Misunderstanding (femoral) adduction

Posted on January 28, 2015


Prefatory note

Depending on your background, this may be one of those posts you have to slog your way through. (I know because writing it made me rethink how I felt about myself.) To some degree, it’s supposed to be. If you get lost at times, don’t feel bad. Especially if you’re lacking anatomy vernacular, there will be some tough spots. Again though, there should be. Stick with it, and you’ll see why.

Somebody performs a squat and this happens:

Squat knees caving in

“We need to stretch those adductors!”

Hip abduction adduction

Fancy word for groin muscles.

Adductor muscles. Fancy word for groin muscles.

Off you go performing exercises like this:

Adductor stretch

Superficially, the logic here makes sense. If something like the knee(s) come together during a squat, adduction, then these muscles are too tight / stiff / overactive (whatever you want to say -we’ll come back to this), so, we should stretch them. Or get them to relax. Again, whatever you want to call it. Concurrently, we should work the leg into abduction -the muscles which “aren’t working enough.”

Anatomy ain’t rocket science, but it ain’t that simple either

Elon Musk Rockets are Tricky

Anatomy can be tricky too. Let’s show why more thinking throws a wrench in this stretch the adductors logic. Again, that squat:

bad squat : standing form

Because the femurs are adducting and internally rotating, the other way to view this squat is, “We need to stretch your internal rotators!” Or, “We need to work your external rotators!”


This usually then goes down the line of “We need to work your glutes!” As the glutes are a good external rotator of the hip. Although, that depends. Some parts of the glute can internally rotate the hip!

Sometimes you’ll also hear to work those deep external rotators, the ones nobody can ever remember the names of.

Deep Hip External Rotators

What are some other lateral rotators of the femur though?

  • Adductor brevis
  • Adductor longus
  • Adductor magnus

In case you’re lost, let’s recap all the above. Instead of this,

Squat knees better

This happens:

Squat knees caving in

And you could say:

“The femurs are adducting! We need to relax the adductors of the thigh.”

You could also say:

“The femurs are internally rotating! We need to work the external rotators of the thigh…Of which the adductors are included…”

What seemed so obvious on the surface becomes a paradox when looking more in depth. At first, it makes sense to relax the adductors. When we look specifically at what the adductors do, it makes sense to work them…and relax them? Or?

In fact, what else is happening during a knees-caving-in-squat?

Squat knees caving in with opposing rotations

The knees -femurs- are going too far in, but the feet -tibiae- are going too far out. The tibiae, especially once the feet turn out and particularly in relation to the femur, are externally rotating. Meaning one could go “We need to work the tibial internal rotators!” And what’s an internal rotator of the tibia? The gracilis (especially when the knee is bent, like in a squat)…which is an adductor.

Keeping everything else fixed; showing the gracilis as a medial rotator of the tibia.

Keeping everything else fixed; showing the gracilis as a medial rotator of the tibia. (Blue arrow.)

Who’s right? Who’s wrong? Where can all the lunatic Youtube commenters go to rejoice in their disdain of human flaws?

Enough anatomy for now, let’s watch some movement

I know eyes can gloss over and minds can numb when so much anatomy speak is used. Let’s look at some common movements, and typical compensatory strategies.

-Hands and Knees; Leg Extension -colloquially referred to as a Bird Dog:

Hands and Knees Leg Extension

-A Side Lying Leg Lift:

Side Lying Leg Lift Good Abduction

-On Back (elevated); Hip Extension -sometimes referred to as a Thomas Test:

tight it band causing knee pain

Tangent: You are a pain in the ass by naming a test or movement after yourself. It makes it harder for people to remember what the test or movement is. Think the difference between Bird Dog versus Hands and Knees; Leg Extension. Or Thomas Test versus Supine (on back) Hip Extension. The latter names tell you what the movement is. It’s in the name. This is how most physicists and engineers name things, and it’s for a reason: It makes life easier. It’s not as ego satisfying, but utility is more important here.

Now, some common compensatory strategies for these three movements.

-In a Quadruped Leg Extension, it’s common to see the leg flare out to the side and or see some opposite rotations going on at the knee:

Craig Leg Extension Abduction with lines

Keenan M Bird Dog Line

-In a Side Lying Leg Lift, it’s common for the femur to internally rotate during abduction:

Side Lying Leg Lift Knee Turning in during abduction with lines

-It’s also common for the leg to not be able to come down -adduct- very far. (I discuss this thoroughly here.)

PGM Test Set Up

When setting the leg up in position…

posterior gluteus medius strength test

…it will want to stay held in abduction.

Note the above is true when the leg is held in extension. If the hip is allowed to flex enough -the TFL often wants to pull things back into flexion, or abduction, to avoid being stretched- then adduction will be no problem. While the adductors do also have a role in hip flexion, we know this is not due to the adductors. I’ll cover why in a few minutes.

-In Supine Hip Extension, it’s common for the leg to either flare out:

Supine Hip Extension Abducted Thigh (Me) with two lines

-And or the tibia to rotate / move laterally in relation to the femur:

Supine Hip Extension Tibia Laterally Rotated with lines

I want to be very clear here. The people whose knees cave in during a squat are the people most likely to engage in the above patterns. (These are extremely common patterns overall.) Hopefully this isn’t tough to conceptualize. If we reorient these images, notice how similar they all look to one another.

Various Internal Rotations Side by Side

While I only put green lines to show where the tibiae would ideally be, in all the above, ideally the knee (femur) would be turned out -laterally rotated NOT abducted- some as well.

This is crucial here, because in these compensatory strategies:

-In the Quadruped Leg Extension:

–The femur is extending and ABducting.

Craig Leg Extension Abduction with lines

–In this one it’s extension and internal rotation:

Keenan M Bird Dog Line

-In the Side Lying Leg Lift:

–When the femur abducts and internally rotates, the ADDuctors are being stretched. (Remember, the adductors externally rotate as well as adduct.)

Side Lying Leg Lift Knee Turning in during abduction with lines

–When the leg is held in ABduction, it shows the ABductors are stiff. The abductors are pulling more than the adductors.

posterior gluteus medius strength test

-In Supine Hip Extension:

–When the leg flares out, the ABductors are responsible. The ABductors are illustrating they’re more stiff than the adductors.

Supine Hip Extension Abducted Thigh (Me) with two lines

–When the femur internally rotates, the ADDuctors are being stretched.

(The main abductor doing all the above: pulling the leg out / internally rotating the femur / laterally pulling the tibia is the Tensor Fascia Latae.)

Do you see what has happened here? All that talk about stretching, loosening, relaxing the adductors…yet in all the scenarios above the ABductors are what need to be relaxed. If anything, in the above scenarios, the ADDuctors need to be working more! They need to be working more as adductors and external rotators.

Because if the adductors were truly overactive, in the Quadruped Leg Extension we’d expect this to happen:

The leg to adduct (orange) and externally rotate (not drawn).

The leg to adduct (orange) and externally rotate (not drawn).

Bird dog with three lines

In the Side Lying Leg Lift, we’d expect the leg to either be pulled more into adduction relative to abduction, and or the leg to want to externally rotate during abduction:

Side Lying Leg Lift Good Abduction with green line

The adductors get stretched during abduction, so one way for them to prevent this stretching would be to externally rotate.

In the Supine Hip Extension we’d expect this to happen (orange):

Supine hip extension with three lines

The difference between a squat and the above movements is in a squat the foot is fixed -it’s on the floor- and, for the most part, can’t move. In the Quadruped Leg Extension, Side Lying Leg Lift, and Supine Hip Extension, the foot can move, which gives us a more accurate representation of what’s going on.

Because what’s happening in a squat is the appearance of dominant adductors. But what’s really happening is dominant internal rotators, which cause the adductors to come along for the ride, due to the fixed foot. (Other things can be causative here as well, such as a lack of dorsiflexion.)

That is, when you internally rotate the femurs on fixed feet, the only way to generate significant internal rotation is for the femurs to also adduct, but that doesn’t mean the adductors are responsible. Again, the adductors can play a role in external rotation. If they were so dominant we’d expect the femur to externally rotate and adduct.

-> You could argue here the adductors are so dominant into adduction that the internal rotation is what’s actually coming along for the ride (because the foot is fixed). However, once the foot is not fixed, we clearly see this isn’t the case. On balance, people are dominant into abduction and internal rotation, not adduction and external rotation. It’s just that in a squat these things are mutually exclusive. You can’t adduct and externally rotate, nor can you abduct and internally rotate. On balance, internal rotation is what usually wins in a faulty squat.

Understanding the adductors and external rotation

Let’s back up a second. I have a feeling hearing the adductors play a role in external rotation is new for some, or something that’s tough to visualize.

In the interest of accuracy, let’s remember the adductors are not one muscle. And, that all adductors are not the same. Let’s look more closely at the adductor longus.

The muscle starts at the superior ramus, and ends at the medial lip of the linea aspera of the femur.

Image credit: www.studyblue.com

Linea Aspera in blue, on right. Image credit: http://www.studyblue.com

The insertion is on the posterior side of the femur. Look at the line of pull. It’s towards adduction and external rotation, because of its posterior insertion. When you pull the posterior aspect of the femur medially, the superior aspect of the femur rotates laterally.

Adductor Longus GIF

Notice the femur rotating laterally. (Use the patella as a reference if needed.)

In case the animation is still tough to see in your own mind, I constructed a simple model of the hip and adductor longus.

Adductor longus simple model

Here are two slo-motion videos, pulling on the rubber bands to represent the adductor longus contracting. Top view and bottom view:

Note the similarity in this line of pull compared to what we’re trying to do in the Supine Hip Extension exercise:

Supine Hip Extension Abducted Thigh (Me) with two lines and arrow

Adductor Longus GIF

To reiterate, if the adductors were stiff, we’d expect when the leg is flared (above) for the hip to go into hip flexion, because the adductors can play a role in hip flexion. That is, when the leg is abducted the adductors are put on stretch, and one way to avoid being fully put on stretch would be to flex the hip (or externally rotate the leg). But this is not what happens.

Instead, when the leg abducts, it then falls into some hip extension, further stretching the adductors. When the leg is adducted though, it then goes into hip flexion (in someone who is stiff in the TFL -an abductor and hip flexor). Something like this:

it band syndrome stretch

Notice how high thigh is off bench when leg is in rightful adducted position. (Hip is held in flexion.)

Illustrating the TFL is what’s causing the hip to flex, not the adductors. Not to mention, in these scenarios, you typically see the femur internally rotate or the tibia externally rotate, both things the TFL has a role with. You pretty much never see the femur externally rotate and the tibia internally rotate -both things the adductors can play a role in. (I’m ignoring the long head of the adductor magnus right now. I’ll get there in a second.)

If I had one do a Quadruped Leg Extension, but we fully insure the lower back does not go into extension -we avoid anterior pelvic tilt / hip flexion; we take the hip flexion compensation out of the equation. The leg will then be much, much more prone to abducting than adducting. Illustrating the abductors are stiff relative to the adductors. Again, this:

Craig Leg Extension Abduction with linesThe adductors, particularly into adduction, are not stiff. The abductors are.


Why this matters

We now know, more often than not, we want to be avoiding focusing on abduction exercises. (And internal rotation.) In a squat, abduction is ok -it can help us externally rotate- but we don’t need to go crazy with it. You only cue someone “knees out” to generate the effect of their knees not turning in. There is no need for say, the knees to be outside the feet.

Really, we’re cueing the avoidance of internal rotation. We are NOT cueing abduction. It just happens to be that telling someone who is prone to internally rotating their femurs to “open your knees” often places their femurs in neutral during a squat. (That said, you can often get by with telling someone to “Not let the knees come together.”) This is part of the art of cueing. How your words are interpreted don’t always coincide with their literal meaning.

Beyond a squat though, such as non-feet-fixed exercises, there is no need to be going crazy with abduction. In fact, more often than not, we want to be avoiding it.

When we do a Quadruped Leg Extension, we want the leg going straight back, not out to the side. When we do Supine Hip Extension, we want the leg going straight down, not out flaring out. When stretching the hip flexors / quadriceps in a more conventional manner, we want the knee pointed straight down, not abducted. The green lines, not the red lines.

Abducted Side by side by side

There’s no need to be doing something like a split stance adductor mobilization:

Split Stance Adductor Stretch

In this position we’re significantly abducting the leg. So much so, barring the ballet type, not a single person is going to be able to do this without moving their lower back.

Split Stance Adductor with arrow 2

Notice one side of back higher than other. (Lumbar spine is rotated to right side to compensate for lack of hip mobility.)

Split Stance Adductor with hip to shoulder lines

Hips having to rock asymmetrically.

Split Stance Adductor with line 4

Split Stance Adductor Stretch with line 3

Split Stance Adductor with line 2

Split Stance Adductor with line

Split Stance Adductor with arrow

Split Stance Adductor lower back with circle

You only have about 45 degrees of abduction to play with. Beyond that, the only way to get the motion is to do something like rotate your lower back. And of course you feel a stretch in your groin when doing the above. You’re at your absolute maximum abduction range of motion. That doesn’t mean it’s a good thing to be doing though.

I randomly found a bunch of videos for “Split Stance Adductor Mobilization.” Video after video mentions keeping the lumbar spine in neutral. But neutral is not just a extension / flexion matter. It’s a rotational one too. As all the photos above depict, nobody is in neutral! But this isn’t the person’s fault, it’s the exercise’s.

There are more issues with this. Looking at how gravity is working here, where is the most vulnerable location in this position? The knee.

Split Stance Adductor Stretch Split Stance Adductor Stretch with medial displacement line Split Stance Adductor Stretch with medial displacement line and gravity

Gravity can pull down at the knee in a medial fashion. Where the distal femur and proximal tibia are both moving medially. This is what we’re trying to avoid in a squat.

Squat knees caving in with medial displacement lines

Split Stance Adductor Stretch with medial displacement line and gravity

Squat knees caving in with medial displacement lines

Notice the feet in the squat and split stance adductor mobilization are both over-pronating too!

To those familiar with the split stance exercise, yes, you could do something like turn the foot up to (somewhat) help with the knee and foot issue:

Split Stance Adductor foot up

But then you’re still dealing with the fact the leg is more abducted than 99% of people can achieve, -trying to achieve this is actually a good way to pinch the labrum- the fact one probably has stiff abductors (primarily TFL), and the tibia is still lateral to the femur.

Split Stance Adductor foot up with lines

We haven’t even talked about timing yet

If knees turn in at the transition stage of a squat -where you are moving from down to up- there is a good chance something like adductor magnus’ long head is contributing to this. Remember how the adductor magnus is a lateral rotator of the femur? I don’t see anything wrong with saying the adductor magnus is also an internal rotator of the femur.

The short, horizontal fibers, connect to the femur much like the adductor longus and brevis.

Adductor Magnus front view with adductor circle

Anterior view; adductor magnus connecting posterior to femur, but anterior to pelvis. (Dotted line versus solid line.)

adductor magnus with adductor circle

But the long fibers, starting on the back of the pelvis, connect distally on the femur, and medially. This is the hamstring portion of the adductor magnus.

Adductor Magnus front view with hamstring circle

Anterior view; adductor magnus long head starting at posterior side of pelvis; connecting to posterior and medial aspect of femur.

adductor magnus with hamstring type circle

We know something like the medial hamstrings, which connect just below the adductor magnus, play a role with internal tibial rotation. Much like the gracilis muscle we went over earlier.

It makes sense then that the adductor magnus long head, which has the same line of pull as the medial hamstrings, but connects at the femur, plays a role with femoral internal rotation. And when is this aspect of the adductor magnus needing to contract? At the bottom of a squat, when we engage the hips into extension.

This is in contrast to someone who immediately upon bending their knees allows them to cave in. That is, the person who engages say, the adductor magnus upon transitioning from down to up, is using the adductor magnus in a faulty way concentrically. Whereas the person who immediately brings the knees together may be using the adductor magnus in a faulty way eccentrically. 

So, even if your adductors truly were stiff / short / whatever, how does stretching them help with our timing issue? It doesn’t. Because it doesn’t address the brain’s role in all this. If you take all the muscles which turn the knees in during a squat, and they were as flexible as possible, that doesn’t mean your knees won’t turn in when you squat.

Stop the madness

The point of all this is not to take some highly technical path to a deduction of “We need to do this one stretch, for this one muscle.” The point is to illustrate the downsides to this type of thinking.

The point of all this is to illustrate when someone gives you a stretch like the split stance adductor mobilization, or a side lunge, or tells you something like “stretch your adductors” -a group of muscles which don’t all have the same function, all because your knees are caving in, that this advice is flawed. It’s a result of lazy thinking. (We’ve all been there.) Of a malformed philosophy. I understand this type of thinking has been BERATED into people, and it’s tough to come out of.

But think of how nice it is to be able to say our main takeaway from all this is, if you’re someone whose knees turn in when you squat, and you want to correct that…then start with not letting your knees come together when you squat.

Enter your email address to follow this blog and receive notifications of new posts by email.