isn’t meaningless), I’m all about transparency.

Examining Bret’s Thesis

So let’s look at Bret’s thesis.   As is the case with PhD dissertations, this thing is a monster.  It’s 291 pages and here is the full table of contents for those who care.

Bret Contreras Thesis TOC

There is no way in hell I am going through the entire thing since it would be longer than my absurd Why the US Sucks at Olympic Lifting series.  Rather, let me briefly address the bits I don’t care about and then focus on the one piece I do care about since it’s the only one that matters.

After a few chapters of introduction, the next 4 are all EMG work.    Well EMG is interesting but surface EMG is rough at best and drawing conclusions from it is questionable.  And since Bret Contreras, Chris Beardsley, James Steele and Andrew Vigotsky penned a paper titled Greater Electromyographic Responses Do Not Imply Greater Motor Unit Recruitment and ‘Hypertrophic Potential’ Cannot Be Inferred the EMG doesn’t matter.  Certainly not in terms of the results from the recent Barbalho et al. paper on the hip thrust versus the squat.   Even Bret admits that EMG isn’t indicative of growth.  So screw it.

All I care about in the context of this article is the effect of hip thrusts vs. squats on glute growth

Chapter 8 deals with a biomechanical (kinetic) analysis of the squat and hip thrust, Chapter 9 a performance study in adolescent athletes (yay?) and Chapter 10 is on the horizontal push force test.  This gets into something called the force vector theory and, let’s face it, nobody cares.  Nor is it relevant to the debate at hand.

Again, all I care about here is the current debate over hip thrusts vs the squat and glute growth

So Chapter 11 is where my focus will lie as this is Bret’s original study that led to his claim that the hip thrust was superior to the back squat for glute growth.  First I’m going to take it apart piece by piece.  Then, for contrast I’ll compare it to the Barbalho study that got Bret’s undies so twisted.

Chapter 11: Effects of a Six-Week Squat versus Hip Thrust Program on Gluteus Maximum Thickness and Horizontal Force Production in Monozygotic Twins: A Single-Subject Design

Ok, so that is the full title of Chapter 11 in Bret’s thesis and yes, it’s a mouthful.  By the wording alone you can probably guess what it’s dealing with.   Bret wanted to test  hip thrusts vs. squats on glute growth and this was basically the paper he used to create an entire freaking industry out of.  So let’s take it apart.

Hip Thrust vs Squats: Number of Subjects

So Bret’s “study” had two subjects.  That’s not a typo, two.  Not even two per group.  two total.  One did squats and the other did hip thrust and each only did the one movement or the other.  There was no crossover, by which I mean the subject that did squats first did hip thrust next and vice versa.  One subject in each group, one movement each.  That’s it.

In defense of this design, the chapter cites a few studies claiming that:

Single-subject designs are useful for strength and conditioning and physical rehabilitation research (Kinugasa, Cerin, & Hooper, 2004; Perdices & Tate, 2009)

which is all good and well I guess.

Edit: Peter Bond in my group took the time to look up the above two papers and had this to add

If you lookup Kinugasa et al., you see it was aimed specifically at elite athletes and more importantly, they note that: ” Single-subject research designs allow us to find out the extent to which a specific conditioning regimen works for a specific athlete, (…)”
Their paper highlights how it can be used to assess how well something works for a specific athlete: the athlete in question.

This does not apply to Brets twin stuy, since it was meant to extrapolate the findings. Bret was wrong to cite this in his context. His twins were not elite atlhetes and the study was not performed to assess how well the hip thrust or squat worked for a specific twin.

The second paper Perdices & Tate, target neurorehabilitation specifically. It’s a big leap to stretch that to investigating the growth response of a muscle. Additionally, the paper lays out some stuff to strengthen your research results when you do apply a single-subject design.

Most of which Bret does not seem to have applied. But most importantly, not to miss the point: Bret was comparing hip thrust to squats between two individuals without cross-over: this paper does NOT discuss such a setup at all and thus can not support his statement in this context.

Me again: Which basically says that neither paper really apply here or act to support Bret’s decision to use this “study” design.  Is anybody really shocked at this?  Did Bret read them?  Did Cronin?  Did Brad?  Either they didn’t and just cited them or they read them, cited them and hoped nobody else would look them up.

Even if it did it would not change the fact that there was ONE subject performing each exercise with no crossover which make any conclusions very very limited.  Barbalho may only have had 11 per group but that’s an order of magnitude (ha ha) more.

Pete Holman made a big issue that the Barbalho paper only had 11 subjects in each group.
Hey, Pete, Bret’s original paper had ONE person per group.  Please discuss.

Of some interest, they were monozygotic female twins.  Monozygotic means that they are genetically identical.  And I will give Bret a point for doing it this way.

The logic here was by using twins, any individual variance in response would be avoided or minimized.  That is, we know that any two individuals on a given training program can show drastically different responses.

The work on aerobic non-responders is has found that the VO2 max increase to training can vary from roughly 0 to 40%.  In every resistance training study there are always some number of subjects that get far less of a response or even lose size For example, about half of the subjects in the Haun et al. study lost quad size but this is obscured when you present only the average response.  All it takes is one massive hyper-responder (or hypo-responder) to screw the average.

Muscle Fiber Size Changes in Haun et al.

Some recent papers on hypertrophy have had subjects do one volume with one arm and a different volume with the other.  This is in contrast to having 10 people use one volume and 10 another.  And what’s being found is that the difference between any two people is ENORMOUSLY more than the difference within any given individual.  Like 10 fold difference in results.

What is being found is that people who respond well respond well to both volumes (even if they do better with one versus the other) and the people who respond poorly respond poorly to both volumes.  And this is likely to shake up the hypertrophy research arena going forwards.

Because all of these previous studies might not have been comparing what they thought they were comparing.  Because rather than comparing volume per se, they may have really been comparing individual responses.  So if by some chance a bunch of hyper responders end up in one group that will make it look like the specific volume or whatever is superior.  When what it actually is is that the subjects are superior.

In any case, with monozygotic twins, the above is at least minimized since the genetics are the same (let’s not get into arguments about epigenetic changes please).  Claude Bouchard has done a metric ton of work showing that twins show a much more similar response to each other than they do to other sets of twins.  Note that I said similar.  Not identical.  But certainly more similar than non-twins.

So I’ll give Bret’s study a point for this.  It’s about the only decent thing about
the design but I should acknowledge it.

Importantly, the women were described as athletic (whatever that means) but were UNTRAINED in terms of resistance training.  They had never lifted before.  And this is a huge problem, one I’ve mentioned repeatedly.  Because, for the most part, untrained people get the same response almost irrespective of what you do and it’s all basically a wash.  I don’t even pay attention to studies in the untrained for that reason.  Not beyond noting that “Hey, they got the same results….again.”

And I will refer to this again and again because it impacts on an enormous number of parts of the study.  Like all of them.

Hip Thrusts vs Squats: The Study

The study was only 6 weeks long. Or perhaps 6 weeks short would be a better description. Don’t get me wrong, a lot of pilot work is short, there is a funding issue to consider, etc.  As I’ve mentioned in previous articles, at least one research group thinks that 12 weeks is the minimal length to find significant differences in hypertrophy.

There is also the issue that in untrained individuals, most of the gains from training tend to be neural moreso than actual growth.  Note that I said most.  There is also the fact that you cannot safely extrapolate results in untrained individuals to trained individuals.

The technique for each exercise was described as such

During the back squat, subjects’ feet were slightly wider than shoulder width apart, with toes pointed forward or slightly outward. Subjects descended until the tops of the thigh were parallel with the floor (Pierce, 1997).

In accordance with Contreras et al. (2011), the barbell hip thrust was performed by having subjects’ upper backs on a bench, approximately 38 cm high with a 2 cm pad. Subjects’ feet were slightly wider than shoulder width apart, with toes pointed forward or slightly outward. The barbell was padded with a thick bar pad and placed over the subjects’ hips.

So it was a parallel squat more or less although not certainly not a full squat.    The hip thrust is described fairly generally in terms of technique.  In addition to the two lower body movements, some basic upper body work was done but it was the same for both of the twins so I won’t bother with it.   The workout is given below.

Hip Thrust vs. Squats Workout

It’s just a basic daily undulating periodization (DUP) program although do note the AMRAP (As Many Reps as Possible) set at the end, something I’ll come back to below.  Loads were increased as subjects got stronger to maintain the intensity in an appropriate range. I’ve certainly seen stupider workout designs and should probably give a point here too.

Hip Thrust vs Squats: The Measurements

Before beginning the training program and afterwards, the twins underwent testing for upper and lower gluteus maximum thickness, maximum horizontal pushing force (this is the whole force vector theory thing I don’t care about) and hip thrust and squat 1 repetition maximum (1RM).

Let’s go through them backwards.

The idea of testing 1RM is rough enough but to test it in rank beginners is simply ludicrous.  I want everyone reading this to remember the first time they squatted.  Do you think that anything you accomplished as a 1RM would have been valid?  Of course not.  You’d be wobbling around more than lifting and you know it.

1RM is likely to go up day to day in a beginner just due to learning the movement.  In fact, doing nothing but practicing the testing for a 1RM improves strength in untrained individuals as well as actually training.  To test 1RM in the squat in a beginner is nonsensical.  Or rather, any result you get is utterly meaningless.

Now, for the hip thrust, I’m willing to buy a 1RM test as being relatively more accurate?  Why?  Because it’s a simple movement.  I could certainly test someone for a 1RM on the leg extension on their first day.  Maybe even a leg press.  It still wouldn’t be a true 1RM because you have to learn how to generate that kind of effort.  But it’d be closer due to the lower technical demands of the movement.    But getting anything approximating a true 1RM in the squat, deadlift or bench on the first day of training?  No way.  And everybody reading this knows it.

So right away there’s a red flag for me.  For this reason, you can’t base training loads on a percentage of 1RM when one and possibly both 1RM measures are likely to be meaningless as hell.

Maximum horizontal pushing force?  Nobody cares.  Moving on.

So what about the glute growth?  Upper and lower gluteus thickness was measured via ultrasound 3 days before and 3 days after the training program.  This is certainly standard enough.

Nowhere does it indicate who did the measurement or if it was or was not blinded (a statement in the discussion mentions inter-rater reliability but I have no idea if this means more than one person took measurements for comparisons or what).    As I have done when I examined previous papers in recent months, I will assume it was not blinded since it was not explicitly mentioned.

I consider this a fair assumption and here’s why.  To make sure I don’t miss it, I always search PDFs for the word “blind” or “blinded” to see where/if it appears.  In the case of Bret’s thesis, one chapter did mention explicit blinding.  So it’s not as if it was left unspecified anywhere.  By extension I assume that if it’s not mentioned, it wasn’t blinded.

Note: I could be wrong about this but I can only go on what is or isn’t explicitly written in the paper.

So once again we have the (I assume) problem with an unblinded ultrasound.  And if Bret did the measurements himself, which I tend to assume was the case (and I could be wrong about that too but it wasn’t described), well then we have the always present spectre of bias.

When you consider that Bret was in the process of setting himself up to create and entire career based around the hip thrust and showing that it was superior well……

Hip Thrust vs Squats: The Results

I’m not going to present all of what was found.  The study mentions that time under tension was higher for the squat versus the hip thrust (longer range of motion) while the volume load was higher for the hip thrust (i.e. they used heavier absolute weights).  It also mentioned that TUT multiplied by volume load was nearly identical for whatever that is supposed to mean.

I’ve presented the volume load data below.

Hip Thrust vs. Squat Load Volume

At least visually hip thrust load volume does seem to go up more consistently than squats but eyeballing data is always a little bit dangerous.  There was no statistical analysis so take my comment for what it is.

Whatever, let’s look at growth.  With only two subjects, standard statistical methods (i.e. P values) couldn’t be used and, honestly, I’m not going to bother pasting what they did do to decide if the results were significant or not.  I’ll just take it at face value that it was good enough and move on.

The overall changes in glute size were in the 20-23% range which was considered significant based on what they described and I didn’t.  I’ve presented the actual values below.

Hip Thrust vs. Squats Results

So for the upper glute, the growth was 4.5 mm in the squat and 5.2 (rounded) mm in the hip thrust.  For lower, it was 4.3 mm and 5.0 mm.     The average difference was just about 0.65mm which is a difference albeit a small one.  By percentage the hip thrust generated 2.7 and 2.9% more growth than the squat.

Edit: I managed to leave out one last thing in the writeup which is this:

The twin that performed the squat lost 1.4 kilograms during the six-week period (64 to 62.6 kg), but the twin that performed the hip thrust did not change body mass (63.3 kg).

Given what we know about the impact of appropriate calorie intake on muscle growth, the fact that the squat twin lost 3 lbs over 6 weeks is important.  Then again, we might actually be surprised that she gained as much muscle as she did.  But this is what happens in beginners.  Would her glute growth, inasmuch as it was barely less than the hip thrust, been higher if she hadn’t lost weight.  We can’t know but this is just another flaw in the study.

Hip Thrust vs Squats: Discussion

Ok, so both the squat and hip thrust caused glute growth…in one untrained individual twin in each group (let me keep reminding people of that).  With the statistical method used (that I didn’t describe) both were considered significant.

But were they different from one another statistically?  As the paper explicitly states:

However, it needs to be noted that both exercises produced statistically significant changes to thickness in terms of the 2SD band method, and whether the between exercise percent changes are statistically different is unknown.

The difference in growth was 0.6 mm and ~2.7%.  Is that statistically significant?  Real world significant?  They didn’t know, I don’t know and you don’t know either.

It also notes that:

A greater effect size was noted in the upper gluteus maximus in the squat than the hip thrust (7.72 vs. 6.68); this difference is due to the lower SD in the squat twin.

A much greater effect size was found in the lower gluteus maximus for the hip thrust when compared to the squat (6.32 vs. 3.84), which is partly due to the larger pre-post difference, and partly due to the smaller SD in the hip thrust’s post measure.

And you can try to parse your way through that as well as I can.

In looking at other research, it is noted that this “study” generated greater growth (20-23%) than other studies (12.7-17%).  An explanation for this is given as:

These superior values may be due to the previously discussed genetic variability, and that these twins are hyper-responders to resistance training. Additionally, both twins were athletic, but had not previously performed resistance training with progressive overload; therefore, they were sensitive to this novel stimulus.

Yeah, they were total beginners.  So a novel stimulus to be sure.  The paper also mentions that this seems to contradict the prevailing idea that growth doesn’t occur for the first 6 weeks.  I don’t disagree.  Does that mean something else was going on?  Can’t say and I won’t speculate since that would only represent my bias.  I can only work with the data as it is presented.

It also mentions:

Lastly, the increases in hypertrophy were not necessarily proportional to the differences found in the upper and lower gluteus maximus for each exercise (Contreras et al., 2015a), suggesting that electromyographic activity may not be a good indicator of hypertrophy.

Which is why I didn’t bother with the EMG work in his thesis.  It’s at best roughly indicative and at worst irrelevant.  Bret put his name on a paper saying exactly that (which also means he can’t fall back on it when data he doesn’t like appears).  All of the EMG work doesn’t mean jack squat (or jack hip thrust I suppose).  Damn, I’m funny.

The discussion finishes with what is a reasonably cautious conclusion

In addition, this study was a single-subject design that utilized monozygotic twins, so these results cannot be extrapolated to other populations.

It also interestingly notes

These twins had unique anatomies, which seemed to be more conducive to hip thrusting than squatting, due to their anthropometry. More specifically, studies have suggested that a greater crural index is more beneficial for squatting (Lovera & Keogh, 2015), but the twins in this study had rather low crural indices (0.89 and 0.94 for the squat and hip thrust twin, respectively)

Basically they were built better to hip thrust than squat.  Which I’m sure can’t have impacted on the results.  No siree, Bob.  Not in the least.

Hip Thrusts vs Squats: Practical Applications

The final section looks at practical applications.  Note that there was more presented having to do with the strength transfer and vertical strength bit.  I don’t care about that so I’ll focus on the growth data.

First it is stated

In this respect, the results are telling, in that the hip thrust performs quite well in comparison to the squat for eliciting gluteus maximus growth, hip thrust strength, and maximum horizontal pushing force.

Well in two rank beginners built better to hip thrust than to squat trained over 6 weeks.

Adding

From these data, it is recommended that those seeking gluteus maximus hypertrophy incorporate the hip thrust as a part of their training protocol.

Well except that he didn’t find a triplet to do both movements to see if it had additional benefits to the squat.

I would note that the above showed a great deal of caution in conclusions.  You might compare that to Bret’s later writings online where he said something to the effect that:

Ronnie Coleman would have been bigger if he did hip thrusts.  Andy Bolton would deadlift more if he did hip thrusts.  Usain Bolt would be faster if he did hip thrusts. Hip thrust are that good.

It brings to mind another researcher who’s very weak/anecdotal/meaningless statistics were announced as “data that will BLOW THE ROOF off of previous volume recommendations.”  Oh wait, that researcher was part of Bret’s thesis.  How very curious that…

Anyhow, with that said, it’s time for my comments.

Bret’s Hip Thrusts vs Squats Study: My Comments

First, let’s sum up: Bret’s initial study on hip thrusts vs squats for booty growth had two subjects, twins (one point to Bret) who performed either the squat or hip thrust for 6 weeks, generating a relatively small difference in growth (0.6 mm or 2.3-2.7%) between the two.

Where to start with this mess?

Well clearly the use of two subjects is questionable.  Yes, he cites data saying it can be done and be valid and well, ok.  But it’s hard to see these results as particularly meaningful.  Sure, other exercise science studies usually only have 10-12 subjects but it’s still more than one (in fact it’s 9-11 more than one although someone should check my math).  I will still give him a point for using twins in an attempt to minimize genetic/intra-individual differences.

Tangentially, that might be an interesting route to pursue going forwards with some of the other studies.  Get 12 sets of twins and put one in each training group to compare volume or frequency or something.  But I digress.

Now, let’s talk about the choices of exercise.  The squat is highly technical and depending on someone’s background can take a while to learn well (I’ve taught it to gymnasts in 5 minutes but that’s about it).  At least it was a half squat as opposed to some of the partial squatting that gets used sometimes.

In contrast, the hip thrust is essentially a basic isolation movement that anybody can learn almost immediately.   Growth occurs earlier when simpler movements are done compared to more complex simply because the former doesn’t have a long neural learning phase.

The 1RM Measurement

First this is going to call the initial 1RM numbers into question.  Who can generate a true 1RM in a movement they just learned in their first workout?  Nobody is who.  If the study had tested 1RM daily it would have increased every day.  That’s how meaningless it is in beginners.

In contrast, the hip thrust was probably closer to a true max, inasmuch as beginners can generate a true max effort.  It too would have gone up daily just by doing nothing but practicing the 1RM.  It’s simply a useless measure in beginners and I’d say that about any study that used it.  It doesn’t mean anything for an untrained individual.

The Workout

The workout is certainly not the worst I’ve seen (I’ll be honest that Barbalho’s stock workout is a little bit weird to me) but set poundages according to 1RM.  When you have a useless 1RM measurement, possibly more useful in the hip thrust, this means problematic starting poundages.   Now, since loads were increased as strength went up, this isn’t as big of a deal except for the first workout.

But there’s still a problem here.  Remember the workout, it was a series of straight sets to an AMRAP (As Many Reps as Possible) set.  AMRAP means reps to failure.  Now first and foremost most people don’t know how to go to failure.  At best they go until it gets uncomfortable.   It gets better with practice but it still takes practice.  And it’s far harder to do as many reps as possible in a technically complex movement than a simpler one.

Ask yourself, assuming you know how to squat and hip thrust which one you can take closer to a true limit set in terms of maximum reps?  Hint: it’s not the squat and you know it.

And consider that within the fact that they were beginners.  How close to true limits is a beginner doing a technical movement going to get?  Not very, their  form will fall apart quickly.    In contrast, I could take someone on a simple isolation movement and take them until they gave up without technical breakdown.

Squats to failure is also exhausting which is why nobody has taken me up on my challenge to prove that they can do the workout in Brad’s study.  Hip thrusts are a lot less so.  Anything to limits sucks but isolation movements suck a lot less than complex movements like a squat.

So right there I’d expect the hip thrust twin to have much more success with the AMRAP sets than the squat twin.  And anybody who thinks about it logically will probably agree with me.  I’m not 100% sure, since the training was only roughly described, if the AMRAP was used to set training weights as they went.

IF (I said IF) it were, that would vastly impact the poundage increases on the two lifts.  The load volume definitely went up faster and more linearly in the hip thrust than the squat which might support that interpretation that weight on the bar went up faster in the hip thrust vs. the squat.  Again, I don’t know, it wasn’t stated explicitly in the chapter that I can find.  I could be just as wrong as I am right.

Back to the Twins

That that the twins were apparently biomechanically more well suited to the hip thrust than the back squat.  That alone would predict that the hip thrust would get better results.  I mean duh.

And yes, clearly in a practical sense, the hip thrust may be a “better” movement biomechanically for some than others.  It may have other benefits in terms of not having to shoulder the bar, etc.  This was never up to debate although the number of people I saw misinterpreting this was staggering.

The issue is whether Bret’s claim that “the hip thrust IS superior to the squat for glute growth” is true or not.  Nothing more and nothing less was being said but a lot of people missed this entirely.

And the point of this is that the subjects were more well suited to one movement than the other.  So it’s not surprise that the hip thrust was “superior” for the one twin in this case.  Inasmuch as the ~2.5% difference in growth was superior.

The Results

In that context it’s almost surprising that the squat generated growth at all. Not because it’s inferior but because the twin’s training was likely to have been so inefficient due to having to learn the freaking movement during the study and not being built well for it.

I ignored the strength gains but it was about 17kg in the squat twin over 6 weeks.  Which sounds like a lot but take a beginner learning a new movement doing it three times per week and it’s not that amazing.  Just practicing squatting would have gotten that.

Even there, the results weren’t that different in the big picture.  Non unless you consider a whopping 2.3-2.7% greater growth in the hip thrust meaningful.  It was 0.6 mm difference, nearly 10 fold less than the change in either group (about 5 mm total).  The way Bret has always presented this it sounded like the hip thrust BLEW BOOTY GROWTH OUT OF THE WATER compared to squats and these results just don’t support that.  It was a tiny bit better and that’s it.

My Conclusion

Basically as I see it this study was loaded about as well to favor the hip thrust as it could have been.  Two subjects, built to hip thrust, doing either a high skill complex movement that they had never done which takes a while to get good at or something you can learn in one or two sets, with measurements having been done in an unclear fashion (i.e. who did it, was it blinded or not).

Pretty much if I wanted to load a study to reach my preconceived conclusion this is how I’d do it.  At least I’d use this mess as a roadmap.  I’d take one subject built to do the simple movement I favored and compare it to another subject having to learn a complex new movement that they aren’t well built for and, well…you get the idea.

Basically, I don’t find the results, small as they are that impressive.  It was clearly loaded to favor the hip thrust on every level.  Even then the differences in results was miniscule in the big picture.  There’s not even an indication that the growth differences between movements was significant.  Both grew significantly.  Whether they grew the same or differently is unknown.  It was 0.6 mm and 2.3% difference.  Go see how much 0.6 mm is.  Is that statistically different?   Real world different?  You tell me.

Of course, in beginners, as I said, everything works about the same so maybe it isn’t that surprising that the growth results were nearly identical.   So any conclusions about the hip thrust being superior, much less building an entire career/industry is hardly supported.     The differences were small in one subject doing a simple movement they were built for compared to a second subject doing a complex movement that they weren’t built for.  Yay?

And now to wrap up.

Hip Thrust vs Squats: The Tale of the Tape

So all of this started after the paper Back Squat vs Hip Thrust Resistance-training Programs in Well-trained Women by Barbalho et al. came out.   And I and a bunch of other people reported on it to various degrees.  And well, Bret…Bret was not happy even if his primary approach to rebutting it was a series of appeals to authority, anecdotes and ad-hominems.

But just to finish the savaging of Chapter 11 of his thesis, let’s put his thesis “study” up against Barbalho’s paper methodologically.

Hip Thrust vs. Squat Tale of the Tape

And well that seems fairly clear to me in terms of which is the higher quality research.  Barbalho had more total subjects, well trained versus untrained, a longer study, a blinded ultrasound, a longer time to measurement, all sets taken to failure, a full squat versus a half squat and a much larger and statistically different result.

Barbalho KO's Contreras

Sorry, Bret, it’s a knockout for Barbalho on every level.   He may not even lift but he and his group can out science you all day every day.  And since we’re only focusing on the science here, that’s all that matters.

Similar Posts:

Facebook Comments