So two or three weeks ago I did an unbearably tedious and ranty review of the Barbalho paper on women trainees, the one that I was challenged to examine because “DOES IT ALSO NOT CONTRADICT YOU?”. And as much as I wanted to look at it on fundamental grounds, let’s face it: I was mainly doing it that way to be petty. But I do my best work out of spite and anger.
In any case, I think the paper made a couple of points not the least of which being that, when you’re competent, you can do good science. And when you do good science, you often come up with contradictory results to when you’re not doing good science (i.e. having the lead researcher do the measurements unblinded as a random example). This isn’t to say that the paper didn’t have it’s limitations, all of them do. But methodologically, it crushed a more recent study on the topic which did not do these things. I digress.
Following up on that comes a spanking new paper from the same group that I also want to look at. Thankfully it won’t take nearly as long or be nearly as ranty for the reason that it’s basically identical to the women’s study with one or two very minor differences (one of which is a big improvement on the women’s study). So most of it will just be referring folks back to the first review. Let’s go.
Barbalho, M et. al Evidence of a Ceiling Effect for Training Volume in Muscle Hypertrophy and Strength in Trained Men – Less is More? Int J Sports Physiol Perform. 2019 Jun 12:1-23. doi: 10.1123/ijspp.2018-0914.
To compare the effects of different resistance training volumes on muscle performance and hypertrophy in trained men.
37 volunteers performed resistance training for 24 weeks, divided into groups that performed five (G5), 10 (G10), 15 (G15) and 20 (G20) sets per muscle group per week. Ten repetition maximum (10RM) tests were performed for the bench press, lat pull down, 45º leg press, and stiff legged deadlift. Muscle thickness (MT) was measured using ultrasound at biceps brachii, triceps brachii, pectoralis major, quadriceps femoris and gluteus maximus. All measurements were performed at the beginning (pre) and after 12 (mid) and 24 weeks (post).
All groups showed significant increases in all 10RM tests and MT measures after 12 and 24 weeks when compared to pre (p <0.05). There were no significant differences in any 10RM test or changes between G5 and G10 after 12 and 24 weeks. G5 and G10 showed significantly greater increases for 10RM than G15 and G20 for most exercises at 12 and 24 weeks. There were no group by time interaction for any MT measure.
The results bring evidence of an inverted “U shaped” curve for the dose response curve for muscle strength. Whilst the same trend was noted for muscle hypertrophy, the results did not reach significance. Five to 10 sets per week might be sufficient for bringing about optimal gains in muscle size and strength in trained men over a 24-week period.
The background on this paper is a little bit different, in the sense that they introduce the concept with a discussion of training volume and growth moreso than the adherence issue. Basically it looks at the idea that there is a dose-response relationship between training volume and strength/hypertrophy along with the data both supporting and contradicting it. Not that it doesn’t tie in with the time and adherence issue. If tripling your training time to get only 1.5 times the results is a poor ROI, doing 3-4 times the volume to get no benefit is even worse (and stupid).
Of course they cite their previous paper on women, and I’d once again point out that it is dangerous to directly compare research between sexes in this fashion, but make the point that, with very high volumes of training, there is the risk of overtraining if it is done for extended periods. In this vein they acknowledge that:
Indeed it should be noted that proponents of relatively higher volumes do suggest that very high volumes (should only be performed during short training cycles (~8 weeks; 21). The negative effect of overtraining might only manifest after a longer period and most studies have only relatively short durations (≤12 weeks; 11–13, 15); therefore, it would be interesting to perform longer duration studies to assess that hypothesis.
And I pointed out in my review of the women’s study that, the trash fire that is Raedelli still being ignored (again, in no world do beginners not get growth until 27 sets/week or does a calisthenic group gain more LBM than folks lifting), all of the studies to date were 12 weeks or shorter with the exception of theirs on women. This helps in a lot of ways including to eliminate the potential of water shifts that occur early on to skew measurements.
So a first goal was to look at how volumes might be relevant over much longer periods of training, and this study was 6 months same as the women’s (I have a gut hunch that they did both studies side by side but could be wrong about that).
One factor they also address and I will be addressing this HARD in a week or two, is the control of intensity and “training to failure”. Two authors of this group, Steele and Fisher have written quite a bit about this, arguing for the need to rigidly define muscular failure so that studies can be meaningfully done and compared. The reality is that most don’t and probably haven’t gone to true failure, defined here as working until the bar can no longer be moved no matter how much effort is given. Most stop when it gets uncomfortable and that’s not failure. In fact it’s usually not anywhere close and I’ll be demonstrating this on the site soon.
And inasmuch as volume and intensity are related, the higher the intensity/closer to failure, the less volume you probably need and/or can tolerate. And this is important. If you’re faffing about in the gym, you might need a zillion sets to get a stimulus. If your training has some degree of quality, not only do you not need it, you can’t do it. I’ll say it again, if you can’t get it done on 6-10 hard sets/workout for a muscle group, you’re the problem. Your form sucks, your focus sucks, your intensity sucks. More volume isn’t the problem. You are. If you’re in Austin, I’ll prove it to you by taking you through a workout. You won’t even remotely want to do more work than I give you.
Note: I’m still waiting for those videos of people doing 5 sets of 8-12RM (I mistakenly called this 15RM previously and want to acknowledge my mistake) of squats on 90 seconds by the way. Where RM means that the last rep is a death grinder and the next rep would not be accomplished and you get stuck at the bottom or need a spotter to get you to the top. I’ll be waiting a while. Because if I put you through a TRUE 12RM squat set, you’re not getting off the floor in 90 seconds much less doing it again with anything but a 50% reduction in weight or getting 3-4 reps.
So with that background, they set up the study. Which, as noted, was mostly identical to the women’s study so I won’t go into nearly the same degree of detail.
So I’m not joking that this study was more or less identical to the women’s study. They did the same a priori power analysis with the same statistical requirements to see how many subjects they needed, wait until they had them and ended up with 43 subjects (with 40 required to meet their statistical goals).
The men had to be 18 or older, have been lifting for at least 3 years and be able to perform a 10RM bench with at least 100% of bodyweight and leg press at 150% of bodyweight. No further information was given about their training history or current training practices.
I’ll come back to this below since one issue I brought up regarding the women’s study were the fairly poor strength levels and my feeling that the women were not particularly well trained (i.e. their bench press estimated 1RM was novice after 3 years and achieved intermediate after 6 months with all tested exercises making 50% strength gains). It appears that 3 participants did not reach the required 80% attendance so their data was thrown out and it’s interesting that 2 were in the next to highest volume and 1 in the highest.
Like the women’s study, this one lasted 6 months. As always, diet was not controlled and subjects were told to eat the same as they had been before. This is always a problem with this research but none of the other volume studies have controlled diet so it’s at least a consistent problem. The best you can do is tell subjects to maintain their normal diet and hope for the best.
Once again, pre-study anthropometrics (height, weight, etc) were provided but post-study were not. I still just see this as an absolutely baffling (apparently my new favorite word) oversight for the same reasons as I gave on the previous study. It’s trivial and non-time consuming to do and would give at least some indication of whether any of the groups gained or lost weight (It wouldn’t tell you about body composition). I do not understand why this would be ignored.
The subjects were tested on the same 4 movements: bench, pulldown, SLDL and leg press for a 10RM and here there is a slight difference from the women’s study. To whit, in the women’s study, only pre- and post-testing were done so there were 6 months between measurements. This doesn’t give any real indication of what happened in the middle in terms of did one group grow faster initially but slow later on or vice versa.
In this study, testing was done before the study, at the halfway (12 week) point and at the end (24 week) point. Since part of their goal was to see if volume response differs over different time frames or there is a change from one to the other, this makes sense and I consider it a huge strength (haha) of the study.
That is, what if their results showed that higher volumes were superior over shorter periods and then became negative after that due to overtraining. This would help to explain differences in study designs and results and training recommendations. Perhaps for short periods, high volumes ARE superior but continued long-term they are not as is being asserted. A mid point test allowed that to be examined directly.
Muscle thickness was measured by Ultrasound just as in the women’s study at biceps, triceps, pecs, quads and glutes (again suggesting that they ran the studies side by side and that you can do more than arms and quads if you put your mind to it) and, as before I want to explicitly point out:
All MT measures were performed in a specialized clinical center by the same experienced technician, that was not involved in the study and who was blind to group allocation.
The tech was blinded and not even involved with the study because it is possible to do good science when you put your mind to it or want to do it (HI MIKE!). You have to be pretty incompetent, or biased, to not have figured out to blind your Ultrasound measurements 17 studies in….moreso when you supposedly teach a class on research methods.
The exercise program was the same batty split routine with the same bonkers set counting and weird undulating set, rep and rest interval structure as with the women’s study with none of the exercises achieving the supposed 5,10,15 and 20 set counts so far as I’m concerned. Training was supervised by specialists not involved with the study, as before.
Again, good science can be done if you try. Blind the people involved to as great a degree possible. It is the ONLY way to REDUCE bias and arguing that it’s ok not to do this or that standard guidelines don’t apply is truly horseshit apologism. It’s saying “Exercise science is allowed to be shit.” at which point you might as well throw it out. And it’s worse when what needs to be done is fairly trivial. It doesn’t cost money to blind the Ultrasound tech. And yet….
The same basic statistics were done as before although the addition of a third measurement time point seems to have made it a little bit more complicated. I won’t pretend to understand what they are describing and will presume that it met the same standards as the previous paper with slightly more complication. Since the researchers, as before, actually adhered to what their statistics said, I consider this a safe assumption. I’ll be happy to be corrected if I’m wrong.
So they didn’t present the results nearly as clearly as in the women’s paper. I don’t know why or if this is something about the journal it was published in or what. It might be due to the third measurement point. I don’t know. Basically they provided the raw changed with a billion subscripts my old eyes can’t be bothered to try to find but didn’t really describe it. So here is that actual data if you think you can figure it out.
Good luck with that. Since it’s easier to see, I’m providing the graphical representation of the strength and growth changes, I’ll also provide their descriptive conclusion from their conclusion section.
Which pretty much shows the same results as the women’s study with G5 and G10 being the top lines and G15 and 20 the bottom lines. On all movements G5 and G10 both improved better than G15 and G20 but G5 and G10 were essentially identical. Of some note, this wasn’t a case where the higher volume groups did better for the first 12 weeks and then fell off. They did worse across the board. So the idea that higher volumes are sustainable or even optimal over short periods but become worse over time was not supported by the study’s results. The higher volumes were worse from start to finish.
For growth, here are the results.
This is much harder to see and I wish they’d used the same estimation statistics graphs they used in the women’s study but it may be that the addition of the second measurement time point prevented that. To avoid having to describe the above and repeat myself, I’ll show their conclusion below.
Ok, since the above is pretty damn hard to see and they didn’t describe it well, I’m taking their own conclusion and breaking up the paragraph slightly to discuss strength and growth separately:
The results showed that all groups had significant improvements in all variables; however, for strength, the higher gains seem to be obtained with 5 and 10 sets per week,
Strength also improved across all time points with the differences between 5 and 10 compared to 15 and 20 weekly sets increasing with time,
So as with the women’s study, 5 and 10 sets gave the superior and more or less identical strength response with 15 and 20 giving a lesser response. However, all groups made strength gains across the entire length of the study. They were just worse in the higher volume groups and, again, they were worse from start to finish. The higher volumes weren’t superior over any time frame.
and for muscle size there appears to be no statistically significant interactions between 5, 10, 15, or 20 sets per week over time, though pairwise estimates suggest lower volumes may result in greater changes over time as higher volume may result in overtraining after the initial 12 weeks (Table 5).
whereas muscle size appeared to plateau after 12 weeks in all groups.
So for muscle growth, all groups grew the same and this was different than in the women’s study where the higher volumes were distinctly inferior for growth. It still makes the point, whether set volume was low or high, growth was identical within the setup of this study. Doing 2-4 times more work yielded zero benefit.
Of some interest, all groups showed no further statistical increases in muscle size after 12 weeks (this is in contrast to the strength data) and you can see the growth curves start to flatten out past 12 weeks. After the 12 week mark, there was a suggestion (which I take to mean it didn’t make statistical significance) that the higher volume growth was inferior and that they were becoming overtrained.
I am quite sure that if I could read the data chart above, I could figure out if it reached signficance but I will assume it didn’t due to their wording. Basically, if the higher volumes caused overtraining, it didn’t reach statistical significance of P < 0.05. So it’s not a ‘real’ finding in a statistical sense and I won’t present it as such. It still makes the same point: doing 4 times the volume didn’t generate superior results over any time frame with the possibility that it was becoming too much past week 12.
So the basic conclusion of this paper is similar if not identical to the women’s study: the lower volumes, at least within the context of this study (sets taken to true failure) gave identical results to one another. In terms of strength, the response was superior for the lower volume groups but for growth there was no difference between groups. All groups stopped growing (in a statistical sense) at week 12 with a suggestion that the higher volume groups were still getting a more negative response.
In discussing their results, they hit all the same basic points as in the women’s study and you can tell that a lot of the discussion was just pasted from the other study. This is not uncommon at all and researchers will frequently use the same verbiage across publications when discussing a certain topics with review papers being the worst in this regard. The various meta-analyses, that often contradict one another, the fact that most studies don’t use volumes much above 10 sets per week (though we are now at 7-8 on men depending on if you count Raedelli which I do not and one in women), etc.
But they cite the same set of papers, get Ostrowski the same degree of wrong by misreporting the set count silghtly but still correctly concluding that the results matched theirs (i.e. not reversing the conclusion like someone did). They cite a couple of others, the same ones as before, the Osawagara rat study which I continue to think is interesting but not applicable until it is repliacated in humans, Wernbom, etc. Just go back to the women’s analysis I did, it’s all the same in terms of their comments and mine.
Then they get to Brad’s study.
Here they do report it differently and “properly”, stating that the highest volume groups did get better lower body growth (previously they reported the triceps which didn’t even make a pair-wise comparison). Mind you this is still predicated on the belief that a BF10 of 3 means more than jack shit which it does not. Let me reiterate that it’s described as anecdotal and not worth a bare mention in standard statistical texts and no way in hell would Brad or James accept a study that contradicted them with such weak stats.
Note: We’re still waiting on James Krieger to provide the supposed source that says that a BF10 less than is more than meaningless as he asserted on Redditt. And we’ll be waiting forever because it doesn’t exist and he was bluffing/lying about it. Because if it did, he’d have provided it by now. And used it as the actual reference in his original paper (he cited Rafferty). It was just a bluff/boldfaced lie when he got caught out on how weak the statistics were.
And here is what they speculate as to the difference
The conflict between these results with our results and the previous literature might be in the protocol used, especially the different definitions [of failure] that could have been used for set endpoints.
Which like I said is really the crux of a lot of this. Calling a set “to failure” when it’s just people stopping when it gets hard is going to give a fundamentally different result than going to true failure in terms of how much volume is needed, optimal or can be survived.
This is even more true on lower body than on upper body. I’ve seen a lot of bros take bench or other upper body movements to failure to get those “It’s all you, reps” I’ve never seen anybody that wasn’t me, my old training partner or someone I was coaching deliberately take a set of squats to failure (yes, I’ve seen missed singles, I mean higher rep sets to true failure). NOT ONCE.
I can find one or two videos online but await my critics and detractors to send me video of THEM doing it (c’mon, PROVE ME WRONG). I won’t hold my breath. And, as above, if you’re ever in Austin, let me know. I’ll show you what squats to failure are like and you won’t want to do a second set when I’m done with you. Or possibly any further leg work that day. Or ever again.
As they state
Indeed, it has been reported that, even when trainees are oriented to perform repetitions to momentary failure, they may interrupt the set instead due to discomfort, which might be especially true for lower body(23). In such cases, an increase in training volume might bring additional benefit.
Therefore, whilst there are controversies regarding the need to train to momentary failure to optimize adaptations(24–28), the correct definition of set endpoints might be necessary in order to compare different RT studies(21), since this seems to influence the results during lower volume resistance training protocols (<4 sets per muscle group per week)(15,29,30). Therefore, one important aspect of the present study is that the participants were closely supervised in order to reach the defined set endpoint.
Based on their literature and having talked with two of the researchers on these studies, they define failure as “continuing the set until the bar will not move no matter how much effort is provided.” That’s true failure and most don’t and have never gotten there (again, maybe on bench or some isolated upper body movements). But pushing their subjects to that level may be a big part of why higher volumes were not superior (and in some cases inferior). Or rather, lower volumes worked just as well if not better.
It does raise questions in my mind, the same question I had in the women’s study, regarding the higher volume groups. The G20 group during the low rep section would have been doing 14 sets of chest for 4-6RM. That’s a staggering volume and few could achieve even with long rest intervals. Even during the high rep weeks (all 6 of them), I doubt anybody but the lowest volume groups were able to complete the high volume squats at 12-15RM on 30-90 seconds. 2 sets might be achievable although it would kill most people. Beyond that, it’s an impossible workout in the same way 5 sets of 8-12RM on 90 seconds is an impossible workout. It was also only 6 weeks out of 24 in direct contrast to 8 weeks of that workout done three times per week.
Next they turn to the somewhat odd hypertrophy results where growth between groups was more or less identical although there was an indication (again, most likely not statistically significant) that the higher volume groups were starting to overtrain after 12 weeks. Which still makes the point: if you can do “5” sets of “20” sets and get the same growth, why do “20” sets?
To that they add
As the majority of trainees are engaged in resistance training seeking for muscular hypertrophy(23) it is also worth noting that it is not presently clear to what extent any measurable change in muscle thickness or other measures of hypertrophy do actually translate into perceptible aesthetic improvements. With this being the case and considering our results, perhaps persons with the goal of hypertrophy might take into account whether or not higher volumes of training are worthwhile in this regard.
Which is an interesting point. Eyeballing the data in the big ugly chart above the change in muscle thickness was 5-7mm (smaller for pecs and arms, bigger for glutes and quads) and it’d be interesting to know how much visual change that makes. Not the neurotic physique athletes care, all growth is good growth. But it still makes the point in terms of the average recreational trainee for whom time is often reported as a primary limiter.
If growth is identical or even superior and the small amount of growth doesn’t really change appearance for ‘5’ true limit sets vs. ’20’, why do 4 times the work?
The paper finished by making this point:
The present article has some important strengths, like direct supervision, the blinding of the assessors that measured MT, the sensitive method used to assess muscle hypertrophy, the performance of MT across multiple sites, the relatively long duration of intervention, and the evaluation at the midpoint of the intervention.
Strengths, indeed (ahem), some strengths that other labs really need to get on board with.
Along with these limitations
A limitation of the present study was the absence of dietary control. However, the participants were constantly questioned to see if there were any relevant changes in their dietary habits and no significant changes were reported. Notwithstanding, in addition to the long-term influence of dietary habits in the adaptations to a resistance training program, water and food consumption may alter anthropometric assessments; therefore, the lack of a rigid dietary control might have also acutely influenced MT measures. Lastly, without the inclusion of a time matched control group it is difficult to truly differentiate the changes reported here from test-retest error over that time point. Our reliability data was with respect to day-day test-retest error. However, it should be noted that it is difficult to recruit a population of trained participants and persuade them to abstain from training for a sufficient period to determine this.
I mentioned the diet thing, it’s an inherent problem with all studies like this. It would cost millions to do a diet controlled study for this long. The lack of a true control group is also problematic, perhaps 6 months down the road they’d have gained muscle thickness too. I doubt it. Of course, some who want to believe in the Raedelli study seem to believe that the calisthenic group gained MORE lean body mass than the low volume weight training group so….But like they said, no way are you getting trainees to quit for 6 months. And few other training studies have a non-training control group that I can think of (Yes, I am SURE they are out there, probably in beginners, there so spare me).
Thus they conclude:
The present results suggest that as little as 5 sets per week might be sufficient for attaining optimal gains in muscle strength and size in trained men during a 24-week resistance training program, at least when all training sessions are closely supervised and the sets are performed to momentary failure. The results also suggested potentially negative effects of a training volume exceeding 10 sets per muscle group per week, especially evident after 12 weeks of training, which suggests that the resistance training dose might be especially important in long term.
And I’d note again the use of suggestive rather than absolute language. “Suggest”, “might”, “sufficient” even within their own statistical choice (P < 0.05). And again they note that these results are within the context of sets being taken to TRUE failure with the acknowledgement that more volume might be useful, beneficial, survivable if true failure is not achieved Not some self-determined, I want to stop now definition.
So I don’t have much to add and certainly not much that I didn’t say about the women’s study. The workout was really badly set up, the set counting is bonkers and even stupider than assuming 1:1 from compound to isolated and I don’t really like how they structured the workout with the weekly set and rep changes. But they didn’t ask me. I really do think the studies were run side by side given the identical nature of them.
I still question whether split volume would be superior. Certainly it’s more of an issue for women who generally recover more quickly but, as with that study, consider that the volumes got oppressive in the high volume groups with the low reps. 7 sets of 4-6RM for 2 chest exercises in a single workout is absurd. That’d kill most people. And stronger men moreso than relatively less strong women since men will fatigue much harder under those conditions. A woman benching 100 lbs might get away with it. A dude benching 2.5 times that won’t.
Would 5 sets twice or even thrice weekly give a better response than doing it all in one week? I suspect so but I’d like to see it studied directly. Basically, all this kind of study might support is that there is a per workout limit to volume (and we knew that as far back as Wernbom with 70 reps/workout being 8-10 sets tops to begin with). But that doesn’t automatically give us a per week maximum volume.
Note: Why does my gut say that James and Brad, despite having written meta-analyses that frequency doesn’t matter if volume is equated are dismissing the findings of this study by arguing that the volume should have been split across more days? I don’t know this is the case. But I’ve watched gurus goal post shift for years and wouldn’t be shocked if I’m right.
Mind you, when you’re working at true limits like this study, you have to start to worry about recovery between workouts and doing that many truly limit sets three times weekly might not even be possible to recover from. But twice/week, that’s doable and I think the results might be better. So have subjects do 5 or 8 sets to failure twice/week and see what happens.
I’d note that the one “dismissal” or whatever you want to call it that I’ve seen has been that “well, they trained to failure.” Right. And that’s a problem why? Because it makes the very important point: if you’re training with quality, you don’t need a shitpile of junk sets. When you faff about in the gym, maybe you do need 70 sets/week for back because you’re a shitty trainee with no focus, technique or intensity and your coach doesn’t know how to fix that so he just throws asinine volumes at you. Maybe.
Don’t misread this, I am not saying that everyone should take every set to failure all the time. We know that burns people out done too long (thought the low volume group in this study didn’t at least with regards to strength). Rather, the point is that when you’re doing quality work you shouldn’t be ABLE to do that much volume in the first place. Trust me, if you’re even 1-2 reps from TRUE failure (and that means knowing that failure actually is), you can’t do a ton of sets. Maybe it’s 6-8 instead of 4 or whatever. Maybe you get 10 quality because you’re a god. If you can do many more than that, your training intensity sucks. Come to Austin and I’ll prove it to you in one workout.
I am guessing that the high volume Internet brain trust is making exactly that argument since this study and the women’s equivalent contradict them SO HARD. Perhaps rather than worrying about other lab’s data or how to dismiss it that is done with methodological rigor that they should focus on learning how to do their own research properly: blinding the tech, etc. I still suspect that their results would change A LOT when the fucking head researcher wasn’t doing the goddamn measurements unblinded. Unbiased my ass. What a joke.
Ok, the final question to ask.
Were the Men Really Trained?
So this was a big question I brought up in the women’s study and I want to examine it here for consistency. As I mentioned above, I argued that based on their initial strength levels on bench (which was a straight up novice value) and their strength gains which were 50% across the board in all 4 tested exercises that the women weren’t truly well trained. You simply don’t see that kind of strength gain in 6 months after three years of proper “training”.
It’s also possible (and likely IMO) that this was the first time the women were exposed to low rep sets so they probably had a lot of room to improve neurologically. And pushing up low rep strength can push up 10RM strength. As well, despite reporting “training to failure” well, I’ve been in a lot of gyms for a lot of years and I just haven’t seen it among 99% of trainees. Not true failure. Not women and not men (though men are more likely to do it on upper body to impress their buddies).
So I also think it’s the first time they were truly pushed to their limits. That’s my speculation but it’s based on ages of observation. No, spare me the accusations of falling back on anecdote. Studies show this for both women and men: most self-select inefficient intensities of training and you all know most aren’t going to failure. Go observe trainees in your gym today and you’ll see I’m right (unless you’re in some super hardcore gym).
But what about the men? Were they trained? First let’s look at bench. The average starting bench press was 96kg for all groups which is 211 lbs 10RM. As a rough (and not terribly accurate) estimate, assuming a 10RM is 75%, that gives a 1RM of 281 lbs. The average weight was just about 82.5kg (range was 81 to 86) or 181 lbs. So that gave them a 1.5 bodyweight bench press to start which is considered an advanced value by online strength standards. Mind you, given how much men love to bench, this isn’t a shock. I’d have loved to know their squat poundages but only leg press was measured.
Inasmuch as it mattered, their SLDL was 86kg or 190 lbs 10RM which isn’t awful (except that I have a female trainee doing 245X3) and the pulldown and leg press don’t matter that much since machines differ to much to make meaningful comparisons. Honestly, this is probably a bit low for SLDL but given men’s insane obsession with benching, I’m not that surprised about the differential.
So I daresay that compared to the women, the men were more highly trained. Remember, the women showed novice bench numbers (and certainly that could be due to simply not caring about the bench as many women do not) and made 50% improvements in all movements which suggest a low training level to me. But the men were starting bench at least at an advanced level. At least by a rough estimate of 1RM.
So let’s look at their strength gains and here I’ll focus only on the G5 and G10 groups who made the best gains. Here the men went from a 96kg 10RM to a 120 kg 10RM, estimating out at a 352 lb 1RM or nearly double bodyweight which is considered an elite bench. It also represents a 25% improvement over 6 months. Yes, estimating 1RM is problematic like this but I did it for the women and want to be consistent here. And the ranges aren’t enormous here. In G5, bench started between 93.7 and 98.7 kg and improved to 116.7 to 123.7 kg. So about a 5 and 7 kg spread total. It wasn’t some insane outlier pulling the average up which happens more than you’d think in studies like this.
Eyeballing the poundages, the other three exercises made the same 25% strength gains as the bench press so it was consistent across the board. And that is pretty enormous. One of the online criticism I’ve seen is that this is an impossible result to have made in that time period and there may be something to that.
Mind you, 25 kg over 24 weeks is 1 kg/week or 2.5 lbs. Yes, I know that strength doesn’t go up linearly like that but put in those terms it’s a lot more possible especially if it was the men’s first real exposure to low repetition weight training (since no training information was provided, we can’t know if this is the case or not).
And perhaps moreso if it’s the first time they were pushed to their true limits. No mention is made of whether or not the men were exposed to training to failure or not and it is odd that a lot of details that were provided in the women’s paper were not provided here (maybe it was a journal thing, I don’t know but find it perplexing or even…baffling). But certainly that is a pretty staggering gain for that time period. Is it possible? Maybe, maybe not. Again, this is within the severe limitations of estimating 1RM in this fashion. But even the 10RM went up similarly so it’s a pretty major strength gain (note that improvements in low rep strength tend to push up higher rep strength to one degree or another).
Could someone go from an advanced to elite bench in 6 months on only 4 sets/week of chest? I tend to doubt it. But unless this group, which is methodologically rigorous somehow dropped the ball on their strength testing, I have to take the result at face value. Especially since, unlike some other papers that sort of ignored their own statistical choices, this group sets it’s statistical parameters up front and then actually adhered to them. By the standards they chose, which were the standard standards, the results were significant. Are folks gonna dismiss them out of hand.
I mean, if some groups are going to ignore their own statistics to accept non-results, it seems reasonable to accept results when they met statistical standards. I realize that consistency isn’t in big supply among the gurus or their defenders but you can’t have it both ways. Either the statistics count or they don’t and you don’t get to play fucking silly buggers about it.
So there ya’ go, a second study, this time on trained (and I’d say well trained) men which had the same finding as their previous paper on women that the lowest volumes were as good if not better than higher so long as the subjects were taken to their actual limits or failure. Using the same good methodology (i.e. blinding the Ultrasound tech, exercise specialists, using guarded language, adhering to their own statistical choices) as before. Also using the same batty workout, set count and rep/set/RI structure. Certainly this needs to be replicated by another lab but it MUST use the same high standard of methodology. So not Brad’s lab.
Because I will repeat my suggestion from my women’s writeup where I suspect studies that are longer using good methodology (BLINDING FOR FUCK’S SAKE) are going to find systematically different results than those that do not. Add to that actually using a proper endpoint criteria for failure. Because most studies aren’t getting there. And when you do that, I bet the supposed benefits of higher volumes will disappear. Make the study 3 months, use proper methodological guidelines like blinding and don’t let the folks stop when it gets a widdle bit hard and I predict we’ll see different results.
Mind you, it’s not as if we haven’t seen the identical results in the majority of studies. Let’s count ’em up.
Raedelli: A fucking trash fire. 6 months in beginners but no growth until 27 sets per muscle group and the caliesthenic group gained more LBM than low set weight training. Sure.
Ostrowski: No difference between volumes statistically although there is a percentage trend for MODERATE volume group to be superior to LOW volume group in triceps. But the HIGH volume NO better than moderate volume (4.8% vs. 4.7% growth) even if some people (mis) represented it that way. 14 sets was the same as 28 by this percentage. No group was superior by their own statistics.
The two GVT studies: No benefit to higher volumes of training.
Haun: Cap at 20 sets but using piss-ass intensities (4RIR with a 8-10′ rest interval) so who knows what it would have found not training in such a profoundly non-effective way. Legs possibly needing more volume than upper body, growth possibly sarcoplasmic.
Hackett: No benefit to higher volumes for biceps, limited by allowing subjects to train outside of study. But if we can ‘just trust a researcher’ shouldn’t we ‘just trust the study subjects not to have done more biceps work’?
Schoenfeld: Only study to show that batshit high volumes are beneficial but with unblinded Ultrasound and a profound misrepresentation of the non-results based on their own statistical choices (i.e. BF10 of 3 = meaningless). Even in the case where percentage gains were similar to other studies, can’t explain why they needed 2-4 times as much volume to achieve it.
Barbalho (men): ‘5’ sets to failure gave the same results as ’10’ and as good or better than 15-20.
Again, I don’t think the study on women is necessarily relevant anymore than the studies on men are relevant to woman so I won’t include it here. I didn’t include the men’s studies in that writeup and won’t change my tune here. I’m all about consistency.
But let’s count it up. We have 8 total studies, throw out Raedelli to leave 7, that have examined the dose-response of hypertrophy and/or strength to training volume. Of which six show no benefit to super high volumes. In some moderate volume is better than low but higher no better and in others there’s no difference. Mike’s didn’t either despite his attempts to change his argument in the debate. Doubling the volume for a tiny ass bit of LBM gain is a shitty ROI and doesn’t support anything, especially if it’s just sarcoplasmic growth anyhow.
Which leaves a single paper suggesting insane volumes are superior, well if you ignore it’s actual statistics. What it actually found was that moderate volumes were superior and higher volumes in upper body were not superior and higher volumes in lower body were utterly irrelevantly superior (BF10 < 3 = anecdotal results not worth a bare mention).
And yet people are still hanging on to Brad’s (not statistically supported results) like a drowning man to a life preserver, the SINGLE true exception to the broader base of literature. Finding ways to desperately dismiss EVERY NEW FINDING that contradicts it (while Brad’s is immune to criticism but happily defended by a bunch of apologists). Including Brad and James, who so far as I know, is still trying to throw new statistics at this shit (and who still owes us the claimed source showing BF10 of 3 or less means something). And in response to this, after whatever lame-ass dismissals they use, they’ll still say we need more studies on the topic. Which isn’t wrong, more research always helps to improve the model. But when a preponderance of evidence starts to point in one direction…..
And let’s reverse it. If 6 of 7/8 studies showed that higher volumes were superior, you know as well as I do that they’d present the idea as having been proven beyond a shadow of a doubt because that’s how gurus roll. The rules they apply change based on whether the literature base supports or contradicts them. When 6 of 7/8 studies say they are wrong, we need more research. When 6 of 7/8 say they are right, it’s conclusively proven that they are.
But as it stands we have one outlier study marked by a lie in the discussion (I’m sorry, a misrepresentation), unblinded ultrasound by the lead researcher and what amounts to ignoring their own statistics to make a strong claim that can’t be supported on any level. And the rest of the statistics not supporting that at all.
A reminder: for ALL muscle groups the highest volume group did not reach P<0.05 (THEIR chosen metric which THEY have to live and die by) for a difference from the moderate group. By Bayesian analysis, neither upper body reached BF10 > 1 (meaning NO evidence). The two lower body muscles reached BF10 < 3 which is anecdotal/not worth a bare mention. Greg Knuckols has been crowing about a study or his thesis with BF10 > 150. Those are strong stats. So how can he continue to defend a BF10 < 3? Oh right, he said they “oversold” the statistics. And that’s being generous as hell. I imagine as generous as the number of 5000$ a pop seminar appearances Brad has gotten for him.
Even IF you think the Raedelli paper is worth a shit (and you still have to explain the insane results), that’s still 2 in favor of higher volumes and 6 not in favor. You build scientific models based on the bulk of the literature and 6 out of 8 says one interpretation is correct with no ability to explain why Brad’s study found such different results mechanistically (he sure didn’t attempt to in his paper, relying instead on reversing the results of a paper that disagreed with him).
And that interpretation is NOT that higher volumes are better (at least beyond a certain point) but rather the opposite. Science is never conclusive but the data is sure as shit supporting that result. How many more do we need before people will accept it? C’mon, guru boys, give me a number so you can’t move the goalposts more and more and more every time a study comes out like this that says you’re wrong. Don’t do what the keto folks do and make the number “one more study than currently exist.”
Certainly Barbalho is on the low end of low volume and will only hold with sets taken to true failure. Which may not be ideal or beneficial long term. With a slight reduction in intensity, stopping 1-2 reps short of true failure, you probably need more volume. No argument from me. But that’s not the point here. The point is that with real intensity, you don’t need and can’t survive high volumes.
Yes, I want to see it replicated with a split volume and at least 2X/week frequency because I suspect that the results will differ (I am also prepared to be wrong about this). At best it has shown a maximal per workout volume (which makes logical sense) but you can’t automatically conclude that means a maximum PER WEEK volume until split volumes are studied.
But even if 5-7 sets to failure twice/week ends up being optimal, that’s still within the range of 10-20 and still a low volume compared to the idiocy currently being promulgated by supposed “Evidence based practitioners.” The same practitioners who have now decided that anecdotes are ok when they don’t want to accept science. Well their anecdotes anyhow, you know they don’t accept contrary anecdotes which is its own cute little fucking game.
This includes one person apparently suggesting that there “may be no upper limit of weekly volume.” Only if you don’t know how to coach or train I suppose. Because if you need 70 sets per week for a muscle to improve and/or your coach thinks that you do, you’re incompetent as a trainee and he’s worse as a coach.
I’m still waiting for those squatting to failure videos, folks. I suspect I’ll be waiting just as long as for Jame’s statistical source showing that BF10 < 3 means something. And even longer for the self-proclaimed “evidence based practioners” to stop ignore the majority of the studies in favor of exactly ONE outlier. But I guess that’s what THEIR anecodotes are for.
I said this wasn’t over.
I have been informed that both Brad and James are playing the same deflection that Mike Israetel used questioning “Why didn’t I bring up blinding on other studies I’ve examined?” This is of course just a deflection. Other studies are NOT on trial anymore than I, my website or anything else is on trial. And, mind you, in the last two research reviews, I’ve examined everything including blinding which is why I keep harping on it.
Both Barbalho studies were blinded with methodological rigor that Brad clearly doesn’t even know about. Both found results that contradict his paper and HIS PAPER ALONE. For being “evidence based” he sure doesn’t seem to like any studies he doesn’t produce.
But it’s always easier to deflect than acknowledge your own bullshit. Because, frankly, Brad could shut me the fuck up in a heartbeat (just as ANY of you could shut me the fuck up by sending me a squat video to failure). C’mon, everybody loves proving me wrong. So FUCKING DO IT.
BLIND YOUR NEXT STUDY, BRAD!
Seriously, you want to be above my criticism, do things right. Blind your study. That’s my challenge to you. I’ll acknowledge it on my website in all caps. Of course, you won’t get the same results you want by blinding it so you can’t win no matter what you do. It’s either unblinded and biased to generate the results you want (that were still not supported by the statistics) or blinded and no results.
But prove me wrong, guru-boy. Do a study properly FOR ONCE IN YOUR CAREER and this beef will be over.
I won’t hold my breath anymore than I’ll hold my breath for the squat to failure videos.
But c’mon, children, PROVE BIG OLD MEAN LYLE WRONG. We all know you can’t or you would have already.
- The Growth Hormone Response to Interval Training – Q&A
- Effects of Low Versus High Load Resistance Training – Research Review
- Strength and Neuromuscular Adaptation Following One, Four and Eight Sets
- Low Load Training and Videoing Resistance Training Studies
- Effects of different volume-equated resistance training loading strategies on muscular adaptations in well-trained men – Research Review