Shoulder turn in golf: the range your body has

A shoulder turn is not one joint moving

The line across your shoulders at the top of the backswing is a result, not a joint. Four things move it: the thoracic spine rotating on the pelvis, the pelvis turning on the legs, the shoulder blades sliding around the ribcage, and whatever the lumbar spine will give. The line alone cannot tell you which of the four did the work.

The Titleist Performance Institute makes the point on its seated trunk rotation page. Plenty of golfers, it says, "rotate their shoulder blades back and forth" so that it looks like a good shoulder turn while true thoracic rotation is missing. Shoulder turn minus hip turn is a separate number with a guide of its own.

What the thorax actually supplies

36.6° and 38.8°Seated thoracic rotation, each way, 98 tour-card professionals

A 2025 paper in the South African Journal of Sports Medicine set out to establish normative thoracic values, because none existed for that population. Ninety-eight male golfers were measured, all holding a Sunshine Tour and Big Easy card. Each sat with hips and knees at 90 degrees, arms crossed, a small ball between the knees to keep the lumbar spine out of it.

Rotation one way averaged 36.6 degrees, plus or minus 9.8. The other way, 38.8 degrees, plus or minus 10.5. The difference between sides reached significance at p=0.024, which the authors put down to the one-directional work of practising and playing golf. These are working professionals, and the standard deviations are wide enough that plenty of them sat in the twenties.

The lumbar spine has almost nothing to give

About 2 degreesAxial rotation at each lumbar joint, three-dimensional radiography

Pearcy and Tibrewal put the normal lumbar spine under three-dimensional radiography in 1984 and found approximately 2 degrees of axial rotation at each intervertebral joint, L3-4 and L4-5 slightly more mobile than the rest. Five joints at that rate is a rounding error beside a shoulder turn. The lumbar spine bends and side-bends; twisting is not its department.

Which makes one published golf figure worth a second look. A 2015 study in BioMedical Engineering OnLine ran 15 professional golfers through motion capture and reported peak lumbar spine rotation of 45.5 degrees at the top. The paper computes that angle with the Euler method as the difference in orientation between segments, and the segments it defines are the upper torso and the pelvis.

So the 45.5 degrees is the torso turning on the pelvis, taken from markers on the skin. The discs are not supplying it. Any figure handed to you for spinal rotation deserves one question: between which two segments was it measured?

How much of the turn is pelvis

42 ± 7 degreesPelvis orientation at the top, 25 PGA Tour players

The pelvis turns as well. Zheng, Barrentine, Fleisig and Andrews tested 25 PGA Tour and 25 LPGA Tour players and put pelvis orientation at the peak of the backswing at 42 plus or minus 7 degrees for the men, 49 plus or minus 8 for the women. Most of a right angle, before the spine has been asked for a degree.

Set that beside the 45.5 from the motion-capture study and the sum lands near 88. It is not a measurement: two labs, two groups of golfers, and no single swing produced both halves. Treat it as a plausibility check. What it says is that a right-angle shoulder turn is roughly half pelvis, and the half you might stretch is the smaller one.

The legs pay part of it. A prospective study followed 41 injury-free golfers for six months, over which 17 developed low back pain. Against those who stayed well, the pain group showed 6 degrees less lead knee flexion at the top (P = .05) and 6 degrees less lead ankle dorsiflexion there. An association is not a mechanism, but it is a reason to look below the belt.

Telling a limit from a choice

TPI's seated trunk rotation test measures the mobility half. Knees and feet together, upright, arms extended in a W supporting a bar across the shoulders, two clubs laid on the floor as 45-degree markers. The instruction is to see whether the golfer can rotate past the 45 degree range on both sides. Pinning the knees together keeps the pelvis out of it.

Hold that pass mark against the Sunshine Tour figures of 36.6 and 38.8 degrees. The protocols differ, arms crossed and a ball between the knees rather than a bar and a W, but the idea behind them is the same, and a screen a roomful of tour professionals would average below is information rather than a verdict.

The second screen separates mobility from technique. TPI's torso rotation test has you stand in your five-iron posture, arms crossed on the shoulders, and turn the upper body while the lower body stays put. If the turn is short, someone stabilises your pelvis by hand and you go again. Better with the hands on and the problem is stability, not range.

On your own video the shoulder line is the last thing to read. Film down the line, stop the clip at the top, and check the belt line and the lead knee, because those are what a shoulder line borrows from. Pebble Swing reads joint angles out of the frames and the tempo ratio out of the motion, and no video says which joint supplied a turn.