“Overhead Triceps Extensions are (NOT) Better Than Neutral Arms Triceps Extensions”

Review of Questionable Conclusions
of Scientific Studies (a series)

Part 1

A study was recently done which attempted to compare the resultant muscle growth of two different triceps exercises. The two exercises that were compared were the “Overhead Cable Triceps Extension”…and the standard “Cable Triceps Pushdown”.

The 21 participants did one of the exercises with their left arm, and the other exercise with their right arm — for a 12 week period of time. Then the triceps growth of each arm was measured. The conclusion of the researchers was:

“Triceps brachii hypertrophy (growth) is substantially greater [using overhead elbow extension versus neutral arm elbow extension]”

This study — and the conclusion of the researchers — was reported in “PubMed.gov” (National Library of Medicine), among other journals.

This conclusion was then widely reported in a variety of fitness journals, touted by numerous “evidence-based” personal trainers, and announced as fact by professors of exercise physiology…throughout the world.

“Insider.com” published a report on this study, with the following headline:

“Stop doing triceps pushdowns if you want bigger arms…overhead extensions may lead to significantly more muscle growth, according to a study published July 12, 2022 in the European Journal of Sport Science”.

However, I believe there is a flaw in this conclusion, and it stems from the assumption that the only difference between the two exercises was the position of the arm, relative to the shoulder joint.

Actually, that was not the only difference between the two exercises. It was not even the primary difference, although it might appear that way to most people.

The researchers attributed the “significantly better triceps hypertrophy” to the overhead arm position. However, the “neutral arm” exercise they selected for the study, has a very compromised resistance curve. The “overhead” exercise they selected has a much better resistance curve (i.e., the direction of the resistance is more perpendicular to the forearm, at the beginning of the range of motion). So, the exercise that was more productive was the “overhead” exercise, but that was more likely to due to the fact that the “overhead” exercise had a better resistance curve than the “neutral arm” exercise.

The resistance curve of each exercise was not even addressed, so it appears to have not been considered.

In order to accurately assess the value of a “resistance exercise”, a researcher must have a thorough understanding of the mechanics that occur outside the body (i.e., the physics of each exercise), and ensure that both exercises being compared have the exact same resistance curve — thus eliminating all possible confounding variables.


To see how this oversight occurred —let’s begin by reviewing the triceps anatomy.

The triceps muscle is compromised of 3 parts. These are known as the “long head” (Longus), the “lateral head” and the “medial head” (Medius) — shown below.

The “long head” originates on the outer edge of the shoulder blade (the “scapula”…indicate by the black arrow). The other two parts — the “lateral head” and the “medial head” — originate on the upper arm bone (the “humerus”)…indicated by the green arrow. Then, all three parts converge on one single tendon just above the elbow joint. This tendon then crosses the elbow joint, so — when the triceps contracts — it “extends” (straightens) the elbow.

Since the origin of the “long head” is situated lower than the shoulder joint, positioning the arm overhead causes the distance between the origin and insertion of the “long head” to be farther away from each other, as compared with having the arm down at one’s side. This creates a situation whereby — potentially — the “long head” could be stretched more during an overhead riceps exercise. However, this requires that the elbows be fully bent, while the arms are overhead.

It is well known among students of exercise physiology that allowing a muscle to stretch more during a resistance exercise increases the potential benefit of the exercise. Thus, it’s conceivable that performing Overhead Triceps extensions COULD cause greater development of the “long head” of the triceps — but only the “long head”not the other 2 parts.

The researchers in this study noted that “ALL THREE PARTS of the triceps grew more with the Overhead triceps exercise, as compared with the arms neutral (arms down) triceps exercise.”

This mystified the researchers because they know that the “lateral” and the “medial heads” of the triceps do not change their length, nor their function, when the position of the arm (humerus / shoulder joint) changes. Therefore, there is no logical reason why those two parts of the triceps would benefit more from an overhead triceps exercise, as compared with an “arms down” triceps exercise.

Yet, apparently none of the researchers stated, “the fact that the lateral and medial triceps heads improved more with the overhead exercise suggests that some OTHER reason probably caused the better triceps growth….maybe the better outcome was NOT due to the Overhead arm position.

The two uni-articulate triceps heads (the lateral and medial heads) cannot possibly function differently during an overhead triceps exercise, because their length is not altered — in fact, nothing about their function is altered — by a change in the shoulder angle. So this was a clue that something ELSE caused the triceps to grow better with the overhead triceps exercise.

In fact, the direction of resistance typically applied during an overhead triceps exercise is MORE PERPENDICULAR to the forearm, then that which is typically applied during a standard Cable Pushdown. This results in a more productive resistance curve with the overhead triceps exercise, then the compromised resistance curve of the standard Cable Pushdown. This is the only plausible reason why the uni-articulate triceps heads developed less well with the Cable Pushdowns.

In order for the conclusion of a study to be accurate, the researchers must be aware of how and when OUTSIDE forces (ground reaction force, centrifugal force, centripetal force, friction force, composite force, momentum, varying DIRECTIONS of resistance and variations in “moment arms”) are occurring. It seems this was overlooked in this study.


Students of exercise physiology know that muscles have more strength potential when they are elongated, and they have less strength potential when they are shorted (contracted). This is known as the “strength curve” of muscles.

The triceps has more strength potential when the elbow is fully bent (i.e., because the triceps is elongated), and less strength potential with the elbow is straight (i.e., because the triceps is shortened).

Since muscles are stronger during the early part of their range of motion and weaker during the latter part of their range of motion, the BEST exercises for optimizing muscle development are the ones that provide MORE resistance when the muscle is elongated, and LESS resistance when the muscle is shortened. Selecting an exercise that provides a more perpendicular resistance at the beginning of the range of motion, optimizes the benefit to the muscle.

All levers (which includes human limbs…arms, legs, etc.) follow universal physics principles that influence how much “load” is placed on a particular structure. Longer levers magnify load more than shorter levers, and levers that are more perpendicular with resistance (e.g., gravity) magnify load more than levers that are more parallel with resistance.

In the two LEFT images (upper and lower), the forearm is vertical (parallel with gravity). At both of those points in the range of motion, the triceps experiences ZERO resistance, because the forearm is vertical (in the “neutral” position).

In the two RIGHT images (upper and lower), the forearm is horizontal (perpendicular with gravity). At both of those points in the range of motion, the triceps experiences the maximum percentage of available resistance (which is a factor of the amount being lifted).

In the TOP two images (“Lying Dumbbell Triceps Extensions”), the amount of resistance being experienced by the triceps goes from MOST (right image) when the elbows are bent, and LEAST (left image) when the elbows are straight.

Conversely, in the bottom two images (“Dumbbell Kickbacks”), the amount of resistance experienced by the triceps goes from LEAST (left image) when the elbow is bent, and MOST (right image) when the elbow is straight.

These sequential changes in the percentage of resistance an exercise delivers to a working muscle is called the “resistance curve”. Every exercise has a different resistance curve, or “biomechanical profile”.

When an exercise has a resistance curve that matches the strength curve of your target muscle, the exercise is GOOD. It provides the muscle with more resistance when the muscle has more strength capacity (in the early part of its range of motion), and less resistance when the muscle has less strength capacity (at the end of its range of motion).

When an exercise has a resistance curve that does NOT match the strength curve of your target muscle, the exercise is compromised to varying degrees. A compromised resistance curve typically provides TOO LITTLE (or no) resistance when the muscle has more strength capacity, and TOO MUCH resistance when the muscle has less strength capacity.

Being able to recognize when an exercise provides a good resistance curve requires some familiarity with physics principles (mechanics), and this awareness is significantly different than issues related to physiology.


When evaluating the resistance curve of an exercise that uses “free weight” (dumbbells, barbells, etc.), the direction of resistance is always “straight down” / vertical. So, one would then simply compare that vertical direction of resistance to the varying limb angles during the exercise.

However, when evaluating an exercise that uses “cable resistance”, the direction of resistance is NOT necessarily straight down / vertical. Instead, it is the direction of the cable. For example, when doing a Lat Pulldown, the direction of resistance is “upward” — toward the pulley above you. In cases such as this, you would simply compare the angle of the cable to the angle of the limb being moved by your target muscle, to evaluate the resistance curve of the exercise.

In the image below, I’ve drawn a red line through my forearm (i.e., the limb that is being moved by my triceps), and green line though the cable (i.e., the direction of resistance being provided by THIS particular exercise).

As you can see, the red line and the green line are not perpendicular to each other. They are closer to parallel with each other. This means that — in that starting position (with the elbow fully bent, and the triceps most elongated), the percentage of resistance being delivered to the triceps is very small…perhaps only 15% to 20% of what it would be if those two lines were perpendicular to each other.

Therefore, this exercise (standard Cable Pushdowns) has a compromised resistance curve because it is not “early phase loaded”. Yet, this exercise (with its very compromised resistance curve) was one of the two exercises used in this study.

It is not the “neutral arm” position of a standard Cable Pushdown that compromises this exercise. It is the fact that the direction of resistance is not perpendicular with the forearm, at the beginning of the range of motion. It can be improved, however. If you were to change the direction of resistance (i.e., change the position from where the resistance comes), you can make it “early phase loaded”. This would then provide an optimally productive resistance curve, making this an excellent triceps exercise. In the images below, I show how you would do this, even though the arm motion stays the same. Make the resistance come from slightly behind you, rather than from in front of you.

Below is another way of doing this, with the added advantage of having “back support” (and a seat) for more stability.

The other exercise that was used in the study is depicted below — a (one arm) Overhead Cable Triceps Extension.

Notice how perpendicular the cable is with the forearm, at the beginning of the range of motion (above, and below-left). Compare this with how NOT perpendicular the cable is with the forearm, at the beginning of the range of motion, of the Cable Pushdown (below-right).

It seems the researchers thought they were comparing two triceps exercises which had ONLY one difference— the position of the arm (either “arm up” or “arm down”). But, they were comparing two exercises that ALSO have two drastically different resistance curves.


Acknowledging when the direction of resistance is perpendicular to a limb — versus parallel to a limb — is critically important, because only then can the “resistance curve” of the exercise be identified. Only then can it be determined whether it matches the strength curve of the target muscle. Only then can a comparison between two exercises be accurate.

The conclusion reached by these researchers — that “Overhead Triceps Exercise are more productive than Neutral Arm Triceps Exercises” — ignores the significant difference in resistance curves, between the two exercises.

I believe it’s entirely unreasonable to suggest that overhead triceps exercises are “significantly better” for triceps growth, compared to neutral arm triceps exercises, which was the conclusion of this study.

The lateral and medial heads of the triceps are not the least bit altered in their function, based on the position of the arm / shoulder. Therefore, it should be clear that these two parts of the triceps CAN be optimally developed with any triceps exercise (any arm / shoulder position) provided an optimal resistance curve is used (early phase loaded / perpendicular resistance at the beginning of the range of motion).

In regard to the long head of the triceps, the key question — I believe — is: “Can the long head of the triceps be adequately lengthened with a neutral arm triceps exercise, provided maximum elbow bend is used with each repetition?” In other words, “Does the long head absolutely REQUIRE the maximum stretch that could be achieved with an overhead triceps exercise AND maximum elbow bend with each repetition?

There are two ways of answering this question.

The first way is “Try it and see”, which is a very reasonable strategy. Why would you do an uncomfortable exercise (i.e., an overhead triceps exercise) if a neutral arm triceps exercise (with an optimal resistance curve and maximum elbow bend) works just as well? I am thoroughly convinced it does work “just as well”, and I have proven it with the excellent results I’ve achieved. The thought that “I cannot achieve optimal triceps long head development UNLESS I perform an overhead triceps exercise” seems absolutely ridiculous to me.

The second way of sensing whether it’s “necessary” to do an overhead triceps exercise in order to achieve optimum development of the triceps long head, is by comparing it to the quadriceps. The triceps of the arm, and the quadriceps of the leg, are essentially the same type of mechanism.

The long head of the triceps and the rectus femoris (of the quadriceps) are the only bi-articulate parts of the triceps and quadriceps (respectively) . The other parts of each muscle only cross the one joint. The suggestion that training the triceps with maximum shoulder extension (overhead position) is “essential”, is like suggesting that maximum hip extension is “essential” when working the quadriceps (i.e., to fully stretch the rectus femoris). In both cases — I believe — it’s absurd.

The two triceps exercises I most recommend (the only ones I do…one or the other, but not both, per workout) are “Modified Cable Pushdowns” (using two cables, coming from behind you), and “Decline Dumbbell Triceps Extensions”…both shown below. Both have an optimal resistance curve.

You might want to read “The 16 Factors That Determine Exercise Value”, at my website: dougbrignole.com …in order to see all the factors that can be used to determine the value of an exercise.

I have deliberately avoided overhead triceps exercise for the past 20 years, and — as you can see below — my triceps development has not been the least bit compromised (not even the “long head”). In addition, the numerous clients I have coached, as well as those who have followed the recommendations in my book, have also reported exceptional triceps development doing only one of the two options shown above.


The conclusion reached by the researchers who conducted this study…

“Triceps brachii hypertrophy (growth) is substantially greater [using overhead elbow extension versus neutral arm elbow extension]”

…is misguided and misleading.

The outcome of this study would have been very different if the “arms down” triceps exercise used in the study had had an optimally productive resistance curve (i.e., a perpendicular direction of resistance at the “early phase” of the range of motion), instead of the compromised resistance curve which is typical of a standard Cable Pushdown.

This is not meant to suggest that Overhead Triceps Exercises are “not productive”. They can be productive, but are not superior to “neutral arm” triceps exercises when they are early phase loaded.

It is possible that the triceps’ long head would benefit more from an “arms up” triceps exercise, provided the elbows are fully bent at each repetition. But this does not mean that “overhead” triceps exercises are “essential” for optimal development of the triceps’ long head.

The long head of the triceps can achieve very adequate stimulation with “arms down” triceps exercises, provided the elbows are maximally bent when performing “neutral arms” triceps exercises, and the exercise is early phase loaded.


Doug Brignole is a veteran competitive bodybuilder with over 43 years of experience. He is a former Mr. America and twice Mr. Universe winner, and a respected author. He is still competing successfully at the age of 62 — without injury.

Levels of Complexity in the Biomechanical Analysis of Secondary Limb Participation During Resistance Exercise

The fundamental premise of resistance exercise is to load a target muscle by applying a resistance source to a limb that is operated by that target muscle, and then contracting that muscle against the opposing resistance.

The simplest example of this is grasping a dumbbell in one’s hand, and performing a “Dumbbell Curl” (aka a “Biceps Curl”). The resistance cannot be held directly by the biceps brachii, so it must be held in the hand which is located at the distal end of the forearm, on the other side of the wrist joint. Thus, the forearm is the “primary lever” of the biceps, because the biceps connects directly to the Ulna of the forearm. The hand (which has some degree of length) is the “secondary lever”, because it’s connected to the forearm (at the wrist), but is not directly connected to the biceps. The hand essentially lengthens the forearm. Thus, it extends the “moment arm” (increases the magnification to the biceps), but only to a small degree.

There are occasions, however, when the secondary lever is longer than the hand, and is therefore much more influential to the primary lever. In the simplest of these cases, the secondary lever is used mainly as a “connector”—a means of holding / applying the resistance, because the primary limb cannot hold it directly. In this scenario, the secondary limb (which is holding the weight) is mostly inconsequential (maintained at a neutral angle). However, it could be more consequential if it is allowed to tilt away from its neutral angle.

Further, there are occasions when the secondary lever (limb) is pressing against a “ground reaction force” (an immovable surface), rather than simply holding a free weight. In such cases, the secondary lever cannot be held in a neutral position, and muscles (one or more) attached to that limb are therefore loaded to the degree of its moment arm — its degree of tilt away from the neutral position.

Adding more complexity to the mechanical analysis, there are occasions when the secondary limb encounters a lateral (sideways, forward or backward) “ground reaction force”—rather than a simple vertical ground reaction force, and this alters the direction of resistance from that which might appear to be obvious.

The analysis is further complicated with the addition of a “brace” from which the person must apply their effort by default—and that is also combined with a lateral ground reaction force.

Analysis is most complicated when an exercise combines all of the above, plus a non-linear resistance source that has as an “arc” trajectory and that trajectory is opposite the limb’s trajectory.

Here’s a more detailed breakdown of the various tiers of complexity.

* * * * * * * * * * * * * * * * * * * * * * *

Tier 1:     The Neutral Secondary Lever

Examples: Supine Dumbbell Press / One Arm Dumbbell Row

In the image above, the humerus is the primary lever of the pectorals because the pectorals connects directly to the humerus. The forearm is the secondary lever of the pectorals because it does not connect directly to the pectorals, but is connected to the humerus, and is holding the weight.

In this particular application (i.e., the forearm being held vertically), the forearm (i.e., the secondary lever) is only a connector applying the weight to the distal end of the humerus, because the humerus has no other way of “holding” the weight. It requires the hand, which is attached at the distal end of the forearm.

In the top two examples below, the forearm is in the neutral position; it is neither extending nor shortening the “moment arm” of the humerus (as the pectoral lever). It is also not loading the biceps, nor the triceps, nor the external shoulder rotators, nor the internal shoulder rotators. It is acting merely as a “support beam”—holding the weight directly above the end of the distal end of the humerus.

In the lower two images above, I’ve modified the image by eliminating the forearm, and attaching the hand directly to the end of the humerus. I’ve lined up the upper and lower images, and added lines showing the “moment arm” of each (the upper and lower images).

As you can see, the moment arm is essentially the same—with the forearm and without the forearm. The distance between the shoulder joint (i.e., the pivot used by the pectoral muscles) and where the load is applied, is the mostly same. Thus, the magnification is the same—and the load to the pectorals is the same—with or without the forearm.

In the lower image above, I have removed the hands from the ends of the humerus, and attached a chain instead. Now, the weight is hanging below the “elbow” (the distal end of the humerus), instead of being supported above the distal end of the humerus by the forearm. Still, the “moment arm” (the distance between the shoulder joint and the application of the resistance) is the same. Thus, the magnification is the same, and the load to the pectorals would be the same—whether the weight is held above or below the humerus. It’s the moment arm of the humerus that determines load magnification to the pectorals.

This example demonstrates the simplest use of the “secondary lever”. It acts simply as a neutral connector (a “support beam” or a “pendulum”) of the resistance to the primary lever, without increasing nor decreasing the primary limb’s magnification. It also does not involve the muscles (to any significant degree) which actively move the secondary limb (i.e., biceps, triceps, infraspinatus, subscapularis). It is entirely neutral (“zero moment arm”).

To the untrained eye, it might appear as though the triceps are “extending” the elbow (during Supine Dumbbell Press), but the elbows are actually bending and un-bending passively. The elbow “unbends” simply because the distal end of the humerus rises (as the pectorals contract), and it carries the vertical forearm with it. Of course, there is a small amount of biceps and triceps involvement, preventing the forearm from tipping medially (the triceps) or laterally (the biceps). But as long as the forearm remains perfectly vertical (parallel with gravity, like a plumb bomb or a support beam), there is no need for any significant effort from the triceps or the biceps.

Another example of this passive application of a secondary lever is the One Arm Dumbbell Row.

In the two images below, focus your attention on the distance between the green line on the shoulder versus the dotted line, and the distance between the green line on the elbow versus the dotted line.

In the image above-left, notice that the forearm and the upper arm are both vertical and parallel with gravity. Thus, both limbs have, essentially, a “zero” moment arm. Both limbs are neutral. This means that the muscles which operate the upper arm (in this exercise, ostensibly, the “back” muscles) and the muscles that operate the forearm (the biceps and triceps) are neutral—mostly unloaded—in this starting position.

In the image above-right, you can see that the upper arm lever (the humerus) has changed its angle dramatically, and has moved to the horizontal position, and beyond. Notice how much wider the humerus’ “moment arm” is at this stage (the distance between the shoulder and the weight). This indicates that the muscles that cause the humerus to move from the vertical position to the horizontal position (i.e., posterior deltoids, teres major and latissimus) are most loaded. Conversely, the forearm is still mostly vertical, with a near zero “moment arm”, indicating that the muscles which move the elbow joint (the biceps and triceps) are not loaded much, if at all.

Now, imagine that we could replace this person forearm with a chain or rope hanging from his “elbow” (the distal end of his humerus). Then, we attach a weight on the end of the chain or rope. Doing this would not change the dynamics that is occurring at the shoulder joint (i.e., the muscles that are causing his humerus to rise). In other words, like the previous example, the forearm (during One Arm Rows) is simply connecting the distal end of the humerus and the weight. It’s simply acting as a connector, and if the forearm stays perfectly vertical (at the neutral angle), neither the biceps nor the triceps are loaded.

Tier 2:     The Optionally Active Secondary Lever

Examples: Same exercises as above

In the three images below, you can see how the angle of the forearm influences the degree of magnification of the humerus, which is loading the pectorals.

The image above-left shows a vertical (neutral) forearm / secondary lever. The scale shows 1/1 — the same amount of weight on both sides balances the scale, even though the left side has the weight attached to the “hand” (the hole at the distal end of the secondary lever), while the right side has the weight attached to the “elbow” (the distal end of the humerus).

The image above-center shows how tilting the forearm laterally increases the magnification (the moment arm) of the humerus. The scale shows that half as much weight is needed on the left side of the scale, to balance twice as much weight on the other side.

The image above-right shows how tilting the forearm medially decreases the magnification (the moment arm) of the humerus, such that twice as much weight is needed on the left side of the scale, to balance out the weight used on the right side of the scale.

In addition to increasing or decreasing the moment arm of the humerus, tilting the forearm laterally engages / loads the biceps (commensurate to the degree of the tilt), and tilting the forearm medically engages / loads the triceps (commensurate to the degree of the tilt).

Note: Tilting the forearm sagittally (toward the head or toward the feet) would cause the humerus to ROTATE — either“externally” (toward the head) or “internally” (toward the feet). This would load the subscapularis and teres major (internal rotators), or the infraspinatus and teres minor (internal rotators)—respectively—as they try to prevent further humeral rotation caused by the weighted forearm. This would not be a good thing, as it could strain the rotator cuff muscles if a heavy weight is being used.

A similar thing occurs while doing One Arm Dumbbell Row.

If the forearm (the dumbbell, held by the hand at the distal end of the forearm) is tilted toward the head—decreasing the elbow angle—the moment arm of the humerus is decreased. Thus, the magnification of the load to the muscles moving the humerus, decreases. In addition, the biceps becomes loaded, commensurate to the degree of the forearm tilt.

Conversely, if the forearm (the dumbbell, held by the hand at the distal end of the forearm) is tilted toward the buttocks—increasing the elbow angle—the moment arm of the humerus is increased. Thus, the magnification of the load to the muscles moving the humerus, increases. In addition, the triceps becomes loaded, commensurate to the degree of the forearm tilt.

Tier 3:     The Secondary Lever Meets Ground Reaction Force

Examples: Barbell Squats / Parallel Bar Dips

“Ground Reaction Force” is essentially “an equal and opposite resistance” caused by a person’s effort against an immovable object. For example, the starting blocks used by a sprinter in a track meet, “produce” a ground reaction force that is proportional, and in the exact opposite direction, as the diagnonally-down-and-back force delivered by the sprinter against the starting blocks, when he propels himself forward.

Let’s look at an other example, a bit more closely. A man weighing 200 pounds, preparing to Squat with a 200 pound barbell on his shoulders, delivers a vertical downward force against the ground that is equal to 400 pounds. In turn, the ground “pushes back” with an equal and opposite amount of resistance. This is called “ground reaction force.”

Since the direction of force produced by the man’s weight (plus barbell) is vertically down, the direction of the ground reaction force is vertically up. Thus, the “straight upward” direction of resistance—depicted by the “up” arrow in the images below—is the line of force which will be compared against the angle of the limbs, to determine the approximate percentages of load placed on the participating muscles.

In the image above-left, the lower leg (which can be considered the primary quadriceps lever, as well as the potential secondary lever of the gluteus), the femur (which can be considered the primary hip extension / gluteus lever in this scenario), and the torso (which can be considered a potential hip extension lever, as well as the primary spinal extension lever) are all vertical—parallel with the direction of resistance. Therefore, each of them is “neutral”. Each of the three “levers” are acting simply as “support beams” for the weight being held at the top, in this position.

In the image above-right, you can see that the lower leg is not free to be held vertically. It assumes the angle that it must, by default.

“Allowing” the lower leg to tilt away from the vertical position is not a matter of choice, nor a matter of carelessness—as it could be the case in the previous examples (Tiers 1 and 2). In this scenario, the degree of lower leg tilt is determined by the need to maintain balance. In order to prevent falling forward or backward, an equal amount of mass / weight must be maintained in front of the feet, as behind the feet.

A person performing a standard Squat must produce a direction of force that is directly opposite the direction of resistance—or else risk losing his balance and falling. If he “pushes” in a slightly forward-down direction, he’ll fall backward. If he pushes in a slightly backward-down direction, he’ll fall forward.

The obvious objective (i.e., the requirement) of the exercise is to make the body RISE upward from the descended position, and this can only be done by producing a straight downward direction of force. That direction of force causes the femur to encounter the greater “moment arm”, when the person is in the descended position. It is the 90 degree downward rotation of the hip angle (i.e., the femur) that produces most of the upward movement of the torso. This is exactly like the upward rotation of the shoulder joint (the humerus, when doing Supine Dumbbell Press, shown above) which causes the rise of the dumbbell.

However, unlike the Supine Dumbbell Press, the “secondary lever” (the lower leg) cannot be held vertically—in the neutral position—by choice. It must now “double under” the femur—similar to the medial tilt of the forearm during Supine Dumbbell Press. Given that the quadriceps is essentially the “triceps” of the leg, it becomes loaded—just like the triceps becomes loaded when this happens during the Supine Dumbbell Press. This tilt “shortens” the moment arm of the femur, just as the forearm shortens the moment arm of the humerus, when the forearm is tilted medially during Supine Dumbbell Press.

Note: The forward tilt of the torso, during Squats, isometrically loads the erector spinae. Given that the torso is longer than the lower leg, and given that many people tend to tilt their torso forward quite a bit (in order to maintain balance, given their individual femur and tibia lengths), the erector spine could be loaded MORE than the quadriceps.

This is why the degree of quadriceps participation during Squats is similar to the degree triceps participation during a Supine Dumbbell Press, with a medially tilted forearm.

During standard Squats, the moment arm of the lower leg (i.e., the distance between the foot and the knee, with vertically drawn lines), is very short because the lower leg does not tilt very far from from the mostly vertical position. It does not go far beyond point of parallel with gravity (usually between 20 and 30 degrees from vertical). Conversely, the femur usually reaches a horizontal angle (possibly a few degrees less or more)—so it is mostly perpendicular with gravity. This perpendicular angle is theoretically good for the gluteus, but it’s compromised (to a degree) by the doubling-under of the lower leg, which shortens the femur’s moment arm.

Another example of this type of “Secondary Lever Participation” occurs during Parallel Bar Dips.

In the image below (showing Parallel Bar Dips), I’ve marked the hand as “A” (where the resistance is applied), the elbow as “B” (the joint moved by the triceps), and the shoulder as “C” (the joint moved by the anterior deltoids / pectorals). The green lines are vertical because that is the direction of resistance (as demonstrated on the scale in the earlier demonstration). The direction of anatomical force produced by the person doing Parallel Bar Dips is vertical—on the same plain as the direction of resistance—as it must be, given the straight-down pull of gravity. The person’s downward application of force causes his body to rise straight up—opposite the pull of gravity.

Notice the “moment arm” of the forearm—the distance between A and B. Now, refer back to this similar scenario during Supine Dumbbell Press. It’s the exact same physics, except that now (with Parallel Bar Dips) the forearm is below is below the elbow, whereas it was above the elbow when doing Supine Dumbbell Press. Either way, a vertical forearm is “neutral”, and tilting the forearm toward the shoulder loads the triceps—but only to the degree that the forearm tilts away from the neutral position. If it only tilts to small degree, it only loads the triceps a little bit.

The difference with Parallel Bar Dips and with Squats (Tier 3 scenario)—as compared with Supine Dumbbell Press—is that the person does not have the option of causing his secondary lever (his forearm, and his lower leg, respectively) to be in the neutral position. The application of “ground reaction force” (instead of dumbbells) requires a slight tilt away from the vertical position, in order to maintain balance and to produce force that is directly opposite the direction of resistance.

Notice also, during Parallel Bar Dips, the degree of “moment arm” that occurs with the humerus—the distance between A and C. It has a much greater moment arm, just as occurs during a Supine Dumbbell Press. Therefore, the humerus is loading its corresponding muscle(s)—those that move the shoulder joint (the anterior deltoids and pectorals)—significantly more than the forearm is loading its corresponding muscle (the triceps).

Tier 4:  The addition of lateral Ground Reaction Force (instead of vertical ground reaction force)

Examples:  Barbell Bench Press / Chin-Ups / Lunges

In the images below, we see the descended position, and the extended position, of a Barbell Bench Press.

Unlike the situation which occurs when using dumbbells, having the hands gripped onto the barbell prevents the person from being able to move the arms medially — to bring them toward the midline of the body (i.e., The distance between the hands cannot be moved from “wide apart” at the bottom, to “narrow” at the top, when using a barbell). This results in the person being obligated to produce a linear-outward direction of force. This force originates at the shoulder joint, and moves “through” the hand—represented by the white arrows above.

To help prevent the hands from sliding outward (laterally) the barbell is typically “knurled”, and some people also use chalk. If the bar were not knurled and made even more slick with oil, the hands would certainly slide outward—evidence of the linear-outward direction of force produced by the person.

Just like the sprinter’s starting blocks, this lateral direction of muscle force results in an equal and opposite ground reaction force—represented by the red arrows above. As a result, the direction of resistance is no longer vertical, in terms of the analysis of the “moment arms” of the primary and secondary levers, during barbell Bench Press.

Consider the fact that—in the image above-left—that particular angle of forearm (i.e., vertical) would be considered neutral, when using dumbbells. However, now—when using a barbell—the forearm is not neutral because it is not parallel with the direction of resistance (those red arrows). In this scenario, the triceps are called to action even though the forearms are vertical, due to the angle of the forearm relative to those red arrows. In the image above-left, the forearm angle is about 45 degrees relative to the red arrows (i.e., the direction of resistance). This results in “more than zero” load on the triceps, but “less than maximum”—as a percentage of “the available resistance*.

(* The “available resistance” is defined as the weight being used, multiplied by the length of the operating limb, multiplied by the angle of that limb relative to the direction of resistance…using the simplified reference, that a “parallel to resistance” angle is zero, and a “perpendicular to resistance” angle is 100%).

Note: The lateral direction of force being applied against the barbell—by each arm (the right hand pushing diagonally-right, and the left hand pushing diagonally-left), is itself an equal an opposite force. If it were not for the left hand pushing in a left-ward direction, the force of the right hand pushing in a right-ward direction would cause the entire barbell to shift to the right. That does not happen because each hands’ lateral force (in opposite directions) is equal, and thus keeps the bar centered.

Another example of this type of lateral ground reaction force occurs during Chin-Ups, shown below.

The white arrows in the image above-right indicate the direction of force that must be produced by the person. The red arrows indicate the direction of the ground reaction force that is caused as opposition to the muscular force produced by the person.

Since the grip on the chinning bar does not move (just as the sprinter’s starting blocks, and the ground under the person who is Squatting, do not move), the man’s pulling force results in his body rising toward the chinning bar.

Although it might appear that the direction of muscular force is straight-down-vertical (indicated by the green arrows), that’s actually impossible because no force-producing part of the anatomy is situated directly below the hand. Rather, the pulling-force on the Chin-Up bar originates from the shoulder joint. This results in the right hand pulling diagonally-left (toward the midline), and the left hand pulling diagonally-right (toward the midline), against a bar that does not allow the hands to slide “inward” toward the midline of the body.

This “inward” direction of muscular force—in this scenario—is more perpendicular to the forearms (i.e., the secondary lever) than would be the case if the muscular force could be straight-down vertical (parallel with the forearm). This is the reason why there is a fair amount of biceps participation during Chin-Ups, whereas there is little or no biceps participation during One Arm Dumbbell Rows.

Similar to what occurs during the Barbell Bench Press, the inward pulling of each arm during Chin-Ups opposes that of the contralateral arm. The right arm’s effort of pulling diagonally-left would tend to shift the entire body toward the right, if not for left arm’s effort which is pulling the body toward the left. The equal and opposite diagonal pull of each arm keeps the body in the center.

Notice that any lateral ground reaction force—a sideways force that is separate from the vertical force produced by simple gravity—requires a secondary point of contact, which causes that lateral force. For example, a One Arm Chin-Up (shown below) would produce no lateral ground reaction force (assuming the person is strong enough to do it). The force production required by the person would then only be vertical. However, when the second arm (the second point of contact) is included in the Chin-Up, each arm produces the lateral ground reaction force that opposes the contralateral arm.

Illustrated below is the third example of this type of ground reaction force—Lunges. In this case, it is momentum (initiated by the rear foot, the second point of contact) that produces the lateral force, which then combines with the straight down direction of gravity, and produces a “composite” direction of resistance.

The green arrow in the image above-right indicates the trajectory of the body weight caused by the forward step. This altered direction of resistance (i.e., the vertical pull of gravity combined with the slight forward momentum), is then suddenly halted when the forward foot meets the ground. The angle of “landing” causes an equal and opposite ground reaction force (i.e., the black arrow), which crosses the tibia more perpendicularly than would occur without that forward momentum (initiated from the rear leg). Thus, this ground reaction force (eccentrically) loads the quadriceps more—as a percentage of the weight being used—than would occur with a straight Squat.

Note: When the forward-stepping foot “lands” (makes contact with the ground), you can often feel your foot slide forward into the front part of your shoe, demonstrating the forward momentum and corresponding ground reaction force, which halts your continue forward motion. If the ground were slick and oiled, you would discover (“the hard way”) that this forward momentum is occurring, if you didn’t anticipate it.

If the second half of this Forward Lunge involves returning to the original position, it would require a (concentric) forward thrust from the tibia / quadriceps, such that it would propel the body BACK over the rear foot. That force would cross the tibia more, and engage the quadriceps more, as compared with the person continuing the step FORWARD, out of the Lunge. Stepping forward would engage the gluteus more, because the ground reaction force would then be more perpendicular to the front femur.

Note: Lateral ground reaction force is sometimes incorrectly referred to as “friction force”, possibly because the term “ground reaction force” is thought to pertain only to resistance from the ground. However, “friction force”—as the name implies—is the friction that is created by something rubbing, sliding or dragging across a surface, like a baseball player sliding into home base. If there is no “rubbing, dragging or sliding, there is no friction. If the pushing or pulling is against an immovable “grip” or surface, then it is simply an “equal and opposite direction and amount of force”, which is the definition of “ground reaction force.”

Tier 5:     The addition of a “brace”, combined with lateral ground reaction force

Examples: Swiss Ball Wall Squats / Hack Squat Machine

In the image below, you can see that this person is leaning back against a Swiss Ball, which is against the wall.  This is a “brace” (a second point of contact), and it alters the direction of resistance used during an exercise.

As you can see, her feet–which are placed on the ground in front of her (i.e., not directly beneath her torso and hips)–can only push in a diagonally forward-down direction (indicated by the red arrows), due to leaning back against the brace. If the ground were slick and oiled, her feet would slide forward (indicated by that green arrow). However, since the ground is not slick and her feet are “gripping” the ground, a reaction force is created that is in a diagonally-up-and-back direction of resistance (opposite those red arrows).

Note that, in the image above-right, you can see that her lower legs are vertical. However, they are not “neutral” in this situation, because the direction of resistance is not vertical. Thus, this diagonal direction of resistance loads the quadriceps more, as compared with a vertical direction of resistance. During this exercise, the quadriceps are able to interact fairly perpendicularly (more productively) with the modified direction of resistance.

It’s important to note, however, that the direction of resistance is still not “maximized”. The resistance is not being applied to the lower leg lever 100% perpendicularly. The angle of the resistance, relative to her tibia (in the descended position) is about half way between zero (neutral) and maximum (100% perpendicular). Therefore, the load to the quadriceps is a percentage* thereof.

* Note: Trigonometry would be required to calculate the exact amount of load a muscle would encounter given a particular limb angle relative to the direction of resistance. However, for the sake of simplicity—for the purpose of allowing a general understanding of how a limb angle (relative to the direction of resistance) loads a muscle more, and less—it’s convenient to consider a limb angle that is 45 degrees to the direction of resistance (i.e., half way between zero magnification and 100% percent magnification), a “50% lever”. Thus, it would load its corresponding muscle with approximately half as much load as would occur if that limb were completely perpendicular to the direction of resistance. The objective is to have a sense of how muscle load increases, or decreases it, based on the angle of the operating limb, relative to the direction of resistance.

Also note that a percentage of the “weight” (her bodyweight plus the weight she is holding) is “resting” against the Swiss Ball (i.e., “the brace”). This reduces the amount of “weight” that’s loading the exercise, to some degree. Therefore, calculating the amount of load her quadriceps encounter during this exercise would require an estimation which includes that reduction, plus the “moment arm” of her lower leg angle (i.e., the degree of tilt between neutral and perpendicular), plus the length of her lower leg lever (tibia).

Also, notice the angle of her femur (in the descended position) in relation to the red arrow. The two black lines I’ve placed there show the moment arm of the tibia, but that “space” (between the two lines) is the same as that of the femur. What this shows is that the lower leg essentially negates the moment arm of the femur.

Imagine doing a Supine Dumbbell Press (as shown above), but–instead of medially tilting your forearm a few degrees–you tilt your forearm to the point where the dumbbell is directly over your shoulder (in the descended position). The moment arm of the humerus would then be no greater than that of the forearm. That would cause the triceps to be very loaded, but would essentially eliminate the moment arm (the leverage) of the humerus — so there would be little (if any) pectoral load. This is why attempting to work two muscles at one time is inherently problematic. Either both muscles (both operating limbs) experience a reduced load, or muscle / limb is very loaded while the other one is not loaded much at all.

Thus, the exercise above requires very little dynamic hip extension (i.e., force produced by the glutes, etc.). Rather, it’s the lower leg levers—and subsequently the quadriceps—that are doing most of the work. The knees forcefully extend (from the descended position), which pushes the pelvis/lower spine up-and-back, which causes the ball to roll upward, which carries the torso upward. There is very little moment arm with which the femur can produce force, by way of the gluteus.

In the example below—the Hack Squat Machine—we have a very similar situation as above, except that there are two braces: one on which his back rests, and one which is over his shoulders. Again, his quadriceps produce most of the force necessary to rise out of the descended position. His direction of force (indicated by the red arrow) originates at the pelvis, which leaves the femur with very little moment arm / leverage. That’s what makes this exercise “very good” for the quadriceps.

It’s worth noting that the angle of the sled (apparently 40 degrees, or so) reduces the actual amount of weight with which the exercise is loaded—as a percentage of the total weight (i.e., the sled plus the weights attached to it). If the sled were perfectly vertical, 100% of the weight would load the exercise.

As reference, imagine the sled angle being nearly 100% horizontal. There would be very little (if any) load against which the legs would have to work, even if the sled was loaded with a considerable amount of weight.

During this type of Hack Squat, the erector spinae (the muscles that extend the spine) are entirely without load—which is good, in this application. The weight of the torso is essentially resting on the back-pad, relieving the erector spinae from having to work against a forward pulling resistance (which occurs during standard Barbell Squats).

The “downward” thrust of his legs originates from his pelvis/base of his spine, and pushes through his feet (indicated by the red arrow). That direction of force produces an equal and opposite ground reaction force (the green arrow). That “line of force” is approximately 45 degrees to the lower leg levers, but it also mostly negates the moment arm of the femur, as the operating lever of the hip extension muscles (i.e., the gluteus, etc.).

Again, you can see the inverse relationship between a maximized moment for the quadriceps, and a minimized moment arm for the gluteus. This is always an issue when dealing an exercise that involves two disparate joint functions, and one direction of resistance.

If a person places his feet farther out in front (during Hack Squats), he’ll produce more forward thrust (against the “grip” of his feet on the platform), rather than a downward thrust. However—in this case—either foot position will produce a ground reaction force that loads mostly the quadriceps, and minimizes the gluteus (hip extension). This is because in both cases, the lower leg lever minimizes the moment arm of the femur.

In a sense, you could say that the muscle loading which occurs during a Hack Squat (machine) resembles the muscle loading which occurs during a Sissy Squat, more so than it resembles the muscle loading which occurs during a standard Barbell Squat.

During Hack Squats and Sissy Squats, the quadriceps are predominantly loaded, and the gluteus participation is minimized. During Barbell Squats, the gluteus is loaded more, and there is less quadriceps participation. This can be identified by the examining the moment arm of either lever (the femur and the tibia) in these exercises.

The Hack Squat machine allows more weight to be moved—as compared with Sissy Squats—for several reasons. There is a percentage of load reduction caused by the angle of the sled of the Hack Squat; the lower leg lever usually does not reach the fully perpendicular angle with the direction of the ground reaction force; and the femur angle has a small amount of leverage / moment arm (…there is bit more gluteus participation than occurs during Sissy Squats); and the issue of balance / coordination is eliminated. However, this does not necessarily mean that the quadriceps are more loaded with Hack Squats, as compared with a Sissy Squat. Maximum quadriceps effort is maximum quadriceps effort, either way.

There may also be a psychological component that occurs during Hack Squats, such that the person is compelled to put forth greater effort and perform more reps, given the heightened sense of stability.

Tier 6: The addition of a non-linear (curved) primary direction of resistance, combined with a lateral ground reaction force, plus a brace.

Example: Pendulum Squat

In the image above-left, I’ve placed a large curved green arrow, indicating the trajectory of the lever arm of this machine—a Pendulum Squat. However, that is not necessarily the direction of resistance experienced by the person, because he has a brace (second point of contact) against his back which allows him to push more backward than upward. This, combined with is foot placement—slightly in front of him, rather than directly under him (i.e., in line with his torso)—causes a modified direction of resistance. This is a type of ground reaction force that is similar to the Swiss Ball Squats.

The actual direction of resistance (due to the ground reaction force) is represented by the orange arrow, which is directly opposite the person’s direction of force (lime green arrow), pushing back against the pad. The direction of force originates at his pelvis, and pushes through his feet. That is the line of force, highlighted by the red line at this foot. The other red line (at his knee) allows you to see the moment arm of his tibia (highlighted with a thin white line).

This is a fairly good angle (moment arm) for the tibia, relative to the direction of resistance. The tibia is not  “maximized”—it does not reach a fully perpendicular angle with the line of force—but it is still mostly good. It’s a bit closer to perpendicular with the line of force, than it is parallel with the line of force.

The advantage of this scenario, is that there is essentially no consequence to adding additional weight to the machine, to compensate for the slightly less than perpendicular angle of the tibia, relative to the direction of resistance. His back / spine is totally supported (un-weighted), so there is no risk of excessively forward loading the torso / erector spinae—which occurs during standard Barbell Squats.

Again—similar to what occurs when doing Hack Squats—the angle of the lower leg (which is advantageous for the quadriceps), essentially negates the moment arm of the femur, in terms of gluteus loading. As such, there is much less gluteus loading with this exercise, as compared with quadriceps loading.

More interesting, perhaps, is the fact that the clockwise rotation of the machine’s lever arm, matches the clockwise rotation of the tibia around the knee axis. In essence, the curved downward trajectory of the machine “pushes” the carriage (with the person on it) toward his heels, thereby ensuring a perpendicular angle of force to the tibia.

Ironically, the clockwise rotation of the machine is opposite the counter-clockwise rotation of the femur, around the hip axis. This further minimizes the mechanics that would favor gluteus loading. As the machine rises, it prevents the femur from fully extending—ultimately stopping 40 to 50 degrees short of full gluteus contraction.

In the image below—the final position of the Pendulum Squat—you can see that the knees are fully extended, but the hip joints are not. The knees have traveled approximately 110 degrees, with a fairly perpendicular tibia (relative to the line of force), while the hip joint has only traveled approximately 70 degrees, with femur that had a drastically reduced moment arm (caused by its secondary lever—the tibia).

Also, similar to the Hack Squat, the stability provided by the Pendulum Squat machine, combined with the back support, allows the person to add additional weight without much concern for balance, or risk of injury. This might psychologically incentivize the person to use more effort, as compared with an exercise that does not provide this type of “security” / stability.

Another aspect that is unique to this exercise is that the resistance varies, depending on the location of where, along the curved trajectory, the person and carriage are. When in the fully descended position, the resistance is “least”—because the moment arm of the pendulum has decreased. When in the fully ascended position, the resistance is “most”—because the pendulum has moved laterally farther away from the pivot (as measured with vertical lines), maximizing the machine’s moment arm. This could be viewed as potentially beneficial, because the moment arm of the person’s tibia is maximized in the descended position (where the machine provides a bit less resistance), and his tibia’s moment arm is reduced in the ascended position (where the machine’s resistance is maximized). However, in practical terms, it would likely not result in a significant advantage, as compared with a more “standard” resistance curve.

The exercise below has a very similar biomechanical profile to the Pendulum Squat. The principles are the same—curved trajectory of resistance, back support (brace) allowing the person to push “back” against the brace, creating a ground reaction force that is more perpendicular to the tibia than a standard Squat. But, here again, the mechanics that favors the quadriceps (i.e., the tibia), compromises the gluteus (i.e., the femur). It’s impossible to optimize both at the same time, because they each “need” (would most benefit from) their own, separate direction of resistance.

The disadvantage of this exercise (above)—as compared with the Pendulum Squat—is that it’s more difficult to load (requires hand-holding the weight), and it’s cumbersome to set up (connecting the straps while also picking up the weights). Once in place, however, it’s a good quadriceps exercise.

The above “6 Tiers” illustrate that most exercises have their own, relatively unique biomechanical profile. Yet, each exercise can be dissected and evaluated once the parameters for evaluation are understood.

To an untrained eye, some exercises might seem to have an “obvious” biomechanical profile (i.e., which muscles are loaded seems “obvious”). However, upon closer inspection, there’s more to it than meets the eye. For example, in this exercise below—Sissy Squat Bench—it appears very similar to a standard Squat, but in fact it’s very different.

There are two braces in play here, and they alter the direction of resistance from vertical-down, to ROTATIONAL—pivoting around the brace that is positioned on the posterior side of the proximal tibia (i.e., upper / back side of the lower leg).

In fact, you can see (in this image above) that his feet are not even touching the floor plate—as would occur if the direction of resistance were straight-down vertical (like standard Barbell Squats). In this case, his bodyweight would essentially fall backward, if not for the forward brace in front of his ankle. This rotational force and brace produce an equal and opposite ground reaction force against the front (distal end) of his tibia, which loads his quadriceps.

In the image below-left, you can see that his knee is the axis of rotation—caused by the brace behind the knee, and the shift of his bodyweight posterior to the apex above the knee. This causes a forward thrust of the ankle against the roller–similar to the ground reaction force discussed in Tier 3 (the brace as an additional point of contact, which redirects the resistance).

On a separate note, he is leaning his torso (from this perspective) to the left of (anterior to) the apex above his hip joint (i.e., the base of his torso). This forward leaning “shortens” (reduces the moment arm of) his femur. This is similar to the medial tilting of the forearm during Supine Dumbbell Press, in Tier 2 (the “optional tilt of the secondary lever”).

In the image above-right, the man is allowing his torso to lean all the way back, which is similar to the extreme lateral tilting of the forearm during Supine Dumbbell Press, in Tier 2. It dramatically lengthens the femur, which greatly increases its load magnification to the quadriceps. Notice the two vertical dotted lines, which show the moment arm, and compare that “width” to the moment arm of the image above-left (from his head to his knee).

In the image above-right, there can be no load whatsoever to the gluteus, when the torso is leaning to the posterior side of the apex above the pelvis. The image above-right also has considerable engagement from the rectus abdominis and hip flexors, which are preventing his torso from collapsing backward. The image above-left has a very small amount of gluteal engagement, and a reduced quadriceps load, due to his forward leaning torso.

Identifying and understanding the effects of moment arm changes, and the corresponding increases or decreases in muscle load which are influenced by the angle of a limb relative to the direction of resistance, is critically important to understanding which muscles are loaded, and to what degree they are loaded. Also important is the ability to identify changes in the direction of resistance which are caused by the addition of lateral ground reaction forces, braces, and non-linear trajectories of resistance.

Doug Brignole is the author of “The Physics of Resistance Exercise”, a biomechanics researcher, and a former bodybuilding champion with a 43 year history of competition.

Lombard’s Paradox: Misapplied

In 1903, professor of physiology Warren Lombard wrote an article in which he described (what he believed was) a paradox that occurs when a person performs any type of Squatting movement. It was dubbed “Lombard’s Paradox” because he was unable to understand how a person was able to perform a Squat, despite an apparent conflict of interest.

A “paradox” refers to a situation which appears to violate logic—or at least it violates a person’s current understanding of what logic would dictate. It does not mean that rules don’t apply. It simply means that the mechanism for HOW something works, is not understood.

What perplexed Professor Lombard was how it was that opposing muscles (i.e., the quadriceps and the hamstrings) could contract simultaneously without that preventing the successful execution of the Squat. He reasoned that those two muscles “should” cancel each other out—each preventing the other from producing movement.

The Squat motion is a two-joint action, whereby the knee joint and the hip joint are simultaneously extended.

The quadriceps extends the knee joint, and the hip joint is normally extended by the gluteus, femoral adductors and hamstrings—collectively known as the “hip extension muscles”.

Lombard wondered: IF the hamstrings (which crosses both the hip joint and the knee joint) contracts because of its participation in hip extension, that same contraction (pulling the origin and insertion closer together) would also force the knee to flex (to bend). Yet, during Squats (concentric phase), the knees are extending—they are not flexing.

His question was, “Why are the hamstrings not preventing the quadriceps from successfully extending the knees?” (during the concentric phase).  Wouldn’t the quadriceps and the hamstrings be opposing each other—the quads trying to extend the knees, and the hamstrings trying to flex the knees? The result “should” be that the no knee action can occur, he thought.

Three years later, in 1906, neurophysiologist Charles Scott Sherrington established “Sherrington’s Law of Reciprocal Innervation”, which identified the role that the Central Nervous System plays in controlling the degree of “innervation” that participating muscles receive, when conflicting anatomical actions are required. This is explained in my book, “The Physics of Resistance Exercise”.

Sherrington observed that an antagonist muscle—in this case, the hamstrings—would receive a “relaxation synapse” (a message) from the Central Nervous System, triggered by the activation of the quadriceps, which causes the hamstrings to REDUCE its degree of contraction so as to not interfere with the efforts of the quadriceps.

What’s important to note here is why it is that the hamstrings receives that relaxation message, and not the quadriceps.

The obvious reason is that the quadriceps is the only muscle that extends the knees—the knees cannot extend without activation of the quadriceps. No other muscle participates in that action.

Conversely, the hamstrings is not the only muscle that extends the hip. The hamstrings is only a participant—one of the two lesser assistants to the gluteus, which is the primary hip extensor.

The hip joint can be successfully extended without much participation from the hamstrings, so the CNS selectively “calms” the activation of the hamstrings, in order to allow the coordinated actions which result in simultaneous knee and hip extension—i.e., Squats.

The fact that the knees do extend during Squats, is evident proof that the hamstrings are not fully activated. If the hamstrings were fully engaged, they would interfere with the quadriceps and knee extension would not occur. The notion of “co-contraction”, therefore, is largely negated.

Lombard was obviously not aware that multiple muscles which participate during a compound exercise do not all activate equally (i.e., produce force that is equal to that of the other participating muscles). He did not know that there are mechanisms which allow the collaboration of contributing muscles, so anatomical motions can be executed without interference.

The varying degrees of muscular activation among the participating muscles is determined by neurological factors—such as reciprocal innervation—and also by mechanical factors influenced by the location of muscle origins and insertions, and how that determines the amount of leverage each muscle has during a particular action.

The “mystery” of why the hamstrings allow the quadriceps to extend the knees during Squats is no longer a paradox. Likewise, the reduced activation of the rectus femoris (a hip flexor and knee extensor) during Squats, is also no longer a paradox. These are both known facts now.

The knees can be successfully extended without the participation of the rectus femoris because the other three parts of the quadriceps group can accomplish the task. Conversely, hip extension cannot be achieved without contraction of the gluteus. Thus, the gluteus muscle is prioritized by the Central Nervous System, and the relaxation synapse is sent to the rectus femoris, as well as to the other four hip flexion muscles—the Psoas, Iliacus, Sartorius and Tensor fascia lata.

Reduced activation of the hamstrings, and reduced activation of the rectus femoris (during Squats) is not just theoretical. It has been observed and measured in studies (Chris Beardsley / Brad Schoenfeld). However, these observations tend to be ignored by those who dogmatically defend traditional exercises and are reluctant to alter their training rituals.

It’s ironic that some people cite Lombard’s Paradox as an attempt to “support” their continued preference of, and emotional attachment to, Barbell Squats.

Lombard’s Paradox does not declare that the hamstrings or the rectus femoris are activated, during Squats. Rather, it’s a declaration that Lombard was mystified — bewildered as to why the hamstrings allow the quadriceps to extend the knees, wrongly assuming that the hamstrings “must be contracting” because they participate in hip extension. Sherrington explained how and why antagonist muscles (i.e., the hamstrings and rectus femoris) are de-activated (shut off) during Squats, and studies have since confirmed that.

“The Direction of Force” in Exercise Analysis

The direction of the force you produce, during an exercise, determines which muscle is loaded more, and which muscle is loaded less. How do you determine the direction of force you’ll use during an exercise? It is dictated by the direction of resistance provided by each exercise.

Let’s say you (at a bodyweight of 200 pounds) are about to perform a Barbell Squat. You are in the standing position, with a 200 pound barbell on your shoulders. You are therefore pressing straight downward, against the ground, with 400 pounds. If you were to put a scale under your feet, this would be confirmed.

In turn, the ground is pushing back 400 pounds. This is called “ground reaction force”. The ground is not sinking under the 400 pounds of load, nor is it pushing back more than 400 pounds, in which case you would rise.

In physics, the “ground reaction force” rule stipulates that an immovable brace (i.e., the ground, or the starting blocks of a sprinter, etc.) “pushes back” an amount equal to the amount pressing against it, and in the exact opposite direction. This means that the straight downward (vertical) pressing of your bodyweight plus the barbell, during Barbell Squats, is producing a straight upward (vertical) direction of resistance.

That upward direction of resistance, during Barbell Squats, requires that you produce a straight downward direction of force—otherwise, you would fall off balance.

The image below-left shows what occurs during a standard Barbell Squat. The red “UP” arrow indicates the direction of movement you are causing with your downward direction of force (indicated by the red “DOWN” arrow). Notice that the femur (upper leg bone) is perpendicular with your direction of force, while the tibia (lower leg bone) is mostly parallel with your direction of force. Thus, hip extension (produced by the glutes and other hip extensors) is loaded much more than knee extension (produced by the quads), in order to produce this upward movement.

In the figure above-right, you can see what would occur if you intentionally emphasized knee extension–instead of hip extension–thereby fully activating your quadriceps. Maybe you’d like to try this as a test (just kidding). If you do, be sure you have a soft cushion on which to fall, placed behind you.

Pushing your foot forward against the ground (indicated by the lower green arrow) would result in your hips and torso being thrust backward (indicated by the green arrow at the hips) and you would fall off balance. The reason this does not happen when you perform Barbell Squats (i.e., fully activate quadriceps contraction) is because you are coordinating a straight upward trajectory, directly opposite the vertical pull of gravity.

In the image below, I am performing a Cable Squat. At first glance, it might look similar to a standard Barbell Squat, but the key difference is that the direction of resistance is now coming mostly from a FORWARD-downward angle, instead of the vertical angle. The lower red arrow indicates the direction of resistance, which thus requires that I produce a force that is directly opposite (upper red arrow), or else I’ll fall forward. This new direction of force I must produce is significantly more perpendicular to my tibia. Therefore, the quadriceps are significantly more loaded than they would be with the vertical direction of resistance of Barbell Squats.

It’s also worth noting that—when you do this exercise—you can feel your feet sliding forward toward the front part of your shoes, which reflects the forward thrust of the tibia (i.e., the emphasis on “knee extension”, thus more quadriceps activation). This does not occur with standard Barbell Squats.

Another way of ascertaining the degree to which each participating muscle is loaded (quadriceps for knee extension – or – gluteus for hip extension) is to measure the “moment arm” of each limb / lever.

In the images below, you can see the difference between the moment arm of the femur, versus that of the tibia. “A” indicates the moment arm of the femur (being operated by the gluteus), and “B” indicates the moment arm of the tibia (being operated by the quadriceps). The greater the moment arm, the greater the percentage of load that is placed upon the muscle that operates that limb.

The reason why these lines are drawn vertically, is because that is the direction of the resistance during Barbell Squats. The “moment arm” lines of Cable Squats would NOT be drawn vertically. Cable Squat analysis would require that the “moment arm” lines be drawn diagonally, parallel with that direction of resistance (the cable). Thus, those lines would show a greater moment arm for the tibia, instead of the femur.

Further still, in the image below, notice the distance that must be traveled by the head and hips, in order to go from the descended position to the standing position. It’s a considerable distance. Thus, the question that must be asked is, “which limb/lever (and therefore which muscle) is producing most of that distance traveled?”.

In the image below, I’ve placed a protractor over each photo–one centered on the hip joint, and one centered on the knee joint. I’ve also placed a red dot showing the beginning (1) and ending (2) location of the femur (left image) and of the tibia (right image). On the left image, you can see that the distal end of the femur must travel a full 90 degrees from descended position to standing position. Conversely, on the right image, you can see that the tibia only needs to travel approximately 25 degrees, from descended position to standing position.

Which of these two limb angle changes is more likely to produce the degree of torso lift shown in the previous image?  Obviously, it’s the image on the left image: the femur—gluteus—hip extension that produces the majority of the upward rise of the torso.

When calculating force, one of the factors is “distance traveled”. The more distance that is traveled, the more force that must be produced by the muscle that is causing that degree of travel. “Distance traveled” is directly associated with the length of the moment arm. The greater the moment arm, the greater the distance traveled. The shorter the moment arm, the less distance traveled.

These same rules apply to Parallel Bar Dips. Remember, the direction of the force that you are required to produce, determines the degree to which each participating muscle is loaded—and the direction of force you must produce is mandated by the direction of resistance provided by the exercise.

In the image below-left, you can see what occurs during standard Parallel Bar Dips. The direction of resistance (“ground reaction force”) is straight upward, which therefore requires that you produce a straight downward direction of force. That direction of force is mostly parallel to the forearm, which renders it mostly “neutral”—loading the triceps with only about 11% of the force. Conversely, that downward direction of force is mostly perpendicular with the humerus (upper arm bone), which causes the front deltoids to be the most loaded. Notice the degree of movement the humerus must travel (90 degrees)—from starting position to ending position—in order to cause the torso to rise.

The image above-right, you can see what would happen if you were to intentionally emphasize triceps extension, during Parallel Bar Dips. The arrow pointing left shows the direction your hand / forearm would “push” (i.e., think “Triceps Kickback”). This backward push of your hands/forearms (a force that is perpendicular to your forearm) would result in your torso being thrust forward–not upward. You can try this in your chair, with your hands on the arm rests, and you’ll feel the difference in triceps activation and in the way your hands feel on the “bars”.

In fact, during Parallel Bar Dips, there is no “backward push” being produced by the hands, as occurs during typical Triceps Extensions. There is no sliding sensation on the hand, which would result from posterior “friction force” against the bars—as occurs in all effective triceps exercises (i.e., think Cable Pushdowns with a rope, or Dumbbell Skull Crushers with a hammer grip…the way you must squeeze the handle or else your hand will slide). “For every action there is an equal and opposite reaction”…thus, there would absolutely be a significant “backward slide/push” occurring—against the bar, by your hand—if the triceps activation (during Dips) is as great as some people think it is. Yet, there is none of that during Parallel Bar Dips.

Below is yet another example of how the direction of force produced by the person, determines the degree to which a participating muscle is loaded.

In this example, the person is leaning back against the ball. This “backward lean” changes the direction of force the person must produce. In this case, the person must produce a direction of thrust (with his legs) that is angled diagonally forward. He has no choice, given the backward lean. He does not have the option of pushing “straight down” with his legs. If the floor was oiled, his feet would slide forward–proving that his direction of force is slightly forward now. This forward direction of force automatically loads the quadriceps more, because it is more perpendicular with the tibia than a straight down direction of force would be. In this case, it does NOT matter how vertical the tibia is because now the direction of resistance is no longer vertical. It has been altered by the forward “friction force”, created by the feet pushing forward against the floor, which is mandated by the backward leaning against the Swiss Ball. The farther forward the feet are placed on the floor (i.e., away from the wall), the more forward “friction force” is produced, the more perpendicular that direction of force is with the tibia, and the more loaded the quadriceps are.

The image below—Sissy Squat on a Sissy Squat Bench—shows a direction of force that is somewhat similar to the Swiss Ball Squat above, but to a much greater degree. In this case, the body would fall backward if not for the brace in front of the ankle (similar to the friction force of the feet on the floor, above). Thus, the direction of the resistance is NOT vertical—it is now rotational—moving around the pivot that is created by the pad at the top posterior side of the tibia (just below the back of the knee). In fact, you can see that his feet are not even touching the foot pad under them, proving there is zero downward (vertical) force in this example…until he reaches the fully ascended position.

This rotational direction of resistance requires that the force produced (by the distal end of your tibia / your ankle) be applied straight forward, against the ankle pad. That is precisely what makes this an excellent quadriceps exercise. It does not matter that the tibia is vertical (i.e., which would only be considered “neutral” when the direction of resistance is vertical). The direction of resistance is now essentially horizontal, and thus requires that the direction of force the person produces also be horizontal, directly against the ankle pad.

The example above also highlights one more very important thing that is worth noting. As you can see, in this descended position, the hip joint is bent (“flexed”). So, when this person rises to the upright position, it will APPEAR as though the hip joint was actively extended by way of the gluteus muscle—but that is NOT the case, in this exercise. Rather, the hip is being passively extended.

To understand this, imagine the man above has no torso. The knees would extend (straighten) by way of quadriceps contraction, and—as a result—the femurs would rotate up to the vertical position. Anything that is placed at the upper end of those femurs—let’s say you’ve taped a golf ball on the upper end of each femur—would also rise. Now, just replace the golf balls with his torso. The same thing would happen—the torso would simply rise by virtue of the upper ends of the femurs rising. In other words, the torso is just “going for a ride”, carried upward by the rising end of the femurs. The hip extension muscles (the glutes, etc.) do not need to actively produce that angle change at the hip joint, given the physics of this exercise. Of course, the weight of the torso would load the quadriceps (knee extension) much more than the golf balls did, but that is a separate issue.

To further illustrate this, look at the two images below—showing a Supine Dumbbell Press.

The insertion of the pectorals is on the upper end of the humerus. When the pectorals contract, they pull the upper arm from an outward (lateral) position, medially—toward the midline of the body. Now, imagine that you have had both of your forearms amputated. So, you attach a weight directly to the distal end of your humerus, and you perform this exercise–“shoulder adduction” (i.e., which is the function of the pectorals). You would still effectively load your pectoral muscles, even if you have no forearms or hands.

Like the previous example (Sissy Squat bench), if you were to then place a forearm (with a hand and a weight) at the distal end of each humerus, that forearm (with attached weight) would also be carried upward by the rising end of the humerus. Thus, the elbow goes from bent to straight, passively—without assistance from the biceps or the triceps—as long as that forearm stays in the vertical position…balanced over the elbow. This is because (in this case), a “vertical” forearm is the same angle as the direction of resistance, so the forearm is neutral—requiring no force from the biceps or triceps.

If you were to tilt your forearm “outward” (when in the descended position), you would engage your biceps to a degree (i.e., a small percentage—an amount equal to the degree of outward tilt). If you were to tilt your forearm “inward”, you would engage the triceps to a degree (i.e., a small percentage—an amount equal to the degree of inward tilt). This is due to the forearm leaning to either one side, or the other, of the “apex” (the neutral angle).

Again, this proves that a joint is often passively moved (flexed or extended), even though it may appear that the joint is being moved by the biceps or the triceps, or whichever muscle which “could” actively flex or extend that joint under different circumstances.

Below is a final example of this. You might think that—during any rowing exercise—the elbow is bending by virtue of biceps contraction, but that is not necessarily the case.

Imagine that you have no forearm, no biceps and no triceps. Instead, you’ve attached a rope to the distal end of your humerus, with a weight hanging on the bottom end of the rope. You would still be able to perform the above exercise (One Arm Dumbbell Row) simply by contracting the muscles that move the shoulder joint (i.e., the “back” muscles), thereby raising the humerus from the lower starting position, to the ending upper position.

The rope (with the attached weight) would do the exact same thing the forearm is doing in these photos—hanging from the elbow, going along for the ride, rising simply because it is being pulled upward by the distal end of the humerus.

Some people mistakenly believe that rowing exercises function as a biceps exercise, under the misguided belief that “since the elbow is bending, it must be like when we do Curls”. No—it’s not like Curls at all.  When we do Curls, we cause the forearm to become perpendicular with gravity (i.e., horizontal). Conversely, when we row, the forearm stays mostly parallel with gravity—in the neutral position.

Of course, if you “pull” your forearm–angle your forearm toward your shoulder–you would engage the biceps a little bit. If you were to “push” your forearm–angle your forearm toward your hips–you would engage the triceps a little bit. But as long as that forearm stays vertical (parallel to gravity), neither the biceps nor the triceps would be loaded.

Again, the direction of resistance provided by the exercise dictates the direction of force that you must produce. If the direction of force you produce is mostly parallel with a limb, the muscle that operates that limb is mostly neutral—not loaded much, if at all. If the direction of force you much produce is mostly perpendicular to a limb, the muscle that operates that limb will be very loaded.

It’s not merely a matter of a limb being “vertical” or “horizontal”, because that ignores the factors that influence the direction of resistance. In order to determine whether a muscle is loaded with a significant percentage of the available resistance, you must first identify the direction of resistance provided by each exercise, then determine the direction of force you must produce based on the demands of that exercise, then compare the direction of force you produce to the angle of the limbs involved (i.e., you must also know the anatomy, to know which muscles move those limbs).

Understanding (Not Misinterpreting)
“Knee Strain”: Leg Press vs Leg Extension

This particular study was done in 1993, and was published in PubMed.gov: https://pubmed.ncbi.nlm.nih.gov/8346760/

Unfortunately, some people refer to this study as “evidence” that Leg Extensions are “bad for the knees”. However, that is NOT the conclusion of this study. This is an example of cognitive bias, whereby people hear or read what they WANT to hear or read, rather than what has actually been demonstrated.

Here, below, is the report:

Biomechanical considerations in patellofemoral joint rehabilitation

Patellofemoral joint biomechanics during leg press and leg extension exercises were compared in 20 normal subjects (10 men, 10 women) aged 18 to 45 years. Knee moment, patellofemoral joint reaction force, and patellofemoral joint stress were calculated for each subject at four knee flexion angles (0 degree, 30 degrees, 60 degrees, and 90 degrees) during leg press and leg extension exercises. All three parameters (knee moment, patellofemoral joint reaction force, and patellofemoral joint stress) were significantly greater in leg extension exercise than leg press exercise at 0 degree and 30 degrees of knee flexion (P < 0.001). At 60 degrees and 90 degrees of knee flexion, all three parameters were significantly greater in leg press exercise than leg extension exercise (P < 0.001). Patellofemoral joint stresses for leg press and leg extension exercises intersected at 48 degrees of knee flexion. This study demonstrates that patients with patellofemoral joint arthritis may tolerate rehabilitation with leg press exercise better than with leg extension exercise in functional ranges of motion because of lower patellofemoral joint stresses.html

It’s important to understand precisely what was “found” during this study, and precisely what was concluded based on that information.

A comparison was made between a Leg Press and a Leg Extension (machine), in regards to “patellofemoral stress”. Researchers were trying to determine which of these two exercises, and which degrees of knee flexion, would produce more knee strain—specifically for “patients with patellofemoral (knee) arthritis”.

Knee stress (load) was evaluated—during both exercises—at four different knee positions (degrees of knee bend): at 0 degrees of knee bend (i.e., straight knees); at 30 degrees of knee bend; at 60 degrees of knee bend; and at 90 degrees of knee bend.

The researchers found that “knee stress was significantly greater during Leg Extension, than during Leg Press, at 0 degrees of knee flexion and at 30 degrees of knee flexion”.

In essence, what this sentence states, is that when the knees are straight, and when the knees are only slightly bent, the Leg Extension exercise produces more knee force, as compared with those same knee positions during the Leg Press exercise.

No surprise here. When the knees are straight, during Leg Press, there is zero activation of the quadriceps because the direction of resistance is parallel with the lower legs (i.e., they are in the “neutral position”). However, when the knees are straight during Leg Extension, the quadriceps are fully activated because the lower leg is encountering a perpendicular force against the front of the ankle, from the machine’s lever arm / roller pad.

However, it is essentially useless (i.e., produces almost no muscle benefit) to “perform” a Leg Press with straight knees—or with knees that are only a bent to a small degree. The straight-knee position (during Leg Press) is not even an “exercise”, any more than “standing” (with straight knees) is an exercise. That position is what happens BEFORE the knees begin bending, and the lower legs encounter some degree of perpendicular force, relative to the direction of resistance.

It’s obvious that the knees participate more during Leg Extensions when the knees are straight—because that is not a “resting position”, as it is when the knees are straight during a Leg Press. During Leg Extensions, the quadriceps are fully activated when the knees are straight (image below-right). It is logical to expect more knee participation (“stress”) when the quadriceps are fully activated, and less knee participation when the quadriceps are not activated, or are only activated to a small degree.

What this study also found, but which many people seem to miss, is that: “At 60 degrees and 90 degrees of knee flexion, knee load was significantly greater in the Leg Press exercise than in the Leg Extension exercise”.

Again, this is no surprise. During a Leg Press, the more the knees bend, the more perpendicular the angle becomes between the lower legs and the direction of resistance (i.e., the pathway of the Leg Press’ carriage, which is holding the weight). The more perpendicular that angle is, the more the percentage of the load on the quadriceps increases, as does the participation of the knees.

In the photo above-RIGHT, the red line shows where the upper and lower leg would be when it’s entirely straight (knees at 0 degrees bend). In that position, the lower leg is parallel with the direction of resistance (the trajectory of the sled), so it would provide zero load to the quadriceps. The yellow line shows the angle of the lower leg when the knee is bent at 30 degrees.

There are two things worth noting here: 1) that is a very compromised range of motion for the quadriceps — possibly as little as 1/5 (20%) of its “full range” ability. Also, 2) the angle of the tibia (relative to the direction of resistance) in that mostly straight-knee position, only loads the quadriceps with approximately 1/4 (25%) of the weight that is on the machine, even though 100% of that weight is loaded onto the bones.

In the photo above-LEFT (i.e., the fully descended position in the Leg Press), you can see that the angle of the knees has gone beyond 90 degrees—it’s approximately 118 degrees. Yet, even with this maximum degree of knee bend, the tibia has still not reached a perpendicular angle with the direction of resistance. In fact, the tibia has only reached a 60 degree angle, relative to the direction of resistance. 

What this means is that, in order to load the quadriceps with the same amount of resistance as could be achieved with a Leg Extension exercise (which provides a completely perpendicular direction of resistance, against the tibia), you would have to use approximately 30% more weight on the Leg Press machine (as compared with the Leg Extension), at this particular degree of knee bend.

This is the reason why Leg Presses cause more strain on the knee joint, when the knees are bent beyond 30 degrees. You must use more weight, in order to compensate for the inefficient angle of the tibia, relative to the direction of the sled’s pathway. Using 30% more weight, in order to get the same amount of quadriceps load as you get with a Leg Extension, simply loads the skeleton (the bones and joints) more, with no additional benefit to the quadriceps.

The Leg Extension exercise (using full range of motion) is more mechanically efficient because it provides more muscle load with less skeletal strain, as compared with a Leg Press (using full range of motion). Leg Extensions deliver the highest percentage of the load being used, to the quadriceps muscle. In addition, since Leg Extensions do not activate the hip extension muscles (i.e., the glutes, adductors and hamstrings), “reciprocal innervation” is not triggered—which means that quadriceps activation will not be neurologically compromised, as it is when doing Leg Presses*.

(* Note: This is fully explained in “The Physics of Resistance Exercise”, in Chapter 11)

The stated conclusion of the study was:  “This demonstrates that patients with patellofemoral joint arthritis may tolerate rehabilitation with the Leg Press exercise better than with the Leg Extension exercise in functional ranges of motion because of lower patellofemoral joint stresses.”

The “functional ranges of motion” to which they refer are the degrees of knee-bend that are between 0 degrees (straight knees) and 30 degrees. The term “functional” here suggests that the degree of knee bending that occurs when walking, stepping onto a curb, or walking up and down stairs accurately represents a person’s basic “functional” requirements. However, moving a knee or elbow no more than 30 degrees is extremely compromised for the purpose of muscle development. Imagine being restricted to bending your elbow joint only 30 degrees when working your biceps or your triceps. The normal range of motion for an elbow (or a knee) is approximately 110 degrees.

What this study actually declares, is that the least amount of knee “stress” occurs when there is the least amount of knee movement (almost none), and the least amount of load on the quadriceps, and that can be achieved by doing the 1/5 least challenging part of the range of motion of Leg Presses. Ironically, the same thing can be achieved by simply doing “mini-squats”– tiny knee bends from a standing position. You don’t even need a Leg Press machine.

This study also established that there is LESS knee stress during Leg Extensions (as compared with Leg Press), when the knees are more bent (60 degrees to 90 degrees). That is the range of motion where the quadriceps are working at their most productive / beneficial length (i.e., when muscles are more elongated).

To be clear, this study was conducted for the purpose of addressing the rehabilitation needs of people with patellofemoral knee arthritis. It was not done for the purpose of establishing which exercise is more safe (Leg Extensions or Leg Press) for people with perfectly healthy knees. The “take away” from this study should NOT be construed as “Leg Extensions are bad and Leg Presses are good”. Yet, this is how some people naively misinterpret this study.

An injured joint would obviously inhibit (interfere with) the successful loading of any target muscle, when attempting to use a resistance exercise that involves that particular joint. However, the concessions that need to be made by a person with an injured joint, are not required of a person who does not have that injury.

Again, this study does not suggest that all people should avoid doing Leg Extensions, nor does it suggest that Leg Presses are better than Leg Extensions for people with healthy knees. In fact, it states that “full range of motion” Leg Presses (i.e., going beyond 30 degrees of knee bend) produce more knee stress than “full range of motion” Leg Extensions. Thus, if an exercise should be avoided by a person with healthy knees, it would theoretically be Leg Presses–not Leg Extensions.

Evaluating the Efficiency of a Resistance Exercise

Determining “exercise efficiency” requires an understanding of
biomechanics — and also of your specific goal

Doug Brignole
January 2020

Exercise, by definition, is used as the “means to an end.” It is not meant to be the end unto itself. Exercise is typically done for the purpose of eliciting a specific developmental response. For that reason, we need to clearly identify the response we seek, so that we can select the exercises which are most likely to garner the physical adaptation we want.

Clearly identifying our goal is not enough, however. We also need to know what makes one exercise different from another—what makes an exercise “better than” or “worse than” another, in terms of its mechanics (physics) and its anatomical / physiological factors.

There are parameters by which all resistance exercises can be evaluated, just as there are parameters by which the flight worthiness of an airplane can be evaluated. Every resistance exercise has a bio-mechanical “profile”, which can be used to evaluate its mechanical efficiency and also its anatomical / neurological efficacy. Identifying the profile of an exercise allows you to determine whether performing that exercise is a wise choice for your specific goal.

Unfortunately, most people are unaware of the physics involved in resistance exercise, and of the anatomical and neurological factors involved in resistance exercise. Therefore, they cannot even begin to evaluate how “efficient” an exercise is (energy cost versus degree of muscle loading), nor how anatomically correct the movement is (range of motion, alignment, joint integrity, etc.), nor whether or not there are any neurological compromises.

Further, most people do not realize that the following two goals are polar opposite:

  • Optimizing muscle development with optimum safety
  • Lifting the most amount of weight

People usually assume that there is a direct correlation between the amount of weight being lifted, and the amount of load a particular muscle receives—but this is absolutely false. How much load a muscle receives involves a number of mechanical variables, only one of which is the actual weight being used. The other variables include the length of the limb a person is using, the angle of that limb relative to the direction of resistance, and the angle at which the target muscle pulls on that limb.

Consider the following comparison of two exercises, in terms of how much triceps loading each exercise produces.

In the image below, I am demonstrating the descended position of “Parallel Bar Dips”. Keep in mind that the forearm is the “operating lever” of the triceps.

All levers (i.e., limbs) are “neutral” when they are parallel with gravity (vertical)—and any muscle operating a neutral lever is not loaded (receives zero percent of the available resistance). All levers that are perpendicular with gravity (horizontal) are “maximally active”—and any muscle operating a horizontal limb is fully loaded (receives 100% of the available resistance).

If we were to quantify (in a simplified way) the amount of resistance with which each of my triceps is loaded, we would first acknowledge the gross amount of weight being moved—which is bodyweight, in this case. Then, we factor in the length of the forearm. This is a standard ratio typically used in all lever-load calculations: the distance between the pivot point (the elbow) and where the force is being applied (the triceps attachment on the ulna / forearm) – relative to – the distance between the where the muscle force is being applied (the triceps attachment) and where the resistance is applied (the hand / distal end of the forearm). Then, we factor in the “moment arm”, which can be easily identified (in this case) by the angle of the forearm, relative to gravity. The more horizontal the tilt of the forearm lever, the greater the moment arm—the greater the moment arm, the greater the percentage of load magnification.

So, in this example, we have a bodyweight of 200 pounds, divided by two arms—therefore, 100 pounds per arm. We then multiply that by a factor of 12 (i.e., the approximate length of the forearm, a 12 to 1 ratio). Then we multiply that by the moment arm (approximately 11% – representing the degree of forearm tilt, between vertical and horizontal).

100 (lbs.) x 12 (forearm length multiple) x 11% = 132 pounds

Therefore, this exercise above (Parallel Bar Dips) would load each of the triceps with approximately 132 pounds of resistance. The “cost” of this is 200 pounds of effort (bodyweight).

As a comparison, let us now calculate the load on each triceps, when performing a “Supine Dumbbell Triceps Extension”—shown below—using a pair of 20 pound dumbbells.

Using the same formula, we’ll first factor in the gross weight being used—20 pounds per arm (40 pounds total). Then, we’ll factor in the same forearm length—a multiple of 12. Then, we’ll factor in the “moment arm” (i.e., the angle of the forearm, relative to gravity). In this case, you can see it’s mostly horizontal, rather than mostly vertical. When the forearm reaches the fully perpendicular position, relative to the straight downward pull of gravity (as shown above), the forearm is at the 100% “active” position. Again, let’s do the math, plugging in our new variables.

20(lbs.) x 12 (forearm length multiple) x 100% = 240 pounds

This exercise above (Supine Dumbbell Triceps Extensions), therefore, loads each triceps with 240 pounds of resistance and the “cost” of this is only 40 pounds total of effort. This is almost twice as much benefit (triceps load), with about 80% less effort. Who wouldn’t want that kind of return on their investment?

Clearly, you can see the superior efficiency of allowing the operating limb of a target muscle (i.e., in this case, the forearm) to interact perpendicularly with gravity (the direction of resistance), as opposed to using a limb that is mostly parallel with the gravity (vertical), when the goal is maximum muscle loading with minimum wasted effort and minimum injury risk.

The triceps muscle could never be loaded as much—during Parallel Bar Dips—as it could be by doing Supine Dumbbell Triceps Extensions. In order to get 240 pounds of triceps load while doing “Dips”, an additional 160 pounds would have to be added to the body (80 pounds per arm). But that would drastically increase the risk of injuring the shoulder and elbow joints, plus also the anterior deltoids, and most people wouldn’t be able to do it.

The compound exercise zealot would argue that “Parallel Bar Dips is still worth doing because it works other muscles at the same time”. However, the other two muscles that participate during Parallel Bar Dips—the pectorals and the anterior deltoids—are not worked as well as they could be worked, during isolation exercises.

Participation of the pectorals is compromised (during “Dips”) because the humeral movement that occurs is not the ideal pathway for the pectorals. The ideal pectoral movement would cause maximum elongation and maximum contraction (shortening) of the pectoral fibers, and that requires the arms (i.e., the humerus / upper arm bone) to move laterally (sideways)—away from the sternum, and then medially—toward the sternum, the midline of the body.
During Parallel Bar Dips, the upper arms are not moved in that manner. Rather, they are moved alongside the torso, with almost no lateral or medial movement at all.

Participation of the anterior deltoids is also compromised, during Parallel Bar Dips, because they are over-stretched (in the descended position), yet forced to stop short of their full contraction (at the “top” of the movement). The anterior deltoids are also severely overloaded during “Dips”—strained far beyond their reasonable capacity—because they are the smallest and weakest of the three muscles involved.

So, from the standpoint of “muscle benefit”, and also from the standpoint of muscle and joint safety, each of the three muscles that participate during “Dips” would receive more benefit more safely, by doing three separate isolation exercises. It would thus be foolish to select Parallel Bar Dips—“because it’s a compound exercise”—while ignoring (or being unaware of) its significant bio-mechanical shortcomings, which compromise benefit and pose a high risk of injury.

Another exercise that is popular because it’s a compound exercise, and because it allows a person to move a large amount of weight, is Barbell Squats. Powerlifters typically use this exercise as one of their three competition lifts, but it’s also regarded as a “Holy Grail” by most bodybuilders and people who are pursuing general fitness (because they’ve been mislead).

In the illustration below-left, we see a person standing with a barbell on his upper back. In that figure, you can see that his lower leg (lever), his upper leg (lever) and his torso (also a lever) are all vertical, which means they are in the neutral position. Therefore, none of the muscles which operate those limbs (levers) are yet loaded.

(Note: Keep in mind that the lower leg “lever” is operated by the quadriceps muscle group—which is to say that the quadriceps extends the knee joint. The upper leg “lever” is operated by the muscles which produce “hip extension”—most notably the gluteus maximus. The gluteus is assisted in this action by two other muscle groups—the adductors and the hamstrings.)

In the illustration above-right, the person has descended to the “bottom” position of the Squat. Notice that the angle of the lower leg has tilted forward about 30 degrees from the neutral, vertical position. As such, it is far from loading the quadriceps with 100% of the available resistance, which would require that it be horizontal. It is far closer to the vertical (neutral) position, than it is to the horizontal (fully “active”) position—and this is the farthest that the the lower-leg-lever can tilt forward, during standard Barbell Squats.

The angle of the lower leg lever does not even reach 45 degrees, which would be halfway between vertical and horizontal. A 45 degree angle could loosely be called the “50% active” angle (i.e., half way between fully neutral and fully active). Because the lower leg only tilts forward approximately 30 degrees from the neutral position, the load reduction to the quadriceps is approximately 60%. This is absurd, given that one of the primary reasons for doing a Barbell Squat is quadriceps development.

To compensate for this drastic load reduction (i.e., the person senses that his quadriceps can handle more resistance), more weight is added to the barbell—but the added spinal compression increases the risk of spinal injury. It would be much wiser to increase the quadriceps load by using a more efficient lower leg lever—doing an exercise that allows the lower leg to be more perpendicular with gravity—rather than continuing to use an inefficient lower leg angle, but with more added weight.

The Compromised Mechanics of the Upper Leg Lever

The upper leg bone (the “femur”), which causes most of the downward thrust during Squats, is powered by the gluteus and a couple of other hip-extension muscles. The femur (lever) does reach the horizontal position—perpendicular with gravity—and that is theoretically good. However, notice that the lower leg lever, which is the “secondary lever” to the upper leg lever, doubles under the upper leg. This causes a “length reduction” of the femur (actually, of its moment arm), such that its magnification is effectively reduced by nearly 40%. So, instead of the gluteus being loaded with the magnification of a 17 inch lever (femur), it is only loaded with the magnification of a 10 inch lever (approximately).

Again, in order to compensate for this load reduction the person who is Squatting must add more weight onto the barbell, which further increases spinal compression. Yes, adding more weight to the barbell does increase the net load to the quadriceps and to the gluteus, but it does so with the same compromised “cost / benefit” ratio—the same percentage of load reductions. Thus, the skeleton is burdened with a tremendous amount of additional load, which could be avoided simply by using more efficient exercises.

It would be much more efficient to use two separate (mechanically “better”) exercises—one for the knee extensors (i.e., the quadriceps) and one for the hip extensors (i.e., the gluteus, adductors and hamstrings). Using isolation exercises for each of these two muscle groups would allow each to be loaded as much or more—while using less weight—thereby also minimizing wasted effort and unnecessary skeletal strain.

It would thus be foolish to hail the Barbell Squat as a “good” exercise for the legs simply because it’s a “compound exercise”, while ignoring (or being unaware of) the significant mechanical inefficiencies of the movement.

Neurological Interference

Compromised mechanics is not the only problem associated with Barbell Squats. There is also a neurological conflict, and it relates directly to the fact that Squats is a compound (multi-joint) movement.

The human body utilizes a neurological function called “reciprocal innervation”, which prevents opposing muscles from contracting simultaneously, with full force. This is meant to prevent self-defeating efforts. It is an integral part of the body’s ability to function with coordination.

For example, when you hold a weight in your hand with your elbows bent—as shown below—your biceps are loaded. Loading the biceps triggers the CNS (central nervous system) to send a “relaxation synapse” to the triceps, which is the “antagonist” muscle (i.e., the opposing muscle). This relaxation synapse prevents the triceps from simultaneously contracting with much (if any) force, thus preventing any interference which would conflict with the biceps’ optimal effort.

Whenever we load any skeletal muscle, the CNS (central nervous system) automatically causes a relaxation of the opposing muscle(s). This happens involuntarily—without us even being aware it. We cannot prevent it from happening, even if we wanted to do so. It’s automatic, and “autonomic”—part of the “autonomic nervous system”.

When we perform Barbell Squats, we unknowingly cause the relaxation (partial shut down) of some of the muscles we are actually intending to engage, because opposing muscles are trying to work simultaneously.

Typically, when we do Barbell Squats, our intention is to optimally stimulate the knee extension muscles (i.e., the quadriceps) as well as the hip extension muscles (i.e., the glutes, adductors and hamstrings). However, the simultaneous engagement of those two joint functions causes some of those muscles to be compromised (to shut off), due to reciprocal innervation.

Hip extension is opposite hip flexion, so activation and loading of the hip extension muscles (i.e., the glutes, adductors and hamstrings) automatically causes a relaxation of the hip flexion muscles. One of the hip flexors is the rectus femoris, which is also a primary quadriceps muscle (i.e., the rectus femoris has two joint functions because it crosses the knee joint as well as the hip joint). The rectus femoris cannot fully engage while the gluteus is loaded. This has been demonstrated in EMG studies.

Needless to say, it’s more than a little ludicrous to do an exercise that causes nearly one quarter of your target muscle to be de-activated, precisely when you are struggling so hard to optimize its stimulation.

More Neurological Interference

Knee extension (which is caused by the quadriceps) is opposite knee flexion (which is caused by the hamstrings). These two muscle groups cannot be optimally engaged at the same time, because they are antagonist muscles. Reciprocal innervation prevents their simultaneous activation.

However, the hamstrings muscle group also has a secondary joint function, because it crosses the hip joint as well as the knee joint. The secondary function of the hamstrings is assisting in hip extension. During Barbell Squats, all hip extension muscles are “called to action” to participate in that part of the movement. However, the hamstrings are conflicted during Squats due to the simultaneous activation of the quadriceps. So, during Barbell Squats, “hip extension” becomes partly impaired, because the hamstrings are mostly deactivated.

It’s also likely that the “desire” (the effort) on the part of the hamstrings to participate in hip extension, causes a degree of deactivation to occur in the quadriceps ….because the quadriceps oppose the hamstrings. Compromised activation of the quadriceps (all four parts) is certainly not what you want when you’re doing heavy Barbell Squats, with the intention of developing the quads.

It would be far better to work the quadriceps without the neurological interference that is caused by simultaneous hip extension. And it would be far better to work the gluteus without the neurological interference that is caused by simultaneous knee extension. In fact, it’s far more productive to do isolation exercises for each of those two primary joint functions (knee extension and hip extension), and doing so has the added benefit of not needing to load and compress the spine at all.

It would be foolish to believe that a Barbell Squat is a “great” because it’s a compound exercise that theoretically “saves time” and “burns more calories”, while ignoring (or being unaware of) the significant mechanical and neurological inefficiencies that occur during Barbell Squats.

Squats do not save time because the benefit each participating muscle receives is significantly less than could be achieved using more mechanically efficient exercises.

Squats do indeed burn a significant amount calories. However, the same amount of calories could be spent simply walking on a treadmill with an incline of 20% and a speed of 6.0 mph—without the tremendous skeletal load on the spine, knees and hip joints.

Interestingly, the fact that Squats are “difficult” is often what compels people to mistakenly assume that Squats are good—“because they’re so tough”. Many people live by the “No Pain, No Gain” mantra, as if anything that is tough must have a commensurate amount of reward—but that is not necessarily true. In the case of Barbell Squats, as well as many other compound exercises, the reward does not match the amount of effort required, due to mechanical inefficiencies and neurological conflicts.

Squats are safe enough when performed without a heavy barbell placed on the spine, and they are productive for the purpose of cardiovascular fitness, improved endurance, body fat reduction, and general lower body conditioning. Adding weight, however, significantly increases the cost without an equal increase in benefit.

Barbell Squats can also be useful for enhancement of sports performance—injury risk be damned—when a particular sport requires movements that are very similar to Squats. But proprioceptive training (practicing a skill with added resistance) for a specific sport is considerably different than training for improved physical fitness and muscular development.

Selecting the Most Efficient Exercises for
Fitness and Muscle Development

The following ten bio-mechanical criteria allow us to determine and select the “best” (most efficient, most productive and most safe) resistance exercises for the purpose of fitness and muscle development.

Not utilizing these criteria, or employing the services of a trainer who does not fully understands these criteria, essentially guarantees that your efforts will be largely inefficient (costing more effort and energy than necessary), that your results will be compromised (less productive), and that you will be exposed to a higher risk of injury than would occur if you used a better exercise selection.

  1. The exercise uses mostly “active” limb angles (i.e., perpendicular with resistance), instead of mostly “neutral” limb angles (i.e., parallel with resistance).
  2. The exercise uses mostly full lever (limb) lengths—without unnecessary “moment arm” reductions caused by a secondary lever (limb).
  3. The exercise allows the person to move the limb (operated by the target muscle) directly toward the target muscle’s origin during concentric contraction, and then directly away from the target muscle’s origin during eccentric contraction.
  4. The exercise allows dynamic muscle contraction (i.e., exercise with movement) and full range of motion—avoiding static / isometric contraction (exercise without movement), as well as incomplete range of motion.
  5. The exercise allows proper alignment (i.e., the direction of movement, the direction of resistance, and the origin and insertion of the target muscle, are all on the same plane). This also ensures “opposite posi tion loading”—that the target muscle is positioned directly opposite the direction of resistance, thereby getting its full load.
  6. The exercise avoids reciprocal innervation—the attempted engagement
    of two opposing muscles. Unintended reciprocal innervation occurs
    when a compound exercise requires the simultaneous movement of
    two joints (each with a different function), and the associated engagement of one or more opposing, two-jointed muscles. In such cases,
    the two separate joint functions compete for muscle innervation, which compromises muscle activation (i.e., weakness) of target muscles.
  7. The exercise avoids “active insufficiency”, which is the over-shortening
    of a two-jointed muscle, upon its contraction. It also avoids “passive insufficiency”, which is the over-stretching of a target muscle’s antagonist (opposing) muscle.
  8. When practical, the exercise avoids “bi-lateral deficit”, which is the neurologically caused strength compromise that often occurs when both (left and right) side limbs work simultaneously.
  9. When applicable, the exercise avoids the simultaneous occurrence of “mechanical disadvantage” (when a muscle that flexes a joint, pulls on its corresponding limb from a fully parallel angle), and a fully “active” limb—a limb / lever that is fully perpendicular with resistance. These two “magnifiers” of resistance, occurring simultaneously, result in a drastic increase of resistance, which significantly increases the risk of tendon rupture.
  10. The exercise utilizes an optimally beneficial resistance curve—providing more resistance during the early part of the range of motion (when the muscle is more elongated and stronger), and less resistance during the latter part of the range of motion (when the muscle is more contracted and less strong). This allows the “resistance curve” of the exercise to match the “strength curve” of the skeletal muscle, which is optimally productive.

This may seem very complicated, and it obviously requires more explanation before the significant benefit of following these criteria—when selecting exercises—can be understood. But it demonstrates that there’s much more that determines what makes an exercise “good” than simply how much weight “can” be lifted, what dogma dictates, and what the current trends encourage.

There is a plethora of misinformation in the fitness industry, due to its over-commercialization. This has lead many people to mistakenly believe that certain exercises are “gold”, when often they are not.

“Compound exercises”—like Barbell Squats, Parallel Bar Dips and Deadlifts”—have been hailed as essential and foundational, but they are compromised for a variety of reasons. Other exercises are often better.

We need to look carefully at the bio-mechanical profile of each exercise, in order to properly evaluate them. Only then can we appropriately select the most efficient exercises for our intended goals.