For what it's worth, this will have no impact on my own training, but it might impact how I work with newbies at the gym.
Research Review: Is More Really Better? Comparing 1 Set to 3 Sets for Building Strength & Muscle Mass
Ever wonder if you’re wasting your time at the gym?
If you routinely debate other gym members about the effectiveness of government spending on economic recovery during your workout –- then yes, you’re wasting your time (and probably theirs, unless they’re federal auditors). But flagrant time wasting isn’t what I’m talking about –- I’m talking about wasted sets. I want to know:
What is the minimum number of sets I need to do to see gains in strength and muscle mass?
Since I haven’t done a “training” review in this column, I thought I’d jump right into a controversial subject. Most of you have probably heard of a strength training technique called high intensity training, or HIT (not to be confused with high intensity interval training, aka HIIT, which is a cardiovascular training technique). The basic method of HIT is to train with very low volume but to take each set to absolute fatigue or “failure”. While not the norm for HIT, in the most extreme cases this method utilizes only 1 set per exercise, and that’s what I’m going to look at in this research review.
Many studies have looked at single versus multiple sets, but none of them kept as many training variables constant as this one did. This study made sure that 6 out of 7 training variables stayed the same: exercise selection, exercise order, frequency (how many times/week), load (aka intensity), repetition, and rest. Only the loading volume –- i.e. the number of sets –- changed. This way, researchers could truly compare the difference between 1 set and 3 sets.
Rønnestad BR, Egeland W, Kvamme NH, Refsnes PE, Kadi F, Raastad T. Dissimilar effects of one- and three-set strength training on strength and muscle mass gains in upper and lower body in untrained subjects. Journal of Strength and Conditioning Research 2007 Feb;21(1):157-63.
This experiment used 24 untrained men who were, on average, 26.5 years old. “Untrained” was specific to strength training and was defined as training fewer than 3 times per month. The fact that these guys were untrained is very important and it likely impacts the outcome of the study — more on that later.
Generally, most training studies have untrained participants. There are two common reasons for this. First, it’s a lot easier to find participants that are untrained –- they are a larger part of the population. (Unfortunately.) Second, people who are trained are much less willing to change their existing training to do what the study prescribes. Most trained individuals have gotten comfortable with a certain type of training program, and they don’t like to try something different that might limit their progress or what they might perceive as inferior.
All the study participants did 3 workouts a week over 11 weeks. The number of repetitions (reps) changed over the 11 weeks.
Weeks 1-2: 10 reps/set
Weeks 3-4: 8 reps/set
Weeks 5-11: 7 reps/set
(That’s what the authors reported, but I calculate 10 weeks total (2wks@10 reps, 2wks@8 reps and 6wks@7 reps.) Either way there was a progression of fewer reps and increasing load over the course of the training.
In order to compare 1 set versus multiple sets, each participant was randomly assigned to one of two groups: the 1L-3UB (1 lower body set – 3 upper body sets) or the inverse 3L-1UB (3 lower body sets – 1 upper body set) group. Here’s a chart of what the two groups had to do:
Exercises 1L-3UB set breakdown 3L-1UB set breakdown Leg press 1 set 3 sets Leg extension 1 set 3 sets Leg curl 1 set 3 sets Seated chest press 3 sets 1 set Seated rowing 3 sets 1 set Latissimus pull down 3 sets 1 set Biceps curl 3 sets 1 set Shoulder press 3 sets 1 set
The idea behind this design is that each group would have equal load volume -– meaning if you multiply the number of reps, sets, and load for each exercise and total them, they would be equal between groups.
For example: Leg press 10 reps x 1 set x 100 kg (load) = 1000 kg
You can think of it as the total weight moved during an exercise session. Since most people can move more weight with their legs than their upper body, the study was designed with more upper body exercises to allow for equal load volume while having the same number of sets and reps. This worked fairly well: 1L-3UB had a total training load volume of 9,201 kg while 3L-1UB had a total training volume of 9,676 kg for the first weeks. There was no difference in load volume for the final weeks.
Many of you might be wondering: why not just have one group do 1 set of everything and one group do 3 sets of everything –- isn’t that how people train? But having a design of that nature -– equal sets for every exercise -– would result in huge differences in volume and load volume. The results would be confounded, or muddled up, by the effects of sets and volume. Although it’s a subtle difference, sets and volume are different. Thus, to figure out the importance of sets, volume (specifically load volume) had to be the same.
Researchers measured several things:
- 1 repetition maximums (1RM) on all exercises
- Muscle cross-sectional area (a measure of hypertrophy, aka muscle size increase)
- Lean body mass
- Fat mass
Both lean and fat mass were measured by DEXA (dual x-ray absorptiometry).
Here are the results.
- All participants increased the percent of 1RM of both upper body and lower body exercises over training (0 week compared to 11 week). In other words, they got stronger overall.
- 3L-1UB group had greater lower body gains in 1RM compared to 1L-3UB. 3L-1UB lower body 1RM increased 41% compared to 21% in the 1L-3UB group.
- No differences in 1RM gains in upper body exercises between groups.
- A bigger increase in thigh (knee extensor & flexor) muscle cross section area in the 3L-1UB group compared to 1L-3UB group.
- Participants gained lean body mass (about a 6% change) and lost fat (about a 10% change).
- No difference in overall lean body mass or fat mass between groups, though only 5 participants in each group underwent these tests.
It seems that 3 sets of lower body exercises (3L-1UB) is better than 1 set of lower body (1L-3UB) with more muscle and more strength gains in the 3L-1UB group. But the same thing didn’t happen with 3 sets of upper body exercise compared to 1 set. In that case there was no difference in muscle hypertrophy or in strength between 1L-3UB and 3L-1UB.
Based on this study it seems that multiple sets are “better” for lower body exercises, but more sets aren’t better for upper body exercises. Weird. Isn’t a muscle a muscle? Wouldn’t one muscle respond the same way as another, regardless of where that muscle is?
The authors propose a few explanations for this:
1. “Daily life training effects”
Since you use your leg muscles more regularly than your upper body, your legs are trained. All the walking, standing and climbing stairs that the “untrained” participants did every day may have led to a level of training in the leg muscles that changed how the lower body responded to exercise. Thus, the researchers argue, trained individuals/muscles respond more positively to more sets. Fair enough. While I buy that trained muscle does better with more volume, I don’t know how much I believe that modern daily life challenges anyone in a “trained” state.
The next two explanations consider the role of hormonal receptors and hormones in upper and lower body muscle.
2. Variations in androgen receptors
Upper body muscles have more androgen receptors than lower body muscles. Thus, the researchers suggest, lower body muscles are more sensitive to sets than upper body muscles, because lower body muscles have relatively fewer androgen receptors.
Hmm, I don’t know about this one. The authors are proposing that the muscles that would be most sensitive to androgenic hormones are the least responsive to multiple sets. How are these two variables connected? Could you argue that the upper body is closer to the brain and therefore has greater neural drive and is more effective at muscle recruitment, thus less responsive to multiple sets? I guess, but I don’t think there’s any connection between the two. Just because something is true it doesn’t mean it has any bearing on the experiment. I don’t believe this possibility explains the difference between groups.
3. Hormonal response
The last possibility is that 3 sets of lower body exercise causes a bigger hormonal response than 1 set and these hormones “spill over” to the upper body. 3 sets of lower body work might release more anabolic hormones than 1 set of lower body, and more than 3 sets of upper body. I lean toward this possibility as the most likely. 3 sets of lower body exercise would increase the amount of anabolic hormones (growth hormone and testosterone), which would lead to increases in upper body strength and hypertrophy. So, the similarity between 1 set and 3 sets in the upper body exercises is not because of the upper body exercises but because of the lower body sets.
While this might be true it doesn’t address one problem: that the upper body gains are not the same magnitude as the lower body gains. If the lack of difference between 1 set and 3 sets is hormonal you would still expect proportionally the same gains as seen in the lower body.
If everything else is kept equal 3 sets are better than 1 set for hypertrophy and strength when training the lower body, even in untrained individuals. But for some reason, in untrained individuals 1 set seems to be just as effective as 3 sets for upper body training –- at least for the first 11 weeks.
So if you’re untrained and have a limited amount of time, you can get away with 1 set for your upper body exercises, but you should use 3 sets for your lower body exercises.