Schoenfeld BJ et. al. Effects of Low Versus High Load Resistance Training on Muscle Strength and Hypertrophy in Well-Trained Men.J Strength Cond Res. 2015 Apr 3. [Epub ahead of print]
The purpose of this study was to compare the effect of low- versus high-load resistance training (RT) on muscular adaptations in well-trained subjects. Eighteen young men experienced in RT were matched according to baseline strength, and then randomly assigned to 1 of 2 experimental groups: a low-load RT routine (LL) where 25-35 repetitions were performed per set per exercise (n = 9), or a high-load RT routine (HL) where 8-12 repetitions were performed per set per exercise (n = 9). During each session, subjects in both groups performed 3 sets of 7 different exercises representing all major muscles. Training was carried out 3 times per week on non-consecutive days, for 8 total weeks. Both HL and LL conditions produced significant increases in thickness of the elbow flexors (5.3 vs. 8.6%, respectively), elbow extensors (6.0 vs. 5.2%, respectively), and quadriceps femoris (9.3 vs. 9.5%, respectively), with no significant differences noted between groups. Improvements in back squat strength were significantly greater for HL compared to LL (19.6 vs. 8.8%, respectively) and there was a trend for greater increases in 1RM bench press (6.5 vs. 2.0%, respectively). Upper body muscle endurance (assessed by the bench press at 50% 1RM to failure) improved to a greater extent in LL compared to HL (16.6% vs. -1.2%, respectively). These findings indicate that both HL and LL training to failure can elicit significant increases in muscle hypertrophy among well-trained young men; however, HL training is superior for maximizing strength adaptations.
For years, it’s been taken basically as an article of faith that the best way to stimulate muscle growth is with relatively heavy loads for lower repetitions. The hypertrophy zone is usually defined as 8-12 (5-15 is better) repetitions per set with heavy weights. Muscle fibers are recruited in an order according to the size principle with only the high threshold Type II muscle fibers (the ones most responsible for growth requiring heavy loads of 80-85% for maximal recruitment.
At the same time, in recent years, research has started to suggest that there is more to it than this. Early studies using blood flow restriction (BFR or Kaatsu training) found that relatively light weights at 20% of maximum appearing to recruit muscle fibers and generate growth similarly to heavy weight training.
Similarly, some recent papers have found that relatively low-load training but doing a ton of reps to failure has the same potential to recruit the high threshold muscle fibers as heavier training; growth in these studies was typically similar as well but they all suffered from the same basic flaw which was that they used untrained subjects. We know that beginners pretty much make the same gains no matter what is done and this didn’t really say anything about trained lifters.
Enter the following study sent to me; I want to thank Brad Schoenfeld for sending it to me in pre-publication format. Since I know that this is going to get misrepresented by folks who only read the abstract, I wanted to look at in detail.
24 male subjects with training experience ranging from 1.5-9 years were recruited to take part in the study with 9 subjects in each group. They were matched for strength and then assigned to either a traditional resistance training program of 3 sets of 8-12 repetitions to failure or 3 sets of 25-35 repetitions to failure.
Both groups did the same 7 exercises: flat bench, military press, wide grip pulldown, seated cable row, barbell back squat, machine leg press and machine leg extension and all 7 exercises were performed three times per week for 8 weeks. The workouts were overseen and repetitions were done at a cadence of 1 second up/2 seconds down with weights adjusted to keep trainees in the failure range for the desired repetition range and an attempt to increase weight each week.
The subjects were told to maintain their diet at the same level (this is always a bit of a crapshoot but it’s impossible/expensive as hell to keep everyone in a metabolic ward) and they were given 24 grams of whey protein after training.
Measurements were made of muscle thickness for biceps, triceps and quadriceps to measure changes in muscle size. Maximal strength was measured by 1RM bench and squat and muscular endurance was measured as the total number of repetitions at 50% of 1RM in the bench press.
Basically this was a very well controlled study with the only variable being load/repetition range; everything else was kept the same (or as the same as it could be in terms of diet).
18 of the original subjects completed the study with; 4 for personal reasons and one in each group due to minor injuries.
Of some interest, the increases in muscle thickness were more or less identical between groups: 5.3% vs 8.6% for biceps, 6.0 vs. 5.2% for triceps and 9.3% versus 9.5% for quads for the low- and high-load groups respectively. None of the differences were statistically significant.
The high-load group outperformed the low-load group for bench press 1RM, gaining 6.5% vs. 2.0% while the squat changes were 19.6% versus 8.8% for the squat. Even adjusted for their starting strength levels, the high load group still won out.
In terms of muscular endurance, only the low-load group improved (by 16.6%) compared to the high-load group.
Ok, where to start. It’s pretty clear that heavier weights increased strength more than lighter weights but this is no real surprise given that there is an enormous neural effect for maximal strength. Only the low-load group improved muscular endurance and this also isn’t much of a shock either.
The growth responses are more interesting as there was no apparent difference between the groups (this is consistent with other research on blood flow restriction and low-load training in beginners) although it was mentioned that there might have been differences in Type I vs. Type II fiber growth (not measured).
This idea has been thrown around for years, that different duration sets will preferentially grow different fibers types, I mentioned this in my never finished periodization for bodybuilders series. The researchers also mentioned that the combination of heavier and lighter work might give the best results (especially if there is fiber specific growth occurring) but that wasn’t tested here.
But before everyone runs out and starts pumping out the reps, let’s make a little bit of a reality check here. First and foremost, just as with BFR, both groups gained to the same degree in terms of muscle size with the heavier weights being superior for maximal strength.
But there are some other considerations: it’s interesting that the low-load group did nearly three times as many repetitions as the heavy-load group. They both did 3 sets so that’s 75-105 repetitions for the low-load group versus 24-36 for the heavy load group. You do a ton more work for the same results.
Adding to that is the fact that very high rep training is miserable and painful. It’s like doing UD2 depletion at every workout. At 3 seconds per repetitions, the set times are 24-36 seconds per set for the heavy load group and 75-105 seconds (1:15-1:45) per set.
75-105 seconds is smack in the middle of anaerobic glycolysis and the amount of acidosis generated makes these types of sets just painful as all hell (in private correspondence Brad told me that about half of the subjects in the low-load condition threw up and that’s common with this type of work).
I’ve personally done sets of 50 in the back squat and sets of 25 in the one legged leg press (with a 2 second pause making it about a 2.5 minute set) and I wouldn’t wish either on my worst enemy. Heavy weight work is done before it really starts to hurt. It’s also interesting to wonder what would happen if the heavy group had done 9 sets instead of 3 to generate the same total number of repetitions.
Make no mistake, this is an interesting and well-controlled study clearly showing that the same amount of growth can be generated without using heavy loads. But let’s look at this in real-world terms.
Because despite what will be claimed and the amazing new e-books that will appear based on this, I’d argue that low-load training is far less efficient and effective for generating growth.
You do three times the work at about 10 times the pain level for the same amount of end result growth. And you don’t look macho fooling around with light weights and we all know that getting attention in the gym is the key to greatness.
Like with BFR, that generates no more growth while requiring specific setup (and only being usable by certain muscle growth) and generating excruciating pain, I just don’t see the point from a practical standpoint.
About the only time I could see the use of either would be in the case of injury where high joint forces have to be avoided or possibly by an athlete who needs to improve local muscular endurance (that’s why we did the 25 rep sets in speed skating) on top of muscle growth. For everyone else I’m just not seeing the point.
Thanks again to Brad for sending me this paper ahead of publication and fielding my comments and questions in email. It’s good to see someone doing applied/well constructed studies on strength training for a change.