Maximizing Muscle Growth

At The Gym

• Untrained people should work out 3 or more days per week. Trained people can get away with 2. dose response for strength development. Training twice per week is better than training once per week, and this probably remains true even after controlling for weekly volume Effects of resistance training frequency on measures of muscle hypertrophy: a systematic review and meta-analysis.Training 2 or 3 days per week is better than training once per week, but this result stops being significant after controlling for volume Grgic - still the sign in this study is in the right direction, so both studies show the balance of the evidence suggests working out multiple times per week is better than working out once per week.
• You should perform about 4 sets for each exercise dose response for strength development. Multiple sets causes 40% more growth than a single set, and this gain seems almost entirely concentrated in the second and third sets Single vs. multiple sets. Multiple sets yield greater gains than a single set Bågenhammar Durall.
• Resting 3 minutes between sets promotes more growth than shorter durations rest periods. Other research says the rest interval doesn't matter de Souza Jr.
• Repetitions lasting 0.5 - 8 seconds have similar effects; reps lasting more than 10 seconds might be less effective repetition duration.
• There's no reason to train to failure repetition failure.
• Novices should lift as many reps as they can while lifting 60% of their max, while trained lifters should lift 80% dose response for strength development. According to the Brzycki formula, this corresponds to about 23 and 8 reps, respectively.
• Trained lifters should lift as many reps as they can while lifting 85% of their max Maximizing strength development in athletes. According to the Brzycki formula, this corresponds to about 5 reps.
• The free-weight bench press activates more muscles than the machine bench press, weakly suggesting that free-weights might prove better at stimulating muscle growth than machine exercises McCaw.
• Full-body workouts appear as good as split-body workouts Ratamess Benton. However, its plausible split workouts can allow for higher volume given the same amount of effort, which could plausibly make split workouts more effective. More research is needed Effects of resistance training frequency on measures of muscle hypertrophy: a systematic review and meta-analysis.
• Working out once per week may be enough to maintain muscle mass Trappe. A third the original volume is enough to maintain strength Bickel
• When comparing dumbbells, barbells, and machines, it's complicated... Farias.

In the Kitchen

• Protein up to 1.6 grams per kg of body weight per day improve muscle gains protein supplementation - that's about double the 0.8 recommended for the general population. A literature review recommends higher levels for people simultaneously losing weight: between 2.3 and 3.1 dietary protein during caloric restriction g/kg/d.
• Protein supplements work Protein supplementation, especially whey advantage whey,
• Three studies have found that there is a limit to how much protein you can consume in one meal before its effect on muscle growth goes away. The estimates for this threshold generally fall between 20g and 30g of protein moderate serving of high-quality protein Ingested protein dose response of muscle and albumin protein synthesis after resistance exercise in young men Protein ingestion to stimulate myofibrillar protein synthesis requires greater relative protein intakes in healthy older versus younger men. However, a study in older adults did not reproduce this result, suggesting that benefits exist far past this 30g limit for that demographic Quantity of dietary protein intake, but not pattern of intake, affects net protein balance primarily through differences in protein synthesis in older adults.
• Consuming protein around the time of your workout has no effect on muscle growth except insofar as it may boost your total protein consumption protein timing.
• You should definitely take creatine supplements, which have been found to increase muscle mass by something like a sixth of a pound per week among older adults. The only side-effects found were minor gastrointestinal issues creatine supplementation.
• Caffeine doesn't increase strength Astorino.

Though I haven't found convincing evidence on muscle mass or strength, some other supplements do boost athletic performance:

• Exercise lasting more than 60 seconds is improved with β-alanine consumption Hobson. These improvements were later confirmed Elliott-Sale. It is generally safe Painelli.
• "Citrulline supplementation confers a significant performance benefit for high-intensity strength and power tasks in comparison to placebo, with a pooled SMD of 0.20 [0.01, 0.39]" Trexler.
• Nitrate supplementation improves exercise tolerance but not performance Van De Walle.

Literature Reviews

[TODO: Integrate this section into the "At The Gym" section]

One literature review Manipulating resistance training program variables to optimize maximum strength in men: a review found

Another literature review is (TODO): Bird

A third literature review (TODO) Wernbom concludes

General Health Recommendations

Creatine Dosage

I couldn't find any meta-analyses on the optimal creatine dosage so I decided to take a look at the data on my own. I looked at the creatine meta-analysis creatine supplementation and assumed the average American weight of ~180 lb to convert between studies using creatine in g vs. g/kg. The resulting data can be viewed here.

I looked at the three metrics included in the study (lean body mass, chest press, leg press) and analyzed how the effect size (d) changed as dosage changed. I found no statistically significant difference between low- and high-dosage creatine usage and found point estimates with both negative and positive signs depending on the metric and whether I used regression or a difference-in-means test. In other words, I found no real evidence that dose mattered. Note: this analysis was low-powered because (a) the effect estimates had significant variance (b) I only had ~27 estimates and (c) the doses fell in a relatively narrow range (3.4 g/d - 9 g/d).

Next I looked at the actual changes in those three measures (rather than effect sizes). I repeated my analysis of dosage with these measures and found similarly weak and mixed results.

Finally, I found a linear relationship between changes in the placebo group and changes in the creatine group and the slope was ~1 for all three measures. This suggests that creatine merely increases your muscles/strength by a fixed amount, as opposed to making working out more effective. The mean difference was ~3 extra pounds of muscle mass and increased chest and leg press strength by ~12 and ~15 pounds, respectively.

If you're interested in looking at the data yourself, you can either see the meta-analysis or this convenient spreadsheet i put together.

Suspicious Studies

Now I want to move on to the controversial part of this page where I disagree with the academics with decades of experience. I think several of these meta-analyses suffer from significant problems.

Take dose response for strength development, for example. The authors found that trained athletes saw smaller returns from exercising 3 times per week rather than 2.

However, I'm suspicious of this study. It didn't look at a bunch of studies that investigated this question. Instead, it looked at a bunch of studies that investigated weight-lifting generally and compared how much strength increased relative to the group's standard deviation.

This choice of analysis doesn't make much sense to me. Suppose the meta-analysis had two studies:

1. A set of 90-year-old apathetic women do 2 sets of bench press 3 times per week. Each lift is 3 inches. Another similar group is the control group.
2. A set of driven 20-year-old men do 1 sets of bench press 2 times per week and are given 90g of protein as supplements. Each lift is 2 feet. Another similar groups is the control group.

It's all but certain the second study would find a larger effect size due to the demographic and procedural differences. The analysis in dose response for strength development would then conclude that 1 set is better than 2, and 2-days-per-week is better than 3. This is absurd reasoning.

They mention that they also did an analysis based on gender and age, but (1) they don't present this analysis and (2) this still ignores other factors such as protein intake.

The analyses Effects of resistance training frequency on measures of muscle hypertrophy: a systematic review and meta-analysis, Maximizing strength development in athletes, and Grgic also suffer from the same problems.

Instead, I think we're better off finding studies that keep the same sets and reps and just change the number of days worked per week.

I should note that some of the studies don't fall victim to these issues. For instance, Single vs. multiple sets and Bågenhammar both look at studies with groups that differ only in one variable and look at the differences in muscle growth between the groups.

At some point I should come back to this and re-examine the other meta-analyses (TODO).

These short-comings are a big problem since these were the studies we were relying on for optimal reps and optimal days-per-week estimates. For this reason, I think it's worth diving into these two topics a bit more.

Optimal Number of Reps (Bench Press)

This table lists studies where participants performed 3 sets of the bench press 3 times per week. "X > Y" means that performing X reps led to greater gains than Y reps. Some of the results are not statistically significant.

It looks like 3-9 reps is probably around optimal.

Optimal Number of Reps (Biceps Curl)

Optimal Number of Reps (Legs)

It looks like ~9 reps is around optimal.