When does training volume reach the point of diminishing returns?

Back in 2017, Schoenfeld and colleagues published a meta-regression suggesting that there was a roughly linear dose-response relationship between training volume and muscle growth, up to ~10 sets per muscle group per week: performing 10+ sets per week resulted in more muscle growth than performing 5-9 sets per week, and performing 5-9 sets per week resulted in more muscle growth than performing <5 sets per week. However, when that meta-regression was published, everyone understood that this roughly linear dose-response relationship couldn’t be extrapolated infinitely. Logically, there must be some point at which further increases in training volume don’t result in more muscle growth. However, the vast majority of the research to that point had only investigated relatively low training volumes (<10 sets per week) so we lacked the data that would be required to determine the point of diminishing returns.

In recent years, however, more and more studies have investigated the effects of training volumes exceeding 10 sets per week. In fact, these days, 20+ sets per week isn’t uncommon for a “high volume” group in a study designed to investigate the impact of training volume on muscle growth. With that in mind, Baz-Valle and colleagues recently performed a meta-analysis looking at the impact of training volume on muscle growth, beyond the upper bound of the original Schoenfeld meta-regression. Specifically, they wanted to compare the effects of weekly training volumes of 12-20 sets per muscle group per week versus weekly training volumes of >20 sets per muscle group per week. To be included in the meta-analysis, a study had to meet these criteria:

1. The study needed to include at least one group training the quads, biceps, and/or triceps with 12-20 sets per muscle group per week, and at least one group training the quads, biceps, and/or triceps with >20 sets per muscle group per week.
2. The study needed to use the same training intensities for all groups; weekly set volume was the only difference in training prescription between groups.
3. The training interventions needed to last at least 6 weeks.
4. The subjects needed to have at least one year of training experience, and be between 18-35 years old.
5. The study needed to directly assess hypertrophy. In other words, it needed to measure changes in muscle thicknesses or cross-sectional areas, rather than merely assessing changes in fat-free mass.
6. The study needed to be published in a peer-reviewed journal.

Six studies met these inclusion criteria. They’re summarized in Table 1. From this figure, you can see why the 2017 Schoenfeld meta was unable to provide recommendations for volumes exceeding 10 sets per week – five of the six studies investigating relatively high training volumes were published in 2017 or later (and we’ve reviewed three of those five studies in MASS; one, two, three).

After identifying these studies, the researchers performed a series of random effects meta-analyses comparing the effects of 12-20 versus 20+ sets per week on quad growth, biceps growth, and triceps growth. The results of these meta-analyses can be seen in Figure 1.

As you can see, 20+ sets per week resulted in significantly more triceps growth than 12-20 sets per week (SMD = 0.50; p < 0.05). Quad growth also tended to be greater with 20+ sets per week than 12-20 sets per week (SMD = 0.20), but the difference wasn’t statistically significant. Finally, biceps growth wasn’t meaningfully different between the two weekly volume categories.

The present meta-analysis is a bit of a Rorschach test. You could use it to credibly argue that 20+ sets per muscle group per week are generally advisable for people aiming to maximize hypertrophy. All pooled effect sizes at least leaned in favor of higher volumes producing more muscle growth, and the difference in triceps growth cleared the threshold for statistical significance. Furthermore, the study with the longest training intervention (the study by Radaelli and colleagues lasted 24 weeks; all other studies lasted 6-8 weeks) reported the largest effect sizes in favor of higher training volumes, which may suggest that the rest of the studies underestimate the “true” superiority of higher training volumes. However, you could also use the present meta-analysis to credibly argue that 12-20 sets per week are plenty, and drifting up toward 20+ sets per week is likely unnecessary. Two of the three comparisons reported trivial effect sizes (SMD ≤ 0.20), and the statistically significant triceps comparison may have been influenced by how the researchers accounted for training volume. In the present meta-analysis, a set of a compound lift that targets a particular muscle was counted as a set of training volume for that particular muscle. So, for example, sets of pressing exercises were counted as sets of triceps training, sets of rows and pull-downs were counted as sets of biceps training, etc. We know that pressing exercises and single-joint triceps exercises target the triceps differently, and we also know that single-joint exercises sometimes result in more muscle growth for a particular muscle than compound exercises that train the same muscle. So, it’s possible that >20 sets of triceps training only results in more triceps growth than 12-20 sets of triceps training when a hefty proportion of your triceps training comes from pressing exercises, because the stimulus provided to the triceps per set is slightly submaximal. If that line of thinking is correct, it’s therefore possible that 12-20 sets of direct triceps training (skullcrushers, push-downs, etc.) would be sufficient to maximize muscle growth, and doing >20 sets would provide no additional benefit.

My personal take is somewhere in the middle. I think this meta-analysis, on the whole, suggests that doing >20 sets per muscle group per week will likely lead to a bit more muscle growth than doing 12-20 sets per muscle group per week, but I also think it suggests that ~20 sets per muscle per week is approximately the average point of rapidly diminishing returns for trained lifters (the threshold for novice lifters is likely quite a bit lower). To be clear, diminishing returns aren’t zero returns; rather, “diminishing returns” implies that it takes more and more of a particular input (training volume, in this case) to move the needle on your desired outcome (muscle growth). Remember, the Schoenfeld meta-regression suggested that there was a roughly linear dose-response relationship between training volume and muscle growth with relatively low training volumes (<10 sets per week), but the present meta-analysis suggests that with higher set volumes, the marginal benefit of each additional set is smaller. In other words, if you double your training volume from 4 sets per muscle group per week to 8 sets per muscle group per week, your rate of muscle growth might approximately double. However, if you double your training volume from 15 sets per week to 30 sets per week, you probably won’t double your rate of muscle growth; maybe your rate of muscle growth would only increase by 20%. Basically, doing more than 20 sets per muscle group per week may help you grow a bit more, but you’ll need to weigh the costs and benefits for yourself. If you’re fine doing twice as much work for slightly better results, then more power to you; however, if you value efficiency (maximizing results per unit of time and effort invested into training), weekly volumes in the 12-20 set range may be more appropriate.

More importantly, though, I’d encourage you to use this meta-analysis as a starting point for your own troubleshooting. Knowing the amount of training volume that works the best, on average, for the subjects in the six studies included in this meta-analysis is perfectly fine, but we also know that individual volume needs can vary considerably, and personalizing training volume tends to produce better results than one-size-fits-all prescriptions. Research provides us with a starting point, but directed self-experimentation is necessary to optimize training volume for the individual. You may find that you need considerably more or considerably less training volume than the average person to maximize your results. If you’re not making progress and you feel worn down, there’s no harm in reducing your training volume, even if that means doing less volume than the research suggests you “should” be doing. Conversely, if you’re not making progress and you still feel fresh most of the time, there’s no harm in increasing your training volume, even if that means doing more volume than the research suggests you “should” need.

Note: This article was published in partnership with MASS Research Review. Full versions of Research Spotlight breakdowns are originally published in MASS Research Review. Subscribe to MASS to get a monthly publication with breakdowns of recent exercise and nutrition studies.