Standing Band Shoulder Extension with External Rotation
The same set-up as the neutral version: two light loop bands on the door at hip height, the loops around your wrists, your back to the door. The difference is the rotation. Turn your arms out until your thumbs point outwards, let them move a little away from your sides, out to the side, and draw back from there. Sternum lifted, shoulder blades back and down, chin back. The bands are lighter than in the neutral version.
The only difference from the neutral version is the rotation — and it shifts who does the work. With your thumbs turned out, the shoulder rotators take on a larger share, and the shoulder blade tilts back more readily.
The closer your arms stay to your body, the more this is an exercise for the rotator cuff. The further they travel out to the side, the more the posterior deltoid takes over. This version sits between the two.
The bands here are lighter than in the neutral version. Turning outwards is the weaker part of the movement — with too much pull you take the force from the posterior deltoid and from twisting the forearm instead of from the shoulder. The inside of your elbow turns outwards with the arm, and that is how you know it is right.
From practice: I use both versions as a pair and let clients work out for themselves which one feels better. Those who feel a pull at the front of the shoulder in the neutral version usually get on better with this one. That is what I see in the studio.
Starting position
The set-up is the same as in the neutral version: two loop bands on the door at hip height, the loops around your wrists, hands open and relaxed, your back to the door, feet hip-width apart. The door is shut and cannot swing open.
Now turn your arms outwards until your thumbs point out. The inside of your elbow turns outwards with them — that is how you check that the rotation is really coming from the shoulder. Your arms move a little away from your sides as you do, further out to the side than in the neutral version. From there, draw back, lift your sternum, draw your shoulder blades back and down, and bring your chin back.
The bands here are lighter than in the neutral version. Turning outwards is the weaker part of the movement, and if the bands are too heavy you produce the force with the posterior deltoid and by twisting the forearm, instead of with the shoulder rotators.
If your shoulder is sore, practise the rotation only for now, without pulling back.
The three cues
- Same set-up, but lighter bands
- Turn your arms out until your thumbs point outwards
- The inside of your elbow turns outwards — that tells you the rotation comes from the shoulder
- Arms a little out to the side, then draw back
- Lift your sternum, shoulder blades back and down, chin back
Common faults
| Fault | How to fix it |
|---|---|
| The rotation comes from the forearm instead of the shoulder | Watch the inside of your elbow — it has to turn outwards with the arm. If only the hand turns out while the elbow stays put, the forearm is rotating. |
| Using the same bands as in the neutral version | Use lighter ones. Turning outwards is the weaker part of the movement — with too much pull the posterior deltoid takes over. |
| The arms travel too far from the body | A little further than in the neutral version is enough. The further out they go, the more this becomes an exercise for the posterior deltoid. |
| The arms go too far back | Stop as soon as your ribcage tips forward. Going further back works against the exercise. |
| The chest collapses | Lift your sternum before the arms start. Upright, the shoulder blade tilts back more readily. |
| Doing the exercise while the shoulder is sore | Leave out the pull backwards and practise the rotation only at first. If it keeps hurting, get it looked at. |
Variations
| Variation | When to use it |
|---|---|
| Rotation only, no pull backwards | Arms at your sides; turn them outwards and back to neutral. Also the first choice if your shoulder is sensitive. |
| Rotation plus pull backwards | Turned out, arms a little out to the side, drawing back to the point where your shoulder blades stay back and down. |
| Short hold at the back | Hold for two or three breaths at the end of the range. |
Reps and load
Twelve reps, one or two sets, calm tempo. Runs as a pair with the neutral version — usually the neutral one first, then this one with the lighter bands.
As homework for desk days and as a warm-up before pressing. Anyone who feels a pull at the front of the shoulder in the neutral version should do this one only.
You progress with a short hold rather than heavier bands.
Which muscles do the work
| Muscle | Role | What it does here |
|---|---|---|
| Infraspinatus M. infraspinatus | Prime mover | Turns the upper arm outwards — the leading muscle with the arm at the side |
| Teres minor M. teres minor | Prime mover | Turns the arm out with it and helps keep the humeral head centred |
| Posterior deltoid M. deltoideus pars spinalis | Prime mover | Draws the arm back — takes over more the further the arms travel from the body |
| Middle and lower trapezius M. trapezius pars transversa et ascendens | Synergist | Draw the scapula back and down and hold it there |
| Rhomboids Mm. rhomboidei | Synergist | Draw the scapula towards the spine |
| Thoracic erector spinae M. erector spinae | Stabiliser | Keep the ribcage lifted — upright, the scapula tilts back more readily |
| Deep cervical flexors M. longus colli, M. longus capitis | Stabiliser | Keep the head over the shoulders while the arms pull |
Which muscle leads depends on arm position: with the arm at the side, infraspinatus and teres minor come first; with the arm further out, the posterior deltoid does. Latissimus dorsi and teres major work against this movement, because they turn the arm inwards.
Seeing it done right is not the same as doing it right
Video, cues and variations take you a long way. What they cannot give you is an outside eye at the moment the movement isn't quite there yet. That is what training with me is for — in the studio or in a workshop.
See the workshopsNew exercise pages are added all the time — the newsletter keeps you posted.
Sources
Reinold MM, Wilk KE, Fleisig GS, Zheng N, Barrentine SW, Chmielewski T, Cody RC, Jameson GG, Andrews JR. (2004). Electromyographic analysis of the rotator cuff and deltoid musculature during common shoulder external rotation exercises. J Orthop Sports Phys Ther 34(7):385-394. DOI: 10.2519/jospt.2004.34.7.385 — Needle-electrode recordings in 10 healthy people across seven external rotation exercises, capturing infraspinatus, teres minor, supraspinatus and posterior and middle deltoid. Sidelying external rotation with the arm at the side produced the highest values for infraspinatus (62 per cent of maximum) and teres minor (67 per cent), while prone horizontal abduction at 100 degrees of abduction with full external rotation produced the highest values for supraspinatus (82 per cent) and middle (87) and posterior deltoid (88 per cent). The basis for saying arm position decides which muscle leads. Only 10 people, a young healthy sample, no band exercise and no shoulder extension against horizontal resistance tested, needle electrodes, and peak rather than mean values. (Evidence type: EMG)
Saeki Y, Kubota A, Kishimoto K, Inoue M, Inoue T, Takazawa Y. (2025). Effect of different load of shoulder external rotation exercises on changes in muscle activity and exerted torque. Front Sports Act Living 7:1527296. DOI: 10.3389/fspor.2025.1527296 — Crossover study of 24 arms from 12 healthy men (22.5 plus or minus 1.9 years) performing external rotation with elastic bands at three tensions and matched total workload; torque and activity in infraspinatus, teres minor and posterior deltoid were measured before and after the bout. Under low load, torque and activity in both rotators were unchanged while the posterior deltoid took on noticeably more; under medium and high load all three muscles rose in activity, but the high-load condition produced no torque gain. The basis for the lighter bands: too much pull shifts the work onto the larger muscle without producing more. Young men only, elbow position unfixed, compensatory movements not recorded, everyone on the same band type, and no direct comparison of the neutral against the rotated version. (Evidence type: EMG)
Suzuki Y, Muraki T, Sekiguchi Y, Ishikawa H, Yaguchi H, Suzuki Y, Morise S, Honda K, Izumi SI. (2019). Influence of thoracic posture on scapulothoracic and glenohumeral motions during eccentric shoulder external rotation. Gait Posture 67:207-212. DOI: 10.1016/j.gaitpost.2018.10.022 — Electromagnetic tracking in 15 asymptomatic people during eccentric shoulder external rotation at 90 degrees of abduction, comparing upright with slouched trunk posture, measured at 75, 80, 85 degrees and maximum external rotation. In the upright posture the scapula tilted markedly further back and rotated further out, the demand on the shoulder joint itself was lower, and the external rotation achieved was greater. The basis for the posture cue: the thoracic spine supplies part of the scapular position. 90 degrees of abduction rather than an arm-at-the-side position, an eccentric throwing simulation in young asymptomatic people, and the manipulated variable was trunk posture rather than rotation. (Evidence type: Biomechanics)
Shirai T, Ijiri T, Suzuki T. (2024). Scapular motion during shoulder joint extension movement. J Biomech 166:112019. DOI: 10.1016/j.jbiomech.2024.112019 — Three-dimensional analysis of scapular motion in 22 healthy men (25.8 plus or minus 2.7 years) during seated shoulder extension, assessed from 0 to 50 degrees. In most participants the scapula tilted posteriorly up to roughly 30 degrees and anteriorly beyond that, while upward and external rotation increased throughout. The basis for limiting the range backwards in both versions of the exercise. Young men only, a small sample, seated and without band resistance, and with no rotational condition — the study does not say whether external rotation additionally brings the tilt with it, and above 50 degrees there are no data. (Evidence type: Biomechanics)
Kim H, Kim B, Shim J, Kwon H, Jung J. (2014). Comparative analysis of acromiohumeral distances according to the locations of the arms and humeral rotation. J Phys Ther Sci 26(1):97-100. DOI: 10.1589/jpts.26.97 — Ultrasound measurement of the distance between acromion and humeral head in 34 people without shoulder pain across three arm positions, each while pushing maximally against a table in internal, neutral and external rotation. Significant differences appeared in abduction and in the scapular plane between internal rotation and neutral, and between internal and external rotation — so it is chiefly internal rotation that narrows the space; a gain from external rotation over neutral was significant only at 90 degrees of flexion. Included here as a correction to a widespread rationale. The abstract reports no millimetre values, what was measured is an isometric pushing task at 90 degrees of elevation rather than shoulder extension with the arm at the side, age, sex and training status are missing, and the sample is pain-free. (Evidence type: Biomechanics)
Beard DJ, Rees JL, Cook JA, Rombach I, Cooper C, Merritt N, Shirkey BA, Donovan JL, Gwilym S, Savulescu J, Moser J, Gray A, Jepson M, Tracey I, Judge A, Wartolowska K, Carr AJ. (2018). Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. Lancet 391(10118):329-338. DOI: 10.1016/S0140-6736(17)32457-1 — Placebo-controlled surgical trial at 32 UK hospitals with 51 surgeons; 313 patients with at least three months of subacromial shoulder pain and intact tendons were allocated to arthroscopic enlargement of the space beneath the acromion, arthroscopy alone as sham surgery, or no treatment. The Oxford Shoulder Score at six months did not differ between the two operated groups (32.7 plus or minus 11.6 versus 34.2 plus or minus 9.2 points; difference minus 1.3 points, 95 per cent confidence interval minus 3.9 to 1.3). Included as the decisive counter-evidence: actually enlarging that space conferred no advantage over sham surgery, so more room does not work as an explanation for relief. A surgical population after conservative care had failed, a high rate of assigned treatment not received, and a surgical trial says nothing directly about the effect of an arm position in training. (Evidence type: RCT)
Lafrance S, Charron M, Dube MO, Desmeules F, Roy JS, Juul-Kristensen B, Kennedy L, McCreesh K. (2024). The efficacy of exercise therapy for rotator cuff-related shoulder pain according to the FITT principle: a systematic review with meta-analyses. J Orthop Sports Phys Ther 54(8):499-512. DOI: 10.2519/jospt.2024.12453 — A review with meta-analyses across 22 randomised trials and 1,281 people with rotator cuff-related shoulder pain, organised by frequency, intensity, type and time. Motor-control programmes reduced disability slightly against non-specific programmes but not short-term pain; for eccentric and scapula-focused programmes and for high versus low intensity, certainty stayed low to very low, and not a single comparative trial addressed frequency or duration. The basis for not claiming any preferred status for this version in people with a history of shoulder trouble, and for marking the repetition count as coming from experience. Heterogeneous populations with complaints rather than healthy trainees, and individual arm positions are not analysed at all. (Evidence type: Meta-analysis)
Lewis JS. (2009). Rotator cuff tendinopathy/subacromial impingement syndrome: is it time for a new method of assessment? Br J Sports Med 43(4):259-264. DOI: 10.1136/bjsm.2008.052183 — A critical review asking whether the terms rotator cuff tendinopathy and subacromial impingement syndrome represent a secure mapping of structure to symptoms. The author argues that the common tests and models do not reliably identify the structure and that the labels are better read as clinical hypotheses. Included here as the basis for not presenting the mechanical space explanation as a cause on the page and speaking of subacromial shoulder pain instead. No primary dataset, no sample of its own, and a position paper remains an argument rather than a measurement. (Evidence type: Review)
A note on context: EMG studies show which muscles an exercise recruits — that is a good reason to pick it, but it does not replace adjusting things to the person in front of you. See a doctor for persistent or severe pain.