Scapular control is one of the most confidently explained ideas in shoulder training and one of the least well supported. The story runs: your blade sits wrong, that pinches something, and here are the movements that put it back. Almost every link in that chain has been examined, and most of them did not hold.
The work is still worth doing. The explanation is not worth repeating. This page puts the evidence first, because a reader who takes the practice without the evidence will come away believing something we cannot back.
The short answer
- No consistent relationship has been found between shoulder blade orientation and shoulder pain.
- Abnormal blade movement is treated by consensus as a physical sign, not a diagnosis.
- The test people use to justify this work — the pectoralis minor length test — has poor accuracy.
- Nothing in Limber's battery reads the blade. There is no scapular reading, and the app does not offer one.
- The work is still reasonable as general shoulder exercise. Four moves in the catalogue do it.
What the evidence says first
Four findings, printed before any practice, at full strength.
Ratcliffe and colleagues, in the British Journal of Sports Medicine in 2014, found no consistent relationship between scapular orientation and shoulder pain. They reviewed the studies comparing blade position in painful and painless shoulders and the results did not line up. That is the finding the whole "your blade sits wrong" argument depends on, and it is not there.
Hickey and colleagues, in the same journal in 2018, found that asymptomatic athletes with scapular dyskinesis had a modestly increased risk of later shoulder pain. That is a real prospective finding and it is worth stating accurately: wide uncertainty, small effect, and a long way from there to "your shoulder hurts because your blade moves wrong". It says something about a population's odds, not about your shoulder's cause.
Kibler and colleagues, in the same journal in 2013, published the scapular summit consensus: dyskinesis is a physical sign and an impairment, not a diagnosis in itself. The people who study this most closely say it names something they observe, not something that explains a symptom. Anyone using it as a diagnosis is going further than the consensus does.
Lewis and Valentine, in BMC Musculoskeletal Disorders in 2007, found the pectoralis minor length test has poor accuracy. So the measurement most often used to justify this work — a short pec minor tipping the blade forward — is itself unreliable. Borstad and Ludewig had described the relationship between that muscle's length and scapular kinematics in the Journal of Orthopaedic & Sports Physical Therapy in 2005; a real relationship in a laboratory does not make a bedside test accurate.
Grade: weak for scapular position as an explanation of symptoms; moderate for control work as part of general shoulder exercise. Both halves of that sentence are meant.
What scapular control means
Strip the claim back to the mechanics and something defensible remains.
The shoulder blade is not fixed to the skeleton at the back. It floats on the rib cage, held by muscle, and it has to move for the arm to travel. As the arm rises, the blade rotates upward and tilts backward so the socket stays underneath the arm bone. If it does not, the arm reaches the top of its available travel early and the rest of the movement comes from somewhere it was not asked for.
So control means the blade does that reliably under load, at speed, at the end of a set, on both sides. It does not mean the blade sits in a particular place at rest. Resting position varies enormously between people who have never had a sore shoulder in their lives, which is the practical shape of the Ratcliffe finding.
That distinction matters because it changes the work. Chasing a resting position leads to squeezing the blades together and holding them there, which is a posture instruction. Chasing control leads to moving the arm through its range slowly and deliberately while the blade does its job, which is exercise.
What the blade has to do
Three actions, and they happen together.
Upward rotation. The socket turns to face more upward as the arm rises, so the arm bone keeps a bed under it through the whole arc.
Posterior tilt. The bottom of the blade swings forward against the ribs and the top edge tips back, which clears space at the top of the shoulder as the arm passes overhead.
Controlled protraction and retraction. The blade slides around the rib cage as the arm reaches forward and back. Held rigidly, it stops the arm; left slack, it gives the arm nothing to push from.
The upper back sets the surface all three happen on. A thoracic spine held in flexion holds the blade forward before the arm has moved at all, which is why blade work and upper back work belong in the same session. Shoulder mobility exercises sorts the whole catalogue by which of those links each move loads.
Nothing in the battery reads it
Stated plainly, because this is a page about a thing we would like to sell you a measurement for.
There is no scapular reading in Limber's battery. Nothing in the app scores blade position and nothing scores blade movement. The five built tests read the posterior chain, ankle dorsiflexion, shoulder reach, hip flexor length and squat depth, and none of them is looking at your shoulder blade. There is no camera, so nothing observes you at all.
The nearest thing in the design would have been the wall angel, and it is not built — in the app either. It is test 7 of the seven, designed with four bands, and designed to supply 40 per cent of the shoulders area. With it missing, the weight drops out rather than counting as a zero, and the Apley scratch reach carries the shoulders area on its own. The Range Score names the test that is absent rather than renormalising quietly and calling the area surveyed. Wall angels sets out the four designed bands and why you cannot score them today.
The shoulder reach does move when the blade moves — a blade that will not glide closes the gap between your hands — but it reads three joints at once and cannot tell you which one was short. Treat it as a reading for the region, not for the blade.
The work that is worth doing
Four moves from the app's catalogue. None of them corrects anything, and all of them are reasonable shoulder exercise.
M031 Wall Slides. Back, hips and head against a wall, feet a step out, arms in a goalpost with the backs of the hands toward the wall. Slide overhead keeping the hands near the wall and the ribs down. Dynamic, 45 seconds, level 2, cleared for warm-ups. Ease off at any pinch in the front of the shoulder. This is the one that also shows you something: what leaves the wall first is a finding, even unscored.
M032 Prone Swimmer Sweep. Face down, arms overhead with the thumbs up, lift the hands an inch and sweep them out and down toward the hips, drawing the blades together as they travel low, then back overhead with control. Dynamic, 45 seconds, warm-up cleared. It loads upward rotation and retraction through a long range with no equipment beyond a mat.
M041 Kneeling Thoracic Extension. Hands well forward on a mat, hips stacked over the knees, chest sinking toward the floor. A 45-second hold. This is the surface work — the blade cannot tilt back over a rib cage that will not extend.
M036 Standing Shoulder CARs. Reach an arm forward at shoulder height and draw the largest slow circle you control, up, behind and down, turning the palm as it travels. Forty seconds a side. It is the closest thing here to a rehearsal of control: slow, deliberate, at the edge of the range you own.
Six to twelve repetitions on the dynamic moves, one per breath cycle, four seconds in and six out. Mild stretch, never sharp pain. And do the loaded work too — pressing, rowing, carrying — because control is a quality that only appears under load, and range you cannot hold is a number rather than an ability.
Pain that persists, wakes you at night, radiates into a limb, comes with numbness or weakness, or followed a fall or an injury is a matter for a clinician who can examine you. Nothing here diagnoses anything and nothing here is treatment.
How to judge it
Since nothing scores the blade, judge the work by the things you can actually record.
The unscored wall check. Set up against a wall, slide once, slowly, and note what left the wall first and how far the arms travelled before it did. Write it down with the date. Three weeks later, same setup, same tempo. More of the arc with contact held is progress; more repetitions is not.
The scored reading next door. The shoulder mobility test takes the Apley scratch — five bands, both sides, a 0–100 score off the app's own anchors, nothing stored and nothing sent. It will not isolate the blade, and it will move if the blade starts gliding, which is as close as this battery gets.
What you can do at the end of a set. Control is what survives fatigue. If the last two repetitions of a press look like the first two, that is the quality this page is about, and no home test measures it better than watching.
The rest of the cluster is shoulders and upper back, and rotator cuff mobility is written on the same footing as this page.
Questions
Is a winged or uneven shoulder blade a problem?
Usually not on its own. Blade position and movement vary widely among people with no symptoms, and the review evidence has found no consistent relationship between scapular orientation and shoulder pain. Abnormal blade movement is treated by expert consensus as a sign that may accompany a problem rather than as a diagnosis, so it is worth noticing and not worth alarm.
Does scapular dyskinesis predict shoulder pain?
Modestly, in athletes, with wide uncertainty and a small effect. A 2018 review of prospective studies found asymptomatic athletes with dyskinesis had a somewhat higher risk of later shoulder pain, which is a statement about a group's odds rather than about the cause of any one person's symptoms. It is a long way from that finding to the claim that a blade moving unusually is why a shoulder hurts.
Should I be squeezing my shoulder blades together?
Holding them squeezed is a posture instruction, not control, and the evidence for resting position as an explanation of symptoms is weak. What is worth training is the blade moving properly as the arm travels — upward rotation and backward tilt through the range, under load. Rigidity at rest gives the arm nothing to work with.
What exercise builds scapular control?
Slow, full-range arm movement with the trunk held still, plus the pressing, rowing and carrying that load the position. In the app's catalogue that is M031 Wall Slides, M032 Prone Swimmer Sweep and M036 Standing Shoulder CARs, with M041 Kneeling Thoracic Extension for the rib cage the blade sits on. None of that corrects anything; it is ordinary shoulder exercise with a sensible target.
Does Limber measure my shoulder blades?
No. There is no scapular reading in the battery, nothing scores blade position or movement, and there is no camera. The wall angel, which would have come nearest, is designed and not built in the app or here, so the Apley scratch reach carries the shoulders area alone.
Is a short pec minor tilting my blade forward?
Possibly, and the bedside test for it is not accurate enough to tell you. A 2007 study found the pectoralis minor length test has poor accuracy, so the measurement usually offered as proof of this idea cannot carry the weight put on it. Stretching the front of the chest is harmless and reasonable; presenting it as a correction of a measured fault is not.