1. I don't think the authors can make such a broad conclusion, given the study design. The only conclusion that can be drawn is that 4 sets of failure at 30% 1RM stimulates more protein synthesis than 4 sets of failure at 90% 1RM. This result doesn't surprise me given the much greater volume of work in the 30% 1RM group.

I know Stu Phillips personally and worked with him on a protein advisory committee. He does very good work, but I'm surprised they did this study in the way they did. Kumar et al (2008) already have a study like this, and it was better in the sense that they kept the volume load equal between conditions. They also did leg extensions, and they found an increasing protein synthesis response as you went from 20% 1RM to 60% 1RM, with no further increase beyond that.

However, when you look at Kumar's data, the young subjects showed a trend for increases up to 75% 1RM until a plateau was finally reached; I think the lack of significance in the 60-75% range was more an issue of statistical power. So that data would indicate that the middle range that is so popular likely is optimal for hypertrophy.

Another problem with this particular paper is they only looked at protein synthesis. However, hypertrophy is the result of protein synthesis exceeding protein degradation. Really you need to look at net protein accretion and not just the synthesis response in isolation. Without protein degradation data it's really hard to say which one did better.

I think this paper does indicate that there is value to "light" days, and you can obviously stimulate protein synthesis even if you're not using heavy weights. I think too many people are caught up in the notion that every work out must be heavy to get benefit out of it.

2. In terms of legs, there certainly is a lot of anecdotal evidence that people tend to do better with slightly higher repetitions and volumes for the legs. However, I don't know if anyone has truly put this to the test experimentally. Intuitively, it makes sense...since you're walking around on your legs all day, it takes a bigger stimulus to the system to see an increase in protein synthesis. Like you, Martin, I've tended to do better with legs in the 10-12 rep range while better with upper body in the 6-10 range.

1. There are a few caveats with this study. The first and most obvious one is that it was a short-term study only looking at markers for protein synthesis, to conclude anything at all you would need a longer-term study showing actual muscle accretion. There are studies showing how endurance training can lead to rebound MPS (muscle protein synthesis or myofibrillar protein synthesis), but you don’t really see any crazy hypertrophy in these protocols over time, do you?

Another one is the mode of exercise. Leg extensions are quite a different animal than e.g. squats or leg presses and I would like to see someone surviving 4 true sets to muscular failure with squats at 30% of 1RM.

With leg extensions, you can keep constant tension on the muscle, and thus reach a hypoxic state from the occlusion effect. We know that the occlusion effect leads to higher fiber recruitment and more metabolic/oxidative stress. So you really have to look at ‘effective’ reps in this protocol.

The first reps of the first set are essentially just work needed to reach this state of higher recruitment, the latter sets will reach higher recruitment levels earlier. This is exactly what my Myo-reps protocol is based on, and you would probably need do a lot less total reps to achieve the same effect, if the rest periods between sets were 10-20 seconds instead of 3 minutes. The 30%FAIL condition probably ended up around 50-70 ‘effective’ reps, whereas all 20 reps in the 90% group were ‘effective’ from a recruitment point of view.

All in all, the study shows that there is value in metabolic type training, but with the lack of actual long-term measurements of hypertrophy, I wouldn’t disregard the heavy stuff we all know works for getting big and strong. And remember the Goto et al study where the combination group of heavy + light had even better results than the heavy or light group. So do both.

2. Yes and no. You still need heavy progressive loading with sufficient volume and frequency to get a muscle to grow. The problem with legs is that it’s painful to push yourself closer to failure, and most people (yes, me too) will usually chicken out just when it begins to hurt, way before hitting true failure. And be rewarded with chicken legs for the effort.

For safety reasons, I wouldn’t want people to get stuck with their knees behind their ears in the bottom position of a leg press or at the bottom of a squat with their spines sticking out of their back, either.

So while you can argue that legs probably need more work since you walk around on them all day, it’s also a matter of doing more reps and sets to compensate for the inability to get enough ‘effective’ or ‘quality’ reps in any given set.


Alan Aragon





1. I'll echo Borge's mentioning that this was an acute study whose long-term effects are strictly speculative. Additionally, I'm baffled about the load intensities the authors chose to compare (30% vs. 90% of 1RM), especially for the purpose of investigating what might ultimately be better for hypertrophy (as opposed to performance measures like strength or endurance).

A detailed literature review by Wernbom et al suggested that 60-85% of 1RM tends to be the most effective at causing hypertrophy [1]. Specifically for the quads, they found the highest rates of hypertrophy occurred at intensities over 60% of 1RM. Importantly, this review examined the results of research lasting well beyond the acute phase, where measuring changes in muscular cross-sectional area are possible.

This conflict of data makes the present study's short-term outcomes highly questionable. In principle, this reminds me of an acute study by Deldicue et al, who found that fasted training increased molecular markers of anabolism to a greater extent than training in a fed state [2]. While this data is interesting, we just can't firmly conclude anything concrete from it. It's still speculative whether or not these were merely compensatory responses to a suboptimal protocol.

2. This varies with the individual, but I've found that it's best to go with a combination of what's been seen to work in the field as well as the research, taking sets to fatigue at roughly 6-12 reps (although Wernbom et al suggest 8-10 is best for the quads, which is appx 60-85% 1RM). As long as an uptrend in strength (increases in reps and/or load) is maintained over time in this range, hypertrophy will occur.

Bodybuilders have traditionally stuck to higher rep ranges for quads, regularly going into the teens. However, I think the increases in mass are far more attributable to progressive overload with sufficient total volume rather than the rep range per se.

For building hamstring and glute mass, I haven't personally experimented with reps outside of conventional ranges. For calves, I hate to sound fatalistic, but think that it's a matter of having picked the right parents. This is not to say that calves are impossible to bring up if they happen to suck; it's just that in my observations, people predisposed to crappy calves tend to struggle equally regardless of the load intensity imposed.

1) Wernbom M, et al. The influence of frequency, intensity, volume and mode of strength training on whole muscle cross-sectional area in humans. Sports Med. 2007;37(3):225-64.

2) Deldicue L, et al. Increased p70(s6k) phosphorylation during intake of a protein-carbohydrate drink following resistance exercise in the fasted state. Eur J Appl Physiol. 2009 Nov 18.


Lyle McDonald



1. Well, after Blade and Alan's comments, I don't really have much to add to the topic. I too would have liked to have seen an intermediate intensity group (either matched or not matched for volume, possibly both). Certainly other work (e.g. the Wernbom review that Alan referenced) suggests that intermediate intensities are probably better than either extreme.

As well, with an acute study like this, it's hard to say what the long-term results would be in terms of growth. I agree that it is at least interesting given the bias most have towards heavier work for hypertrophy.

2. Empirically, many do seem to think that legs (or at least quads) respond better to higher reps. I have to think that part of this may be biomechanical especially in terms of squats. For some people, heavy low-rep squats turn into a pseudo good morning meaning that it's mainly low back stress and/or more stress thrown onto the glutes and hamstrings. For those people, higher reps allow a more upright torso meaning the opposite: less low back stress and more thrown onto the quads.

In keeping with that, Olympic lifters (for whom high reps in squats are 5) usually have pretty decent sized legs; of course most of them are also built to squat (and squat high bar).

Beyond that, like Alan, I can't say I've ever paid much attention to training glutes or hamstrings differentially to quads, typically using some mix of lower rep work (5-8, sometimes lower) for heavy compounds and following it up with additional higher rep work (10-12 or more) on more isolation work.

I do agree that 20 rep squats can be amazing but as you point out, it's more of a heavy rest-pause approach than what was being described in the paper in question.

---

1. 4 sets to failure at 30% 1RM increases protein synthesis more than 4 sets to failure at 90% 1RM. This is not surprising given the much greater training volume in the 30% 1RM group. The 30% 1RM group almost did five times as many reps as the 90% 1RM. 5. As for our personal experiences, no one outright claimed greater results with either low reps or high reps exclusively. If anything, there was a tendency to favor intermediate rep ranges or using low(er) and high(er) reps in combination.

James: "I think this paper does indicate that there is value to "light" days, and you can obviously stimulate protein synthesis even if you're not using heavy weights. I think too many people are caught up in the notion that every work out must be heavy to get benefit out of it."

Good point which goes in line with my own experiments with high frequency training.

Borge: "The problem with legs is that it’s painful to push yourself closer to failure, and most people (yes, me too) will usually chicken out just when it begins to hurt, way before hitting true failure".

Very true. I can count the people I've seen squatting close to their limit in the 10 rep+ range on...two fingers.

Alan: "In principle, this reminds me of an acute study by Deldicue et al, who found that fasted training increased molecular markers of anabolism to a greater extent than training in a fed state [2]. While this data is interesting, we just can't firmly conclude anything concrete from it."

Alan points out that acute response does not predict long-term results. I've actually written an article about the study Alan mentions. Read: "Fasted Training Boosts Muscle Growth?"

Lyle: "For some people, heavy low-rep squats turn into a pseudo good morning meaning that it's mainly low back stress and/or more stress thrown onto the glutes and hamstrings. For those people, higher reps allow a more upright torso meaning the opposite: less low back stress and more thrown onto the quads."

Great point by Lyle. Some people just aren't made for squatting and are better off using the leg press as their main leg movement. Check out Lyle's article on this topic: "Squat vs Leg Press for Big Legs".