Unlike my previous articles which have tended to have a specific focus or question, this one is just a collection of tidbits that have come up in discussions recently.
For reference – I have the privilege of training a decent number of fellow PTs and powerlifting coaches in person and online. The upside of that is that I tend to learn a lot from them, and in as far as there could be a downside to that, they tend to ask questions and so I need to be able to explain or justify things like why I chose a given movement, why I’m prescribing certain volumes and what the overall macro-structure of the training plan looks like.
So in this article I’ll just lay out a few general thoughts about a few topics, each of which could easily be written into full articles themselves to cover all of the nuances pertaining to them. As it happens, a lot of what I am going to write about, I also discussed with Luke Tulloch on Weakly Weights in this episode, so for much more I’d suggest checking that out.
1 – How many hard sets should I do per workout to grow?
There is a dose response relationship between weakly volumes of training and growth. Schoenfield’s analysis (1) on this research question found increases in growth with increasing volumes up to 10+ sets per muscle per week, but there was insufficient data to determine whether the relationship flattened out or (as I and most others would suspect) it was shaped like an inverted U, with a “most effective dose”, past which point gains diminished.
This meta analysis was partly conceived to further explore results of an analysis by Krieger (2) on single vs multiple set resistance training. In short, it found that 2-3 sets/workout produced greater gains than 1 set, and whilst 4-6 sets/workout did not produce significantly greater gains than 2-3 sets, there was a trend for increasing set counts increasing hypertrophy which would indicate that doing more then 2-3 sets/workout likely was better.
As with weekly volumes, I think it’s probable that the relationship between per-session volume and growth has an inverted-U shape. Up to a point more volume is more stimulative, however past a threshold extra hard sets likely add very little stimulation and incur greater recovery costs due to the extra muscle damage that they entail.
As I wrote about here, what follows from that is that appropriate per-session dosage of work allows you to train productively more often, as the excess fatigue that would otherwise impair your performance is minimised. So whilst doing more than the absolute barest amount is obviously better, there is probably a Goldilocks zone of “enough, but not too much”.
A couple of quick asides. Firstly, whilst doing more than the bare minimum is better (if growing more is important to you), I am actually not certain what the minimum dose IS. In the Krieger analysis mentioned above, the single-set group still grew. The fact that some people actually had success on HIT-style protocols also adds some anecdotal weight to the fact that, up to a point, doing only small amounts of work can promote growth. For rank beginners, also, nearly anything works, and one relatively well-known program for school-age athletes involves doing 1×15-20 of a wide variety of movement every training day, again with abundant anecdotes suggesting it works. So in some instances I suppose the bare minimum is 1 set, maybe not even a particularly hard one.
However, as your training age advances the amount of stimulus required to induce growth very probably increases, and the wicked interplay between how much work required to cause adaptation, how much stress you can induce per set/session and how much work you can tolerate in total makes more undulation, periodisation and general fatigue management probably beneficial, and so I don’t think the minimum stays 1 set for long.
So if we have a minimum being “not much”, how much more is beneficial? Again, cited in my other article about hypertrophy, the Wernbom review (3) on muscle hypertrophy found that 30-60 repetitions per session in the 60-90% 1RM range induced maximum growth, with a plateau past this point.
Unfortunately, counting individual repetitions is not enormously enlightening. Sets near failure between 3ish and 20-30ish reps can actually have a similar number of “hard” repetitions, and similar hypertrophic benefit with wildly different rep counts. I’ll cover that more under “how hard should my sets be”. But if you presume sets of 6-12ish reps, the 30-60 rep figure gives you something like 4-10 sets (which is still a very wide range, but comes close to a high end figure). You could maybe narrow that range a little if you presume that the people doing 60ish reps per workout tended to do more reps per set, too, but without looking at all of the papers individually that’s just an assumption.
In another study, performing 8 sets for a muscle group twice per week resulted in superior gains to performing 16 sets once per week (4), which also lends some support to the notion of an optimal per-session dose being in the moderate range (and in this case, not enormously different from what you’d assume from Wernbom).
On top of that, there is a generally-acknowledged advantage to training muscles more frequently than once per week. Another review by Schoenfield (5) found an advantage to training twice per week vs once, but was unable to detect a benefit to higher frequencies. Similarly, Wernbom mentions similar growth in trained lifters from frequencies of 2 or 3x/wk.
As I’ve written before, I suspect that the timecourse of recovery from an adequately-stimulating bout of training should inform your frequency, meaning that 2, 3 or more sessions per week per muscle group could be beneficial provided that such an arrangement allows you to meet the threshold of stimulus per session required to induce adaptation and recover sufficiently to do so again. Given that muscle force-producing qualities have been observed to stay suppressed for 3 days after 8 sets of pressing (6), I suspect that if you’re doing 4-10 sets per muscle group per session, training each muscle group 2-3 times per week is about right.
So, to sum up, very little training can promote growth. Higher weekly volumes are beneficial, and higher per-session volumes are beneficial, but probably only to the moderate range, with some evidence suggesting that >8ish hard sets per session probably don’t benefit you as much as it would to perform that volume later in the week.
Doing higher volumes per session per muscle group is absolutely tenable, especially if your sets aren’t all that hard, but if they ARE then you probably don’t need to do more than 8 or 10, and trust me, that should feel hard enough.
When you place those figures of 4-10 hard sets against weekly frequencies of 2-3 times per muscle group, that gives you in the realms of 10-20ish hard sets per week.
I’m far from the only person to say that that’s about right for most people (see the recommendations of Lyle McDonald, Eric Helms, Mike Israetael and more, across their respective publications), and I tend to say that if your training dose sees either improvements in direct measures of hypertrophy (which most people don’t have access to) or in multi-set performance in moderate rep ranges for most of your lifts, it’s probably serving you well. Even if you’re NOT growing from that dose, addressing the dark arts of hypertrophy (like movement selection, technique/execution and effort per set) and other factors (sleep, nutrition, stress etc) will often yield you gains where extra sets won’t.
2 – How hard is a hard set? Does this differ for hypertrophy training and strength training
A commonly bandied-about figure for hypertrophy training is that sets should be within 3-5 repetitions of failure. I even mention it here myself. It’s based at least partly off of research which found no increase in EMG magnitude (a measure of muscle recruitment) past that point in sets of 12-15 reps (7).
However, that doesn’t mean that there isn’t benefit to training harder per set than this. Muscles are comprised of lots of little functional units (called motor units), each of which are comprised of a collection of fibers that are under common control. They differ in size and composition and therefore force-producing capability, and your body titrates the amount of muscle (via motor units) recruited to do a task to the force required. It can also modulate force production in other ways. One thing that happens as motor units get fatigued is that you can recruit others to pick up the slack and maintain force production. Whilst the total pool of active fibers mightn’t change much in the final few reps of a set of 12-15 to failure, you can still expose individual fibers to more tension by doing more reps, and it is the tension that each individual fiber is placed under that is the stimulus for growth.
The concept of “effective reps” describes repetitions in which there is “true” or maximal growth stimulus being produced, with the criteria being that the maximum number of motor units has been recruited and subjected to tension for a task. For this reason, sets closer to failure subject you to more effective reps (as stopping 4-5 from failure might only have 1 “effective” rep, whilst stopping 2 from failure would subject you to 3, per the above example).
Note that given that people can grow, at least to some point, from training that is very very submaximal, “effective” reps is probably a slight misnomer. For instance, beginners can grow from training that is super duper easy, powerlifters commonly do repeated sets of 10 with 16-20RM loads on the competition lifts and grow muscle between comps (albeit fairly inefficiently, IMO, although that’s another article) and, as a more subtle point, compound exercises stimulate growth in all of the involved muscle groups to some degree even where only part of the system truly “fails” and produces maximal force.
On an also-related note, extremely powerful movements can also require high levels of recruitment and force production, but don’t really seem to make people grow (think jumping or throwing), and so slower contraction speeds are in some manner important. That doesn’t mean that deliberately manipulating lifting tempo is beneficial (beyond simply controlling the weight, it isn’t), but the slowdown inherent in sets taken to a highly-fatigued state or sets that are heavy enough that lifting with maximal intent to speed is still slowish matters.
To very quickly address heavy lifting and hypertrophy – low rep sets with heavy weights are an absolutely effective hypertrophy stimulus. Past a certain relative intensity motor unit recruitment does not increase to produce more force. Instead, rate-coding (the speed of successive impulses to contract) does. This means that for every rep above about 80-85% of your 1rm in a movement, you’re using all available fibers from the get go. Or, put another way, if all of your lifting is with 5RM loads, then by definition every rep is <5 from failure, one of our criterions for saying a set is an efficient use of your time.
Given that, it’d be easy to argue that just lifting heavy all the time is a good idea. However, the disproportionate intra and post-session fatigue from repeated heavy lifting, the injury risk, near necessity to use movements that allow load-shifting between muscle groups (ie a possible reduction in “targeted tension” – more below), longer training/rest times and potential for staleness makes that probably not the case. Instead, it’s enough just to know that even without a sick pump you’re inducing significant growth-signalling.
Either way, for a set to be worth much for hypertrophy, or at least an efficient use of your time, it probably needs to be within 3-5ish reps of failure. Given that people tend to be bad at estimating reps in reserve, particularly in higher repetition sets, and given that data is derived partly from people who were performing sets to failure (ie, 3-5 reps from failure was reverse-engineered from performance, not based on the subjects’ estimations of difficulty) and given that in a research setting even volitional failure (stopping because you can’t do more) is delayed via encouragement from researchers etc, that actually amounts to a pretty hard set.
Furthermore, given that going closer to failure is probably better on a per-set basis, being very wary NOT to undershoot difficulty means that most of your hypertrophy training should be actually quite tough on a set-by-set basis.
However, just as being too far from failure makes training inefficient on a set-by-set basis, being too close has drawbacks too. Similarly to training with near-maximal loads, reaching failure tends to accrue fatigue within and between sessions at a higher rate, such that the tradeoff in total effective work done is actually negative.
That is to say that most people can probably get more effective reps done over time and progress for longer by being 2ish reps from failure than they could were they to take sets to failure every time.
This may involve doing an extra set (eg 4×8 with 2 RIR as opposed to 3×10,8,6 w 0RIR) and yet still carry less fatigue cost. As I said when I first mentioned effective reps, I think the term is a slight misnomer, as even the “ineffective reps” can promote growth. The reps prior to maximal motor unit recruitment in the example above still subject fibers to tension and so still provide some hypertrophic stimulus. Training methods such as myo-reps (or drop sets etc) that aim to maximise effective reps by enabling more work to be done under fatigue are plainly effective, although I suspect that there is still some advantage to doing straight sets for a similar number of effective reps where time and motivation permits.
Despite me saying that relatively moderate volumes of pretty hard sets probably work best for growth, there are a couple of recent studies showing good growth responses from much higher volumes. See this writeup, for one.
My simplest interpretation of this, and the abundant anecdotal evidence of people experiencing good growth with very high volumes, is that if sets are not truly proximal to failure, greater total volumes are likely needed for an equivalent result. Note that in the above writeup subjects rated most sets as 3-5RIR across the study, however I suspect that most of the sets initially fell short of that (truly pushed, most people can get >15 reps with 60% 1RM for at least one set) and that later they were closer to true difficulty measures. That said, the subjects still grew soundly with the initial volumes with sets at that difficulty, so perhaps I’m wrong in my estimation of the difficulty of the individual sets, or perhaps as I wrote before, even “ineffective sets” have some benefit.
Either way, to conclude on how hard sets should be, for the best results I think they should be hard. How hard? Sets that are 3-5 reps from failure or harder are probably the best use of your time and provide the most stimulus, although probably slightly easier sets than that still help, especially if you’re doing lots of them.
How do you know that sets are hard enough? Luke suggests that an involuntary slowdown of rep speed is a good indicator that you’ve reached a fatigued state and are within the target difficulty range.
In a practical sense, if you match the number of reps you do set to set (ie if you bench press 100kg for 3 sets of 8) the difficulty of each set should increase. In the first set you might do 1 slow rep, and by the 3rd have ground out 3. Or, the number of reps that you achieve across your sets should fall (such that you might do 100kg for 10, 8, 6, each with 2-3 slow reps).
3- Movement selection and hypertrophy
Jacob Schepis, of JPS Health and Fitness, also discussed this with us on Weakly Weights and so if you’re interested in his thoughts you should give it a listen.
This also has the potential to be an article in and of itself, and so I’ll instead keep it short.
As I’ve mentioned above, compound movements promote growth in all (or at least many) of the muscle groups involved in their execution, despite a minority actually being subjected to maximal tension. This makes them a very time-efficient way in which to train, and there is some research to suggest that the addition of isolation work to compound exercises does not promote additional growth (8, 9). I also recall recent research, however, suggesting that exercise variety does have some growth and strength improving benefit, although for the life of me I can’t find it. Certainly, there is research suggesting differences in regional hypertrophy for training different joint actions for biarticular muscles and ranges, though, and so having an extremely limited exercise pool is likely suboptimal. Plus, there are practical reasons why isolation work may be beneficial, as I’ll explain.
Compounds can be great. However, they can have some drawbacks in their use for hypertrophy training, particularly more complex ones such as the squat and bench press (as opposed to the leg press or machine chest press, which limit degrees of freedom somewhat).
Taking the squat as an example, under fatigue lifters tend to compensate by shifting load from one muscular system to another. The classic example is the good-morning squat in which knee-extension demands are shifted to the hip by having the knees kick back and the body fold forwards (in this instance, work of the quads is reduced and instead taken up by the hips).
It takes a lot of control and technical awareness to minimise this, and so the ability to truly target tension to a given muscle group is reduced. Whether this is important is another question. Other than poor motor learning/execution, one of the main drivers of this type of technical shift is that the targeted muscle group simply cannot produce the required force (meaning it is still producing maximal force, or being subjected to tension during the movement, at some stage). However, by shifting away from the weak-link muscle group it does reduce the duration of the movement for which it is subjected to tension, and therefore probably the growth signal.
This then raises the question of how we ought define RIR in this instance. If you are squatting for the quads, is 0RIR where you cannot complete another squat, or where your quads give out and your squat begins to resemble a good-morning? If the latter, what degree of technical breakdown is acceptable, and do the reps past the “failure” point provide no additional stimulus?
The flipside of the difficulty in targeting stimulus to just one muscle group, in as far as that is desirable, is that other muscle groups are fatigued by compound movements. In certain instances, this carryover fatigue can impact performance in exercise that ought target them. In others, again especially as pertains to the powerlifts, the additional fatigue from axial loading and postural work limits the total amount of training that you can do and recover from.
The other end of the spectrum entirely would be to do all single-joint work (or at least, all work that is highly targeted and minimises extraneous fatigue). The advantages are exactly that – easily dosed stimulus for individual muscles, the ability to ensure targeted/maximal tension across all of the reps of an entire set (to a greater degree – people can still compensate in isolation movements quite a lot – consider the shifting of pelvic position people do during leg extensions/curls and the tipping of the shoulder joint you see when people do bicep curls) and minimised extra postural stress. However, the downsides are greater training times, the inability to induce growth in multiple muscles at once, and boredom.
That means there is an inherent tradeoff in all movement selections between their fatigue cost and its dispersal, the specificity of the exercise (whether in powerlifting terms, or its specificity to a given muscle group) and its efficiency in time and training multiple muscle groups.
For powerlifters, due to the need to practice technique and the higher fatigue cost of the competition movements, I tend to think that performing mostly moderate-difficulty sets of the competition lifts and then doing targeted higher-difficulty work biased either towards single-joint lifts or compounds w less extra fatigue cost is good practice. The tradeoff for higher specificity of hypertrophy work tends to be a reduction in its recoverability, higher dispersal of fatigue within and between muscle groups, and more difficulty in targeting the stimulus to weakpoints/weaker muscle groups, although it remains absolutely workable.
For pure bodybuilding, however, the choice should be more along the lines of deciding which exercises provide tension to a muscle group along a long ROM (or particular, unaddressed line of pull), are safe to execute, allow progression of load/volume over time, do not have barriers to their execution due to their complexity/difficulty and do not unduly fatigue other muscles, or at least not to a degree that you consider unenjoyable.
So there’s answers to 3 common training questions.
If you LIKED this article, join my mailing list – it’s FREE and I can send you stuff directly to your inbox. You can also follow me on Instagram, where I frequently post training and diet analysis and advice on my stories.
I also have a podcast, Weakly Weights, available on iTunes and Podbean, where I discuss training for powerlifting in-depth. You can join our mailing list, too, for free sample programs when we discuss them.
Finally, I have now made home workout templates available for sale. They’re designed to be used with any combination of bodyweight, bands, kettlebells/dumbbells and weighted backpacks. They individualise volume prescriptions and offer feedback on when to progress your training to get the most out of limited equipment. You can find them here.
1 – https://www.researchgate.net/profile/Brad_Schoenfeld/publication/305455324_Dose-response_relationship_between_weekly_resistance_training_volume_and_increases_in_muscle_mass_A_systematic_review_and_meta-analysis/links/59dc0269458515e9ab4527d6/Dose-response-relationship-between-weekly-resistance-training-volume-and-increases-in-muscle-mass-A-systematic-review-and-meta-analysis.pdf
4 – – Effect of Resistance Training Frequency on Neuromuscular Performance and Muscle Morphology after Eight Weeks in Trained Men
J Strength Cond Res. 2018
6- “Dissociated Time Course between Peak Torque and Total Work Recovery Following Bench Press Training in Resistance Trained Men.”
Physiol Behav. 201