Glute stretches are among the most performed and least examined positions in the whole business. The glutes get stretched a great deal and are rarely the thing that ran out. A figure-four or a pigeon puts the posterior hip capsule, the deep external rotators and the glute under load together, and which of the three you are feeling is not something you can work out from the inside.
That is not an argument against doing them. It is an argument for describing them accurately, which almost nobody does.
The short answer
- A deep buttock stretch loads at least three structures at once and you cannot separate them by feel.
- The glutes are rarely the limit in a stiff hip. Bone shape, the capsule and the hip flexors are more often the answer.
- A pinch at the front of the hip in a deep figure-four is the joint, not the muscle. Ease out of it.
- Only two of Limber's 56 moves are filed to the glutes as a primary area.
- There is no glute test in the battery. The glutes score is fed entirely by the deep squat, and the app calls it a proxy.
- Grade: thin. The dose evidence is borrowed from hamstring work; nothing tested these positions specifically.
Glute stretches, and what they load
The gluteus maximus runs from the back of the pelvis and the sacrum to the back of the femur. It extends the hip and turns it outward. To lengthen it you flex the hip and rotate it inward, which is roughly what a figure-four does.
The trouble is that the same movement also closes the back of the hip joint, loads the capsule at the rear of the socket, and puts the six small external rotators — piriformis among them — on stretch. Four structures, one sensation, no way to attribute it. Nobody can palpate the deep rotators through the gluteus maximus, and neither can you.
So the useful skill is not naming what you feel. It is noticing where you feel it:
- A broad pull deep in the buttock is the position doing what it was meant to do.
- A sharp pinch at the front of the hip or in the groin is the joint, not the muscle, and it does not soften with time in the position. Ease out and change the angle.
- Pain down the leg, past the knee is neither, and it is the point at which this page stops being useful.
Frank and colleagues, reviewing imaging of people without symptoms in Arthroscopy in 2015, found cam-type hip morphology in roughly a third of asymptomatic volunteers and over half of asymptomatic athletes. Extra bone at the head-neck junction of the femur blocks deep flexion mechanically. In a deep figure-four, that block is precisely what the front-of-hip pinch is, and holding harder against it achieves nothing.
Four positions, honestly labelled
Here is what Limber carries in this territory, with what each one actually loads.
| Move | Default | What it loads |
|---|---|---|
| M004 Figure-4 Glute Stretch | Hold 40 s, per side | Glutes, then hips |
| M013 Pigeon Stretch | Hold 45 s, per side, level 2 | Hips and glutes together |
| M016 Supine Twist | Hold 45 s, per side | Upper back and lower back, glutes second |
| M024 Happy Baby | Hold 50 s | Hips, then lower back and glutes |
Read the third column, because it is the honest part. Only the first two are filed to the glutes as a primary area. The supine twist is a spine position that happens to catch the glute; happy baby is a hip position that happens to catch it. If you are doing all four believing you are doing four glute stretches, you are doing two.
The order in the catalogue is also a progression: M004 is level 1 and its stated progression is M013; M013 is level 2 and its stated regression is M004. If a pigeon is too much for the front knee, the figure-four is the same idea taken lying down with the load off the joint. That is the swap the app makes, not a judgement about you.
The pigeon carries one specific instruction worth repeating here: ease off at any pain on the outside of the front knee. The front shin is loaded across the joint in that position, and a knee complaining is a knee complaining, not a stretch deepening.
Two moves out of fifty-six
Limber's catalogue holds 56 moves across ten body areas. The primary-area counts run: hips 10, hamstrings 9, shoulders 9, upper back 9, calves and ankles 8, neck 7, hip flexors 3, lower back 3, glutes 2, wrists and forearms 1.
Two. It is the second-smallest block in the whole catalogue.
That is a deliberate reading of the evidence rather than an oversight. The glutes are rarely the limiting structure in a stiff lower body, they are hard to isolate, and the positions that load them load the hip joint at the same time — so most of the work lands in the hips block, which has ten. A catalogue that listed fifteen glute stretches would be describing the internet rather than the anatomy.
No glute test exists in the score
This is the part that belongs on our own page rather than in somebody else's review.
Limber scores body areas from named tests at published weights. The glutes area is fed by exactly one, and it is not a glute test.
| Area | Fed by | Weight |
|---|---|---|
| Glutes | Deep squat | 1.0, stated proxy |
| Hips | Deep squat | 0.5 |
| Hips | Hip flexor test | 0.5 |
The deep squat test reads four bands from a thirty-second hold: cannot reach parallel is 15, parallel with the heels lifted is 40, a full squat with the heels down is 70, and a squat you could rest in is 90. It sees ankles, knees, hips and lower back at once, and it cannot tell you which of them was the limit — never mind isolating a glute.
So the glutes figure in the Range Score is an inference from a squat, and the app labels it as one. There is no glute measurement in this battery, on the web or in the app, and nothing here will pretend the number is a reading when it is a proxy. It rolls into the posterior group with the hamstrings and the lower back, where a genuinely measured hamstring reading sits beside it.
What the evidence supports
Thin, and worth saying plainly.
On the positions themselves: nothing specific. No trial has compared a pigeon against a figure-four, or established what either does over six weeks. The hold lengths in Limber's catalogue — 40 to 50 seconds — are carried across from the hamstring dose work, where 30 seconds beat 15 and matched 60 over six weeks. That transfer is an assumption. It is a reasonable one and it is still an assumption.
On naming the structure: Hopayian and colleagues, in the European Spine Journal in 2010, reviewed the reported clinical features of piriformis syndrome and found them inconsistent across studies, with no agreed definition and no agreed test. Attaching a muscle's name to a deep buttock sensation is where writing about this area usually goes wrong, and piriformis stretches works through that in detail.
Grade: thin. Stretching the area is low-risk and often feels good. Everything beyond that — which structure lengthened, what it did for you, whether it prevented anything — is unsupported.
Deep buttock pain that travels below the knee, numbness, pins and needles, weakness in the leg or foot, pain that wakes you at night, or anything that followed a fall is a matter for a clinician who can examine you. Nothing here diagnoses anything and nothing here is treatment.
Where to start
Do two positions rather than five: a figure-four on each side, held 40 seconds, and a supine twist as a calm finish. Then spend the time you saved on the thing that is more likely to be your limit.
For a stiff hip, the direction most often missing is internal rotation, and the 90-90 position works both rotations in one setup. Hip mobility exercises covers the joint as a whole, and the rest of the cluster sits under hips and legs.
Take the deep squat reading before you start and again three weeks later. It will not isolate a glute — nothing can — but it is a fixed instrument, and two of your own readings three weeks apart are worth more than any sensation you can describe.
Questions
What is the best glute stretch?
There is no trial ranking them, so the honest answer is a figure-four for most people, because it loads the right direction with the least demand on the knee. A pigeon takes the same idea deeper and asks more of the front knee, which is why Limber files it a level higher and lists the figure-four as the way back down. Hold either for around 40 seconds.
Why do I feel a pinch at the front of my hip?
Because you have probably reached the joint rather than the muscle. Bone contact at the front of the hip is a hard pinch that does not soften with time in the position, and it appears in the same place every time you test it. Ease out, open the angle, and work below it — pushing into a bony end point achieves nothing and irritates the joint.
Are my glutes tight or is it something else?
You cannot tell from the inside, and neither can anyone else without examining you. A deep buttock stretch loads the hip capsule, the deep external rotators and the glute at the same time, and the sensation does not distinguish them. In a stiff hip, the glute is one of the less likely candidates — hip flexor length, capsule and bone shape are more often the answer.
Does Limber have a glute test?
No. The battery has five tests and none of them reads a glute. The glutes area score is fed entirely by the deep squat, which the app labels as a proxy rather than a measurement, because a squat sees ankles, knees, hips and the lower back together and cannot separate them.
How long should I hold a glute stretch?
Around 40 to 50 seconds, which is what the app's positions default to. That figure is borrowed from hamstring dose research, where 30 seconds beat 15 and matched 60 over six weeks — no equivalent trial has been run on the glutes. Treat it as a sensible default rather than an established number.
Can stretching my glutes help my back?
There is no good evidence that it does, and this site is not going to imply otherwise. A forward fold has only low validity as a measure of lumbar extensibility, which is a sign of how loosely the region is measured, and no trial has shown that a glute stretch changes a back complaint. Back pain that persists, radiates or comes with numbness or weakness needs a clinician who can examine you.