One of the most common areas of confusion for bodybuilders and other athletes engaged in resistance training is the question of the appropriate or optimal number of reps and sets for any given workout session or cycle.

This confusion is exacerbated by the common observation that strength and power athletes (weightlifters, throwers, and powerlifters) achieve impressive gains in lean body mass using multiple sets of low (generally 1-3) reps,(1) whereas bodybuilders more commonly employ a smaller number of sets using higher repetition schemes.(2,3) Additionally, numerous books and articles by a host of training experts have advocated a wide assortment of set/rep schemes, all of which have worked well for those who have used them.

All of the above leads to the following questions:

1) Is there an "optimal" set/rep scheme for the acquisition of lean body mass?

2) Or, is there a better question? In other words, does focusing on set/rep schemes lead us away from the answers, instead of bringing us closer to them?

Here are a few possible questions that may bring us closer to the answers we seek:

1) If you're learning a new scale on the piano, how many times should you repeat that scale during any given practice session? Do prominent pianists and/or piano teachers advocate various "optimal" practice schemes, such as "25 repetitions per session" in musical trade publications?

2) If you're practicing the tennis serve, how many times should you repeat that technique during any given practice session? Do expert tennis players and/or coaches advocate various "optimal" practice schemes, such as "50 serves per practice session" in the tennis magazines and journals?

3) If you're a 100 meter sprinter practicing the start from the blocks, how many times should you repeat this skill during any given practice session? Do track coaches advocate various "optimal" practice schemes, such as "30 starts per training session?"

In the case of the above questions, are we looking for an arbitrary number of repetitions, or is there an underlying principle or concept which would lead us to the appropriate number of repetitions or attempts? The answer to this question, I believe, lies in the relationship between volume and intensity.

We all recognize that intensity and volume are inversely related, but how often do we apply that knowledge? Let's explore this for a bit

You're out on the tennis court practicing your serve (if you can't relate to tennis, please substitute your favorite sporting skill). You perform your first serve (read: rep). The serve was absolutely horrible — in fact, you missed the ball, and quickly surveyed your surroundings to ensure that no one else witnessed the blunder. Hopefully, you mentally rehearse the serve prior to doing another one, searching for clues as to what went wrong. Suddenly, you remember a time-honored maxim that your old high school tennis coach loved to quote: "Keep your eye on the ball!" So on your second serve, you do just that. Amazingly, it works — you manage to hit the ball, and you now realize that your second rep is clearly better than your first.

Despite this revelation, there's no time for self-congratulation: even though you hit the ball, it flew straight into the trees on the other side of the court. So now you replay the serve in your mind, and realize that you'll need to slightly modify the angle of your racquet at the moment it contacts the ball on the next serve in order to get the ball into the far side of the court. So on your third serve, you apply this new concept, and sure enough, the ball goes where you want it to go.

Using the scenario above, can we find a way to quantify the quality (read: intensity) of each serve in the practice session, and rank them in order of effectiveness? The answer is "yes." Although the tennis serve has a significant qualitative component, we can translate your skill level on each serve into a quantitative measurement by having 10 highly skilled tennis coaches watch and assign a score to each serve you perform. Then we'll drop the highest and lowest score, and average the remaining scores.

So, let's say you executed 12 serves, and you receive the following scores (the higher the number, the better the serve):

Next question: How many repetitions would have been optimal? There's no exact answer — we're just looking at the principles involved. And in principle, if we can accept the notion that only perfect practice makes perfect, then we might suggest that you should have stopped after the seventh or eighth repetition, because your skill levels began to decline significantly after that point.

OK, How on Earth Does This Relate to Lifting?!

All of the above scenarios involve motor skill acquisition. And I believe it's very useful to view resistance training for what it is: a motor skill!

Not a believer? Have you ever trained a complete resistance training beginner for an extended period of time? If you have, you'll have noticed a commonly recognized phenomenon: on his first day or training, Tom can barely bench press the empty bar for 4 repetitions.

You scratch your head thinking "How is this possible?" After all, you can bench six times that much weight for 10 reps. But your novice lifter improves by leaps and bounds, adding 1-2 reps per set on every single session. Within 6 weeks, Tom can manage 6 reps with 135 pounds.

Quite an improvement, yet, there's no noticeable change in his body. This is because his rapidly improving strength isn't due to muscle enlargement, but rather, neural processesspecifically, the ability of the motor cortex of the brain to recruit greater numbers of motor units, and particularly, greater numbers of high threshold motor units (For a more detailed look at the neural processes involved in force production, please see Table 1).

Table 1:
The Neural Processes Involved in Force Production

1) Motor unit recruitment (intramuscular coordination): All muscle fibers are one component of what physiologists call "motor units" (MU). An MU is defined as a motor neuron (or nerve cell) and all the muscle fibers it innervates or "recruits." There are several essential facts that athletes should further understand about the functioning of MUs:

• All the fibers of a MU tend to have the same characteristics. When all the fibers are type II, the motor unit is said to be a high threshold or "fast" MU. If the fibers are Type I, it's a low threshold or "slow" MU.

• The all or none principle: When an action potential (the command from the nervous system) is sent from the nerve cell to the muscle fibers, one of two events will occur. If the action potential is strong enough, all the fibers of that motor unit will contract maximally. If the action potential is not strong enough, nothing will happen. In a nutshell, muscle fibers either contract all the way, or not at all. When the body needs to apply more force, it simply recruits more MUs, increases the firing rate of those MUs (see "rate coding" below), or both. Generally, untrained people have limited ability to recruit high threshold MUs because their bodies are unfamiliar with high-tension efforts.

• The size principle: When contracting a muscle to overcome a resistance, the MUs involved are recruited in order of size, small to large. This explains why people can use the muscle to pick up something light (a pencil) or heavy (a dumbbell). As resistance increases, the body recruits more MUs.

2) Intermuscular Coordination: This is the ability of different muscles to cooperate during the performance of a motor task. Muscles can function in several different ways depending on the task at hand.

3) Rate Coding: The nervous system can vary the strength of a muscular contraction not only by varying the number of MUs recruited, but also by varying the firing rate of each MU. This is known as rate coding. The tension that a MU develops in response to a single action potential from the nervous system is called a "twitch." As the stimulus from the nervous system becomes stronger and stronger, the twitches per millisecond become more and more frequent until they begin to overlap, causing greater amounts of tension to be generated by the muscle fiber. The mechanism behind rate coding is very similar to the way in which increased vibrational frequency of a sound increases its pitch.

As an example, a muscle comprising 100 MUs would have 100 graded increments available to it. In addition, each MU can vary its force output over about a tenfold range by varying its firing rate (e.g., from ten to fifty impulses per second). For any set of conditions, the force of contraction is greatest when all MUs have been recruited and all are firing at the optimal rate for force production.

The size of a given muscle may in part determine the relative contribution of rate coding to total muscular force development.(4) In small muscles, most MUs are recruited at a level of force less than 50% of maximal force capacity. Forces that require greater tensions are generated primarily through rate coding. In large proximal muscles (such as the pectorals and lats), the recruitment of additional MUs appears to be the main mechanism for increasing force development up to 80% of absolute strength and even higher. In the force range between 80% and 100% of absolute strength, force is increased almost exclusively by intensification of the MU firing rate.

Muscle Fiber Types and Recruitment

By "high threshold", I'm referring to the fact that the recruitment of fast muscle fiber requires more intramuscular tension than what's required to recruit slow muscle fiber.

Note: The traditional classification scheme for muscle fiber types assigns all fibers as either type IIb, type IIa, and type I. However, I've always felt it was more instructive to simply think of all fibers as belonging to a continuum.

A useful way to envision this spectrum is to remember the volume indicator that was commonly used on older models of stereo equipment — it consisted of a vertical column of small lights, and when you increased the volume, the lights lit up from bottom to top, depending on how much you turned the volume control knob.

In the same way, imagine that we arbitrarily assign all muscle fibers into a vertical column of say, 15 categories, or "lights." When you curl a 5 pound dumbbell, only the bottom 2 lights turn on; i.e., the bottom 2 categories are recruited. But if you curl a 35 pound dumbbell, the bottom 6 categories are recruited, and so on.

The importance of targeting fast muscle fibers (even if you speculate that you're a "slow-twitcher") is that a number of studies show that fast fibers have significantly better capacity to hypertrophy than slow fibers.(5,6) Other studies strongly suggest that intermediate muscle fibers can convert "downward," (i.e., taking on characteristics of slow-twitch muscle fibers) when training involves low to moderate resistances for prolonged durations, or "upward" (taking on characteristics of fast twitch muscle fibers) when training involves high tension efforts.(7)

If your goal is to get bigger, you need to gain access to the heavy hitters — the high threshold, fast motor units, because you can't train them until your brain learns to recruit them in the first place.

The Bottom Line:
What Causes Hypertrophy Anyway,
And What's The Best way to Achieve it?

There have been a number of possible mechanisms proposed for the hypertrophy process. These include the muscle hypoxia hypothesis (a deficiency of blood, and therefore oxygen to the muscles stimulates protein synthesis), the blood circulation hypothesis (blood circulation to working muscle provides the stimulus for growth), and the ATP debt hypothesis (ATP concentrations decline during training, which supposedly stimulates muscle growth).(8)

However, the theory which seems to hold the most promise suggests that energy distribution (or lack thereof) creates the stimulus for muscular hypertrophy. The idea is that during rest, muscular energy is distributed between mechanical work and protein synthesis (protein synthesis is a 24-hour a day process, however, it is greatly accelerated by heavy training).

So for example, when you're standing in line at the supermarket, a small amount of energy is used to keep you standing upright, and the rest is diverted toward protein synthesis. However, during a hard training session, a large proportion of available energy is expended for the mechanical work involved in lifting, which leaves relatively little for protein synthesis. It's proposed that this energy deficit is the trigger for hypertrophy of the working muscles.(9)

This hypothesis corresponds well to Selye's general adaptation syndrome (GAS) theory, where, upon being subjected to a stressor, the organism first experiences an alarm stage (here, the energy deficit), and then later, a supercompensation stage (hypertrophy).(10)

If the above hypothesis is correct, we can then say that hypertrophy is a function of how much mechanical work is performed per unit of time. For example, imagine that today's back and triceps workout resulted in a volume of 23,250 pounds performed in a 50-minute time frame. If during the next back and triceps workout you manage to lift 23,320 pounds in 55 minutes or less, you'd have provided the necessary stimulus for muscle growth. Do sets and reps matter? I think they do, but not in the way that you might think. Table 2 illustrates two workouts that both result in the same training volume:

Table 2: Comparison of Volume Versus Intensity-based Approaches
Workout One Workout Two
A-1: Chins: 245 (3x10) A-1: A-1: Chins: 245 (10x3)
A-2: Close-grip bench: 225 (3x10) A-2: A-2: Close-grip bench: 225 (10x3)
B-1: Bent Rows: 205 (3x10) B-1: Bent Rows: 205 (6x5)
B-2: French Press: 100 (3x10) B-2: French Press: 100 (6x5)
Volume: 23,250 pounds  Volume: 23,250 pounds 
Duration: 55 minutes  Duration: 55 minutes 

From this information, you might conclude that the way you arrange your sets and reps will have no bearing on the outcome. After all, the training duration, volume, and even density are identical in both cases. Even the intensity is the same, since the same weights are used in both cases. But wait: is intramuscular tension (the key to accessing, and therefore, training, fast muscle fibers) simply a matter of how much weight you use?

If you answer "Yes," let me propose an experiment: I'd like to place a 25 pound plate gently on top of your foot, and determine your reaction to the load. Then, I'd like you to drop the same plate from 6 feet in the air on your foot. Sounds Okay? The weight is the same in both cases, right? So the outcome should be identical! Of course the outcome will NOT be identical, because the plate which is dropped from a height picks up acceleration as it falls.

In much the same way, accelerating a weightload results in greater tensions on the target muscles than moving the same weight slowly. Further, many sets of low reps facilitate acceleration more efficiently than few sets of many reps (which is the norm in gyms and weightrooms today).

Consider your last workout, where you did an all-out set of 10 reps with 225 on the front squat. How much tension (measured as pounds of pressure on the bar) did you exert on the bar on rep number 10? If you barely managed 10 reps, and you would have missed the 11th rep, would you accept that you exerted just slightly more than 225 pounds of force on the bar — perhaps 226 pounds? If so, would you also accept that you managed slightly more force on rep number 9, and even more on rep 8, etc., since fatigue accumulates from rep to rep? Here is a hypothetical representation of your force output during that set of 10:

Rep One: 244 pounds
Rep Two: 242 pounds
Rep Three: 240 pounds
Rep Four: 238 pounds
Rep Five: 236 pounds
Rep Six: 234 pounds
Rep Seven: 232 pounds
Rep Eight: 230 pounds
Rep Nine: 228 pounds
Rep Ten: 226 pounds

Now bear in mind, the exact numbers may not be completely accurate, but the trend is. The idea is simply that accumulating fatigue limits force output from rep to rep. If we add up these numbers and divide by 10, we get 237 pounds — this represent the average force per rep.

Now let's invert the sets and reps and see what we get. Instead of using 225 pounds for 1x10, we'll use the same weight for 2x5. Now, the average force per rep is 240 pounds, because by keeping fatigue to a minimum, we can accelerate the bar more effectively. Yet the total volume and density are unchanged. Given the following two alternatives, which would you choose?:

First Scenario: 225x10
Load: 225 pounds
Volume 2250 pounds
Average Force per Rep: 237 pounds

Second Scenario: 225 (2x5)
Load: 225 pounds
Volume 2250 pounds
Average Force per Rep: 240 pounds

(Please see Table 3 for a more detailed representation of how to employ these principles into an actual workout.)

Table 3:
Sample Chest & Back Workout Employing the Principles Discussed

"A" Series: Dips and Chins

1) Using the concepts presented in "Warming-Up to a Great Workout — a five-stage event!," warm yourself up by doing several easy sets of both exercises, alternating between dips and chins.

2) Your working weights should approximate 70% of 1RM for each exercise (this may necessitate using additional load on the dips via the use of a weighted belt). Don't get too hung up on 70% — we're just selecting an intensity to illustrate the principles involved.

3) Determine an appropriate lifting speed and a way to monitor it from set to set. The speed will depend on the resistance selected and the repetition scheme. For this example, we'll select 2 seconds OR LESS per rep. Either have a training partner count your reps, or use an electronic metronome to monitor your rep speed.

4) Determine a rest interval. Again, this can depend on the resistance selected and the repetition scheme, but for this example, we'll use 60 seconds OR LESS between sets. Use a stopwatch or a partner to monitor this parameter as you progress through your workout

5) Okay it's "Go Time!" (From Jerry Seinfeld's unsolicited and aged personal trainer, played by Lloyd Bridges): Perform your first set of dips, making sure to stay ahead of the 2 seconds per rep speed (this will require maximal acceleration, but I'm not suggesting that you sacrifice control in the process — stay tight and maintain superb control at all times). Rest one minute or less, and perform your first set of chins. Rest one minute or less, back to dips.

6) Continue alternating between dips and chins until you either slow down to the point where a rep takes more than 2 seconds to perform, and/or where you miss a rep or cannot beat the time limit between sets. This "failure" should occur somewhere between 6 and 12 sets any more, and the load is too light; any less, and the load is too heavy.


7) Depending on how many sets you managed, the "A" series should have taken you between 15 and 30 minutes. Now on to "B" series

8) Your next 2 exercises: Incline dumbbell presses and Hammer rows. Perform 1-2 warm-up sets for each exercise in order to rehearse the motor pattern and to determine an appropriate training weight for each exercise. For the "B" series, I'll often select a slightly reduced load and slightly increased reps per set, for the purposes of local muscle endurance and growth hormone secretion. For this example, we'll use 70% of 1RM and sets of 6.

9) Perform the "B: series in the same manner as the first two exercises, using the same speed parameters.

Progression: Perform this workout 3-6 times (once every 4-7 days), seeking to increase your training volume by 10% each session. On the first workout, leave enough in reserve that you can increase volume by 10% for at least 3 successive workouts (accomplished by adding 1-2 additional sets per session). After 3-6 workouts, create a new exercise menu and start over.

Additional Modifications: Depending on goals, need for variation, and so forth, a variety of loads (from 55 to 85%) and repetition schemes (between 2 and 8 per set) are possible. The key concept is to base the training load on performance quality rather than an arbitrarily selected number of reps or sets.

I should mention that I'm taking a few liberties here to get my point across (for example, when you perform 2x5, you'll have slightly reduced force output on the second set due to fatigue), but I believe the concept remains valid: breaking up your sets into smaller chunks in order to reduce fatigue allows higher force output, and accordingly, more stimulus to high threshold motor units.

I offer this approach not as an exclusive training method (for example, high rep sets have their place in the development of local muscle endurance and in the production of growth hormone, which has been postulated to assist in fat loss), but as a method that's been successful for my own clients, and indeed, thousands of athletes involved in Olympic weightlifting, powerlifting, and other power events.

I urge you to explore the concept. After all, the methods which will bring you the most success in your future training are likely to be the methods you haven't used yet — is this one of those methods?

Conclusions and Recommendations

1) Base the number of reps per set in such a way that fatigue is minimized as much as is practical. I say "practical," because taken to its extreme, this would mean always doing one rep per set. But obviously, in many instances (such as heavy dumbbell presses for example), if you try to limit reps to 1 or 2 per set, you'll end up doing more work setting up for the set than actually performing the set itself.

2) Base the number of sets per exercise on a) how many exercises are on the menu (the more exercises planned for a workout, the less sets you'll be able to perform for each exercise — this argues for multiple daily sessions), and b) the quality of your performance from set to set. This is best measured by monitoring rep speed, usually assessed subjectively, or objectively, using a stopwatch or an accelerometer.

3) There are two ways to increase tension on muscles: lift heavy weights slowly (you'll have no choice in the matter in that particular scenario), or moderate weights acceleratively.(11) The second option is rarely used by bodybuilders, but it offers unique advantages, including improved speed strength and never needing a spotter.

4) Regardless of how you organize your sets and reps, seek continuous, gradual increases in work output from session to session. Hypertrophy is a function of how much mechanical work you do in each session, regardless of what your set/rep scheme is. If you gradually do more and more work with each new session, you're providing the necessary stimulus for muscle growth.

5) Fatigue is not the goal of training, but a sometimes unavoidable result of seeking continued progress from session to session. You'll make more progress avoiding it than seeking it.


1) Zatsiorsky, V.M., Science and Practice of Strength Training (1995) Champaign, Human Kinetics, p.p. 96.

2) Komi, P.V.(Ed.), Strength and Power in Sport (1992) London, Blackwell Scientific Publications, p.p. 378.

3) Fleck, S.J., & Kraemer, W.J., Designing Resistance Training Programs (1987) Champaign, Human Kinetics, p.p. 217.

4) Zatsiorsky, V.M., Science and Practice of Strength Training (1995) Champaign, Human Kinetics, p.p. 78.

5) Komi, P.V.(Ed.), Strength and Power in Sport (1992) London, Blackwell Scientific Publications, p.p. 231.

6) Tesch, P.A., (1998) Strength Training and Muscle Hypertrophy. International Conference on Weightlifting and Strength Training Conference Book, p.p.18.

7) Andersen, J.L., Schjerling, P, & Saltin, B., (2000). Muscle, Genes, and Athletic Performance. Scientific American, Vol. 283, Number 3. p.p. 52.

8) Zatsiorsky, V.M., Science and Practice of Strength Training (1995) Champaign, Human Kinetics, p.p. 64.

9) Siff, M.C., & Verkhoshansky, Y.V., Supertraining: Special Strength Training for Sporting Excellence (1993) Johannesburg, University of Witwatersrand, p.p. 60-61 )

10) Siff, M.C., & Verkhoshansky, Y.V., Supertraining: Special Strength Training for Sporting Excellence (1993) Johannesburg, University of Witwatersrand, p.p. 81-82 )

11) Hartmann, J., & Tunnemann, H., Fitness and Strength Training for All Sports (1995) Toronto, Sports Books Publishers, p.p. 27.