Layer 01 · Anatomy
How the Shoulder’s Parts Work Together
The humeral head articulates with the glenoid at the glenohumeral joint. The scapula moves relative to the thorax and connects through the acromioclavicular joint to the clavicle; the clavicle connects to the sternum at the sternoclavicular joint. Rotator-cuff muscles contribute direction-specific control, while larger muscles generate and transmit pressing force.
Assess scapular movement in context
Clinical interpretation of scapular movement depends on the athlete, task, load and symptoms. Reviews and consensus statements recommend considering those factors together.[1][2]
AMM Research · Anatomical orientation
The shoulder chain and its support boundary

Posterior view
Scapula on the thorax
Both shoulder blades lie on the posterior rib cage. The socket-bearing scapula connects the upper arm to the shoulder girdle.

Anterior view
Clavicle, SC and AC joints
The sternoclavicular (SC) joint connects the clavicle to the sternum. The acromioclavicular (AC) joint links its lateral end to the scapula.

Shoulder detail
Glenoid, humerus and cuff
The humeral head meets the glenoid. Cuff muscles and tendons contribute to control around this articulation.
The bench meets the back of the torso. Contact and edge clearance belong to the interface record.
See the equipment contact example →Force at the hands contributes to the external loading record. Its magnitude needs a defined measurement.
Force along the bar can alter calculated moments even when the visible path looks similar.
Follow the force record →Layer 01 · Anatomy, continued
Why the Moving Socket Matters
The socket of the shoulder joint, the glenoid, is part of the shoulder blade. Wherever the scapula goes, the socket goes with it. When the shoulder blade glides around the rib cage, rotates or tilts, the socket turns, and the upper arm meets it at a different angle even if the arm has not moved relative to the torso.
That is the link between a bench and the joint itself. On a bench the shoulder blades lie between the rib cage and the pad, so whatever changes where and how they are supported can change how the socket sits under the arm.
In ordinary arm movement the two parts never work separately. Ludewig and colleagues fixed motion sensors to pins in the collarbone, shoulder blade and upper arm of 12 people without shoulder problems and recorded them raising and lowering the arm in three directions[9]. All four shoulder joints contributed substantial rotation: the collarbone rotated backward, the shoulder blade rotated upward and tilted back, and the upper arm rose and rotated in the socket. The study measured arm raising rather than pressing, but it establishes what the rest of this article relies on: the arm and the shoulder blade move as one system.
AMM Research · Anatomy · right shoulder
The shoulder blade carries the socket
Three views of the same joint: the socket on the shoulder blade, what happens to it when the shoulder blade moves, and what holds the front of the joint.
1Socket (glenoid)2Upper arm (humerus),shown see-through
- Socket (glenoid)
- Upper arm (humerus), shown see-through
1 · The socket belongs to the shoulder blade. The glenoid, highlighted, is a shallow cup on the outer corner of the scapula, and the ball of the upper arm sits against it. Wherever the shoulder blade goes, the socket goes with it.
Drawn backretractedDrawn forward
protracted1Close to
the spine2Socket faces
the side3Farther from
the spine4Socket turns
toward the front
- Close to the spine
- Socket faces the side
- Farther from the spine
- Socket turns toward the front
2 · Same arm, different socket. Seen from behind. On the left the shoulder blade sits close to the spine; on the right it has glided forward around the rib cage. The upper arm points the same way in both views, but the socket has turned toward the front, so the arm now sits further behind it.
1Upper arm (humerus)2Front of thejoint capsule
- Upper arm (humerus)
- Front of the joint capsule
3 · Different demands on what holds the joint. Highlighted is the anterior band of the inferior glenohumeral ligament, at the front of the joint capsule. In the cadaver study described below[10], strain in this band rose as the shoulder blade was drawn forward. With the arm further behind the socket, the front of the capsule takes more strain, the rotator cuff works to keep the ball centered, and the shoulder-blade muscles and the bench determine where the socket is held.
Three ways to describe the same arm position
Because the socket moves, “shoulder movement” can refer to three different measurements. Keeping them apart prevents the most common misreadings of bench-press research.
| What is described | What it captures | What it cannot show on its own |
|---|---|---|
| Upper arm relative to the torso | The combined movement of the whole shoulder complex, close to what a camera beside the bench records | How much happened at the joint and how much at the shoulder blade |
| Upper arm relative to the shoulder blade | Glenohumeral motion: the angle at the joint itself, with the socket as the reference | Where the ball sits within the socket |
| Ball shifting within the socket | Humeral-head translation, a separate, millimeter-scale motion usually measured with imaging | Anything from bar path, video or arm angle alone |
Two repetitions that look identical from the side can involve different arm-to-socket relationships if the shoulder blades sit differently. A visibly different bar path does not, by itself, show that the ball moved within the socket, and it says even less about whether any such movement was harmful. Studies that answer these questions measure segment motion, joint translation and external force separately.
Position changes what the passive structures do
Weiser and colleagues tested how shoulder-blade orientation affects the ligaments at the front of the joint[10]. They mounted five cadaver shoulders in the apprehension position, with the arm out to the side and rotated back, then simulated 10° and 20° of scapular protraction, the shoulder blade drawn forward around the rib cage, while pushing the ball of the humerus forward with set loads.
As protraction increased, strain rose in the anterior band of the inferior glenohumeral ligament, one of the main restraints at the front of the joint. Forward sliding of the ball decreased at the same time. Under a 15 N load it measured 6.3 mm with no protraction, 4.1 mm at 10° and 2.5 mm at 20°. The tightened ligament held the ball more firmly while carrying more strain.
These were cadaver joints in a fixed position rather than lifters under a bar, so the study does not measure what happens during a bench press. What carries over is the principle. The orientation of the shoulder blade changes the demand on the structures that hold the joint, even when the arm’s position relative to the torso is unchanged. It also shows why tissue demand and visible displacement are separate questions: a joint can move less while one of its restraints works harder.
Layer 02 · Kinematics
First Ask What Moved
Three distinct records
Name the motion before drawing a conclusion
Bar
Path, displacement and timing.
Shoulder girdle
Scapular translation, rotation and tilt.
Upper arm / torso
Humeral excursion and thoracic posture.
Kinematics describes position and motion. Motion capture, imaging and video-based methods track how the humerus, scapula, torso and bar move through the press.
Motion capture · imaging · video-based kinematics
Angles · translations · paths · timing
Movement is not a direct measurement of joint force, tissue stress, pain or injury.
Define the shoulder movement
Specify the direction and phase: retraction, protraction, rotation, tilt or translation. Assess that motion alongside control of the press.
Layer 03 · Kinetics and models
Then Ask What Force or Moment Was Estimated
Instrumented bar · inverse dynamics · musculoskeletal modeling
External forces · net joint moments · model-estimated reaction forces and muscle demand
Model outputs depend on assumptions; they are not in-vivo tissue measurements or clinical outcomes.
Read the participant and measurement protocol
Noteboom and colleagues tested 10 healthy, experienced strength athletes (nine men and one woman, with 6.7 ± 3.9 years of bench experience) across 21 combinations. Grip widths were 1.0, 1.5 and 2.0 times bi-acromial width; shoulder-abduction targets were 45°, 70° and 90°; scapular conditions were neutral, retracted and released. Participants completed three repetitions with a 16 kg bar. Motion capture and an instrumented bar supplied inputs to an OpenSim shoulder model.[3]
Noteboom et al. · 2024 · Table 2
Technique changes the modeled demand
- 10 experienced lifters
- 21 technique combinations
- 16 kg bar
- 3 repetitions per trial
One selected modeled muscle part, compared across three technique components. Every panel uses the same 0–1 scale; BAW means bi-acromial width.
Grip width
SD 0.20 relative activity
SD 0.17 relative activity
SD 0.13 relative activity
relative activity · bars show means; SD printed for each condition
Scapular pose
SD 0.17 relative activity
SD 0.17 relative activity
SD 0.16 relative activity
relative activity · bars show means; SD printed for each condition
Abduction target
SD 0.17 relative activity
SD 0.17 relative activity
SD 0.16 relative activity
relative activity · bars show means; SD printed for each condition
Body and bar kinematics; mediolateral (sideways) bar force, recorded by a strain-gauge barbell.
Assumed input: vertical force, set from the 16 kg bar weight and split equally between the hands.
Selected glenohumeral and acromioclavicular reaction forces; selected muscle activity.
In-vivo joint force, ligament loading, pain, injury, production bench-pad differences or heavy working sets.
Read the joint-force findings and model methods
Narrower grips and scapular retraction reduced selected acromioclavicular compression, glenohumeral posterior shear and modeled rotator-cuff activity during parts of the repetition. Mediolateral bar force materially affected the estimates. The released condition used a pool noodle under the spine, and released versus neutral conditions did not differ in modeled joint-reaction force. Twelve percent of trials were excluded for missing marker or lateral-force data, and ligaments were absent from the model.[3]
Technique changed selected model-estimated shoulder loads under the tested light-load conditions.
The Visible Bar Path Is Not the Whole Force Record
Force direction · measurement framework
A bar path does not contain the whole force record
Record the external forces
Vertical and mediolateral force components at the bar.
Combine with body geometry
Joint positions and moment arms through the repetition.
Interpret the calculated outcome
Net joint moments or model-estimated reaction forces, under stated assumptions.
In 35 strength-trained adults, Mausehund and colleagues found condition differences of up to 12% in load, up to 43% in shoulder net moments with wider grips and up to 26% in elbow moments as grip narrowed. Their central methodological point was that lateral bar forces materially change calculated joint moments.[4]
Related evidence: grip strategy and torso technique
Larsen and colleagues confirmed that grip affected both the weight lifted and the horizontal-force strategy during 1RM pressing in 14 experienced men.[5] Cudlip and colleagues added a torso-technique comparison: in 20 experienced men at 25%, 50% and 75% of self-reported 1RM, the powerlifting arch increased latissimus dorsi activity, while the standardized technique produced approximately 8% larger integrated shoulder moments.[6] The arch redistributed muscular activity and joint moments in this task.
Layer 04 · Muscle activity
Electrical Activity Is Direction-Specific
Anatomy context · Muscle roles
Several muscles contribute to the press
Moves the upper arm
Helps bring the upper arm across the chest during the press.
Assists shoulder motion
Contributes to shoulder motion; its role depends on the task and position.
Extends the elbow
Contributes to straightening the elbow during the upward press.
Surface or fine-wire electromyography
Recorded electrical amplitude and timing for sampled muscles
EMG is not direct joint force, hypertrophy, pain or injury risk.
Wattanaprakornkul et al. · 2011 · EMG comparison
Pressing and rowing recruit the cuff differently
Bench press
Infraspinatus activity exceeded subscapularis activity.
Row
Subscapularis activity exceeded infraspinatus activity.
Read the muscle-recording protocol and full explanation
Wattanaprakornkul and colleagues recorded nine shoulder muscle sites in 15 participants during bench press and row tasks at 20%, 50% and 70% of maximum load. Infraspinatus activity exceeded subscapularis activity during the bench press, while the pattern reversed during the row.[7] The cuff response was reciprocal and task-direction-specific, rather than one generic “stability” signal.
Layer 05 · Surface interaction
Where Contact With the Bench Enters the Mechanics
A bench establishes contact width, contour, friction, compliance, effective height and an edge around the posterior torso. These properties can affect the setup an athlete uses, resistance to sliding, rack relationship and the contact encountered near the scapular region. They are legitimate independent variables.
AMM Research · Visual 25Equipment photograph
Torso support and shoulder clearance

The Launch Pad® supports the upper back while its contoured edges create clearance around the shoulders and upper arms.
Advanced Muscle Mechanics product-image archive. Original image, unaltered.
Where and how much of the thorax and pelvis are supported under load?
Which edge or contour approaches the upper torso through the repetition?
How do friction and compression affect position or drift?
Do measured kinematics, force, EMG or performance change?
The Athlete-to-Bench Interface covers the full research sequence. The acute study below examines pressing activation and bar movement.
Direct product evidence · interpreted at the measured layer
The Acute Launch Pad Study Changed the Support Condition
In the randomized crossover by Kidwell and colleagues, 10 resistance-trained men performed five repetitions at 70% of each participant’s established 1RM on a conventional flat bench and with the Launch Pad.[8] The Launch Pad condition produced higher bilateral pectoralis sEMG, mean and peak bar velocity and vertical displacement. Concentric power was 249 ± 103 W versus 271 ± 93 W, but the difference was not statistically significant (p = .071).
AMM Research · Measurement and meaning
Keep the measurement attached to its meaning

01 · Observation
Bar moves
- Measurement
- Position + time
- Interpretation
- Bar kinematics

02 · Observation
Muscle is active
- Measurement
- Surface EMG
- Interpretation
- Electrical activity

03 · Observation
Support changes
- Measurement
- Contact / pressure
- Interpretation
- Interface condition
Kidwell et al. · 2026 · Table 1
Six measured outcomes, with the full result in view
- 10 trained men
- 70% 1RM
- 5 repetitions
- Randomized crossover
Right pectoralis sEMG
SD 24.8 % MVC
SD 20.1 % MVC
% MVC · bars show means; SD printed for each condition
Left pectoralis sEMG
SD 23.4 % MVC
SD 20.0 % MVC
% MVC · bars show means; SD printed for each condition
Mean bar velocity
SD 0.09 m/s
SD 0.09 m/s
m/s · bars show means; SD printed for each condition
Peak bar velocity
SD 0.11 m/s
SD 0.09 m/s
m/s · bars show means; SD printed for each condition
Vertical bar displacement
SD 7 cm
SD 4 cm
cm · bars show means; SD printed for each condition
Concentric power
SD 103 W
SD 93 W
W · bars show means; SD printed for each condition
View the original compact numerical summary
Pectoralis sEMG · bar velocity · vertical displacement · concentric power
Scapular kinematics · shoulder-joint reaction force · rotator-cuff EMG · pain · injury
The tested support condition changed pectoralis activity, bar velocity and vertical bar displacement during the submaximal set.
Layer 06 · Clinical meaning
Shoulder Mechanics: Findings and Practical Implications
Grip, scapular setup, torso position and the support surface can change the mechanics of a press. Use those variables deliberately: establish a repeatable setup, check shoulder and upper-arm clearance through the intended range, and assess the athlete’s control and response under load.
For coaching, change one variable when possible and record the condition precisely. For research, match the conclusion to the method. For persistent pain, neurological symptoms, postoperative restrictions or return-to-training decisions, use appropriate clinical assessment rather than a mechanical inference.
Continue the mechanics sequenceScapular Motion During the Bench Press explores the motion terms in greater depthReferences
- Ludewig PM, Reynolds JF. (2009). The association of scapular kinematics and glenohumeral joint pathologies. Journal of Orthopaedic & Sports Physical Therapy, 39(2), 90–104. doi:10.2519/jospt.2009.2808. ↩
- Kibler WB, Ludewig PM, McClure PW, Michener LA, Bak K, Sciascia AD. (2013). Clinical implications of scapular dyskinesis in shoulder injury: the 2013 consensus statement from the Scapular Summit. British Journal of Sports Medicine, 47, 877–885. doi:10.1136/bjsports-2013-092425. ↩
- Noteboom L, Belli I, Hoozemans MJM, Seth A, Veeger HEJ, van der Helm FCT. (2024). Effects of bench press technique variations on musculoskeletal shoulder loads and potential injury risk. Frontiers in Physiology, 15, 1393235. doi:10.3389/fphys.2024.1393235. ↩
- Mausehund L, Werkhausen A, Bartsch J, Krosshaug T. (2022). Understanding bench press biomechanics—The necessity of measuring lateral barbell forces. Journal of Strength and Conditioning Research, 36(10), 2685–2695. doi:10.1519/JSC.0000000000003948. ↩
- Larsen S, Gomo O, van den Tillaar R. (2021). A biomechanical analysis of wide, medium, and narrow grip width effects on kinematics, horizontal kinetics, and muscle activity on the sticking region in recreationally trained males during 1-RM bench pressing. Frontiers in Sports and Active Living, 2, 637066. doi:10.3389/fspor.2020.637066. ↩
- Cudlip AC, Maciukiewicz JM, Pinto BL, Dickerson CR. (2022). Upper extremity muscle activity and joint loading changes between the standard and powerlifting bench press techniques. Journal of Sports Sciences, 40(9), 1055–1063. doi:10.1080/02640414.2022.2046937. ↩
- Wattanaprakornkul D, Halaki M, Cathers I, Ginn KA. (2011). Direction-specific recruitment of rotator cuff muscles during bench press and row. Journal of Electromyography and Kinesiology, 21(6), 1041–1049. doi:10.1016/j.jelekin.2011.09.002. ↩
- Kidwell JA, Yamamoto T, Hetherton KJ, et al. (2026). Acute effects of thoracic-spinal elevation via a novel bench press pad on sEMG and barbell kinetics in resistance-trained males. International Journal of Exercise Science, 19(1), 1003. doi:10.70252/IJES2026103. ↩
- Ludewig PM, Phadke V, Braman JP, Hassett DR, Cieminski CJ, LaPrade RF. (2009). Motion of the shoulder complex during multiplanar humeral elevation. Journal of Bone and Joint Surgery (American), 91(2), 378–389. doi:10.2106/JBJS.G.01483. ↩
- Weiser WM, Lee TQ, McMaster WC, McMahon PJ. (1999). Effects of simulated scapular protraction on anterior glenohumeral stability. American Journal of Sports Medicine, 27(6), 801–805. doi:10.1177/03635465990270061901. ↩
