Intensity is not a thing in and of itself, but a measure of something. When we’re talking about “intensity” in the context of exercise we’re really talking about the measure of things like relative effort, relative load, heart rate elevation, etc., and rather than just saying “intensity” we really ought to be more specific, saying “intensity of effort” or “intensity of load”.
In High Intensity Training (HIT) “intensity” is usually used to mean intensity of effort, or how hard you are working relative to how hard you are capable of working at the moment, and that is what I usually mean when I say or write “intensity”. For example, if you perform an exercise using a resistance that is eighty percent of your one repetition maximum your intensity of effort at the beginning of the exercise will be eighty percent, but as you fatigue the resistance becomes an increasing percentage of your decreasing strength, until your strength has been reduced to the point where it equals the resistance and your intensity of effort is one hundred percent.
Intensity of effort is often criticized as being imprecise, since it is not accurately measurable during exercise except at the point of momentary muscular failure, but since relative effort appears to be the most important factor for stimulating improvements in muscular strength and size it is the most relevant, and most useful (1).
Most of the rest of the fitness industry uses “intensity” to mean intensity of load, the percentage of your one repetition maximum used for an exercise. There are several problems with this, however, including problems with accurately testing one repetition maximums. Although all types of exercise intensity are meant as a measure of how hard you are working during an exercise there are numerous factors which affect exercise difficulty and it is possible to work easier or harder with the same percentage of your one rep max by changing one or more of these factors, such as whether the exercise is performed to the point of momentary muscular failure. Also, as long as an exercise is performed to the point of momentary muscular failure (maximum intensity of effort) most people do not perceive any difference in difficulty using different relative loads (2) or even perceive greater difficulty with lighter loads (3).
Regardless of the relative load used, if you are not training to momentary muscular failure you are not working as intensely as possible.
It could be argued that despite these problems intensity of load can be used to estimate the average intensity of effort of an exercise. Assuming a relatively constant rate of fatigue, and assuming the exercise is stopped immediately after momentary muscular failure occurs, the average intensity of effort during the exercise would be halfway between the percentage of one-rep max used and one hundred percent.
This is a lot of assumptions, however, and several things can throw this off. Fatigue during exercise is the result of numerous factors, which can vary depending on protocol, and probably does not occur at a constant rate. When training people on Ken Hutchins’ iMachines with real-time force measurement and feedback (displayed as a line on a graph) it was not unusual for the rate of fatigue to increase towards the end of the exercise. Also, if you continue to contract isometrically for several seconds after achieving concentric momentary muscular failure you increase the relative time spent working at maximum intensity of effort, increasing the average (and I suspect this may be a factor in the increased effectiveness of several advanced high intensity training methods as well as the reason they seem to make greater demands on recovery).
In addition to all these problems there appears to be little difference on average in muscular strength and size increases with different relative loads and repetition ranges as long as exercises are performed to momentary muscular failure, which throws a huge wrench into the claims that load equals intensity, or that ultra-heavy ultra-brief sets are the way to go (1, 4).
The biggest problem with using load to define intensity is thinking that mechanical definitions of work and power or force or load over time accurately describe what is happening in the muscles during exercise.
This is easy to disprove, and Arthur Jones wrote about this decades ago in The Metabolic Cost of Negative Work,
A great deal of confusion exists on this point because of attempts to compare metabolic work with mechanical work, and, secondly, because of a failure to consider several related factors.
By definition, work require movement… no movement means no work; and while this is undoubtedly true in regard to mechanical work, it certain is not true in relation to metabolic work.
Muscles produce force, and it is easily possible for a muscle to produce a high level of force without producing movement; logically, it appears that the metabolic cost of muscular force production would be related to the level of force produced and the time that the force is maintained… rather than the amount of mechanical work performed.
If, for example, a 100-pound barbell is held motionless at the halfway position of a curling exercise, then the muscles will be required to produce a certain level of force to prevent the downward movement of the barbell. Providing that force will certainly entail metabolic cost… yet no work is involved.
Slowly, curling a 100-pound barbell also requires a greater metabolic cost than curling the same barbell at a more rapid pace; even though the amount of mechanical work involved is exactly the same in both cases.
Many other examples could be given to illustrate the same point, but it should now be obvious that attempts to relate metabolic cost to mechanical work are doomed to failure.. there is no meaningful relationship. We must have another standard for comparison.
The only meaningful standard, I can think of is force/time… the amount of force produced by the muscles multiplied by the time that the force is maintained.
Force multiplied by time is a much better measure of exercise difficulty than mechanical work, but still flawed since it is possible to have the exact same force output and duration and even the same mechanical work and still have different levels of difficulty if the repetition cadence, range of motion, or other factors are varied. For example, if you perform ten repetitions of an exercise each taking eight seconds with the same load, over the same range of motion your mechanical work, power, and force multiplied by time would be the same, but the difficulty will be different if you lift in two seconds and lower in six, lift in four and lower in four, or lift in six and lower in two since your muscles are stronger and fatigue more slowly when lowering than lifting and since the force your muscles are capable of producing concentrically varies with contraction velocity.
Differences in strength and resistance curves complicate this further. Although it is possible to get a resistance curve that very closely approximates your strength curve using machines with well designed cams with adjustable timing like Ken Hutchins “Alligator” machines or lifting weights with correct body positioning and movement during some free weight exercises, it is impossible to match your strength perfectly over the full range of an exercise using weight-based machines (the accommodating resistance of motorized machines is a different story, although it has it’s problems, too). Because of this, there will be portions of the range of motion that are harder, and portions that are easier, and depending on the amount of time you spend in each during both the positive and negative the exercise may be more or less difficult. For example, during most compound pushing movements the target muscles are working against a much larger lever at the start point than the end point. Performing the lower turnaround very slowly, or even holding briefly at the start point makes an exercise more difficult, while performing the upper turnaround any more slowly than necessary or holding there makes an exercise easier.
For example, imagine if you were to perform two sets of barbell squats with the same weight, same number of repetitions, and same cadence, one in the top half of your range of motion, one in the bottom. The force multiplied by time and mechanical work would be the same, but the set performed in only the bottom half of your range of motion would be much harder.
This is because the force you produce when contracting against the resistance is the product of both muscular force and leverage. As your muscles contract and cause your bones to move their angle of pull and leverage changes, so the same amount of force applied to a machine or weight through those bony levers can require more or less muscular force input depending on the part of the range of motion you are in. This is the reason some exercises are harder in some portions of the range of motion and easier in others, the reason you are capable of holding much more weight at or near lockout on pushing exercises, and the reason that load multiplied by time is not a valid measure of exercise difficulty.
It is the reason leg pressing something as heavy as a small car over a range of motion of only a few inches near lockout is not more intense, does not require more effort than performing a full-range leg press and avoiding lockout with much less weight. This is not impressive to anyone with a basic understanding of physics. It’s just stupid, and a good way to wreck your knees.
This is extremely important, so I’m going to say it again. The resistance your muscles contract against during an exercise is the result of both the load and the lever you are working against. Two thousand pounds lifted with a two inch lever requires the same muscular force input as only two hundred pounds lifted with a twenty inch lever:
2000 pounds x 2 inches = 4000 inch-pounds of torque
200 pounds x 20 inches = 4000 inch-pounds of torque
Since the muscular force input required is the same it makes little difference to your muscles which you do, but it makes a huge difference to your joints in many exercises, which is why it is just as effective but much safer to perform an exercise over a greater range of motion against a larger average lever using less weight.
Also consider your muscles are capable of producing different amounts of force at different lengths (length-tension curve), with their force output being lowest when fully stretched and fully contracted and highest through the middle of their range of motion. As you approach lockout most of the muscles being worked will be closer to their fully contracted length and capable of producing less force than in the middle of your range of motion, so the muscular force input may actually be slightly lower when doing very short-range partials in the so-called “strong range” of an exercise.
I’m not saying load isn’t important, it appears you need at least a heavy enough load to be able to achieve momentary muscular failure within at most a few minutes. However, working as hard as possible relative to your momentary ability has a much bigger impact on your muscular strength and size gains than the load you use, and it is safer and easier on your joints in the long run to use more moderate loads.
1. James Fisher, James Steele, Stewart Bruce-Low, Dave Smith. Evidence-Based Resistance Training Recommendations. Med Sport 15 (3): 147-162, 2011 DOI: 10.2478/v10036-011-0025-x
2. Vitor L. Silva, Arthur P. Azevedo, Joctan P. Cordeiro, Michael J. Duncan, Jason M. Cholewa; Mário A. Siqueira-Filho, Nelo E. Zanchi, Lucas Guimarães-Ferreira. Effects of exercise intensity on perceived exertion during multiple sets of bench press to volitional failure. J of Trainology August 2014; Vol. 3, No. 2: Pages 41-xx
3. Shimano T, Kraemer WJ, Spiering BA, et al. Relationship between the number of repetitions and selected percentages of one repetition maximum in free weight exercises in trained and untrained men. J Strength Cond Res 2006; 20: 819-23.
4. N.A. Burd, C.J. Mitchell, T.A. Churward-Venne, and S.M. Phillips. Bigger weights may not beget bigger muscles: evidence from acute muscle protein synthetic responses after resistance exercise. Appl. Physiol. Nutr. Metab. 37(3): 551-554, 2012.