Joint rotation is motion around the long axis of a bone. Your femur turning in the hip socket, your humerus turning in the shoulder, your forearm turning palm-up to palm-down, one vertebra turning on the one below it.
It is the plane almost nobody trains, it is usually the first thing to disappear, and the way it disappears is the reason people rarely notice.
Rotation is not one movement
Turn and look behind you. That feels like a single action performed by your torso.
It is not. It is a dozen joints each contributing a small amount, and the amounts are much smaller than people assume.
A 3D CT study of healthy volunteers measured this segment by segment through the thoracic spine. Total rotation of T1 relative to L1 to one side averaged about 25 degrees. Broken down by segment, individual vertebrae contributed between 0.5 and 2.7 degrees each, with the middle thoracic segments doing the most work and the segment at the thoracolumbar junction contributing only half a degree (1).
Half a degree. That is the whole contribution of one joint to the act of turning around.
Which tells you two things. Rotation is a distributed job, and no single joint has much margin. When four or five segments stop contributing their two degrees, the motion does not vanish, because you still need to see behind you. It gets taken from whatever is still willing to move, and the joints that end up over-contributing are the ones that eventually complain.
That is the pattern behind most rotational complaints, and it is why the site of the symptom is so often not the site of the problem.
Where rotation lives
The hip is the biggest contributor and the one people are most surprised by. It rotates internally and externally, and reference data from 200 recreational lifters found meaningful differences by sex and, in women, consistent left-to-right asymmetry across every hip motion measured (2). Asymmetry at the hip is common enough to be unremarkable and important enough to be worth measuring rather than assuming.
The shoulder rotates through a larger arc than any other joint you have, and it is the rotation, not the reach, that most overhead problems trace back to.
The thoracic spine handles most of your trunk rotation, per the numbers above. The lumbar spine handles very little, which is a structural fact about how lumbar facet joints are oriented, and it is why asking your low back to supply rotation it was not built for is a bad long-term strategy.
The forearm rotates through pronation and supination. The ankle and foot rotate more than most people believe, and the loss shows up as a knee that tracks poorly rather than as an ankle complaint. The neck rotates, and it loses range earlier and more quietly than almost anywhere else.
Why rotation goes first
Three reasons, and they stack.
Nothing in a normal week asks for it. Walking, sitting, driving, typing, and most gym training happen in one plane. Push, pull, squat, hinge. All forward and back. A joint holds the range something asks for, and nothing is asking.
Rotation is hard to see. You can watch someone lose shoulder flexion because their arm stops going overhead. Losing 15 degrees of hip internal rotation looks like nothing from outside, and it feels like nothing, right up until a golf swing or a pivot on a court finds the edge of it.
It gets covered for immediately. Because rotation is distributed, the loss at any one joint is instantly absorbed by neighbors. Nothing fails. Nothing feels blocked. The bill arrives years later at whichever joint has been paying it.
What to do about the synovial fluid claim
Worth addressing because the older version of this article made it and because it is everywhere in mobility content.
The claim is that rotational work lubricates the joint by stimulating synovial fluid production. Joint movement does influence the behavior of synovial fluid and the nutrition of cartilage, which has no blood supply of its own and depends on movement for exchange. That much is reasonable.
What does not follow is the neat version, where doing joint circles produces measurably more lubrication and that is why you feel better. The mechanism is more complicated than that and the evidence for the specific claim is thinner than the confidence it is usually stated with. I would rather make the argument that holds up: rotation is trained because you lose it otherwise, and losing it costs you options. That is enough.
How to test yours
Three quick checks. Do both sides and compare, since asymmetry is more informative than any absolute figure.
Hip rotation. Sit on the floor with knees bent and feet wide, then drop both knees to one side and hold. One hip goes into internal rotation, the other into external. Compare sides. Most people find one direction on one side that clearly does not go, and that combination is usually the one causing trouble.
Thoracic rotation. Sit in a chair, cross your arms over your chest, and rotate as far as you can without the hips moving. Sitting removes the hips from the equation. Compare left to right and note where you feel the motion stop.
Shoulder rotation. Arm out to the side, elbow bent to 90 degrees, rotate the hand up and back as far as it goes, then down and forward. Both directions, both arms.
Then repeat all three while producing the motion yourself, with no help from momentum or from the floor. The gap between the range you can get into and the range you can produce is the number that matters, and it is the number we measure joint by joint in the Functional Range Assessment.
How to train it
Start with control before adding range.
The basic input is slow, deliberate circles at each joint, taken to the edge of your current range in every direction, moving that joint and nothing else. Keeping everything else still is the hard part and it is the part that makes it training rather than a warm-up. The full explanation of why we start here covers what those are doing.
From there you load the outer range, which is what turns range you can visit into range you can produce. Isometric work at the end of the available rotation is the tool, and it is uncomfortable in a way that nothing about a stretch prepares you for.
Then it gets connected to something. Rotation trained in isolation is a start, and the point is to have it available while you are moving, which is where rotational strength work and the coiling and rope work from the WeckMethod side come in. Trunk and hips organizing against each other is the closest thing to a fundamental human pattern there is, and it is entirely rotational, which is what coiling actually changes about how you move.
Who this matters most for
Anyone whose sport involves a swing, a throw, or a strike, which covers golf, tennis, pickleball, baseball, and every martial art.
Anyone who sits for a living, because eight hours in a chair is eight hours of no rotation at all, and the thoracic spine and hips take that hardest.
Anyone over forty who wants to keep doing the things they do now, because rotation is what you lose first and it is what makes people start moving like they are old before anything hurts.
And anyone whose complaint keeps coming back to the same spot. A joint that keeps getting irritated is often a joint covering for a neighbor that stopped rotating, and no amount of work on the sore one will change that.
References
- Fujimori T, Iwasaki M, Nagamoto Y, et al. Kinematics of the thoracic spine in trunk rotation: in vivo 3-dimensional analysis. Spine. 2012.
- Cheatham S, Hanney WJ, Kolber MJ. Hip range of motion in recreational weight training participants: a descriptive report. International Journal of Sports Physical Therapy. 2017.
Written by
Brian Murray, FRA, FRSC
Founder of Motive Training
Motive Training is a coaching studio in South Austin built around assessment, joint health, and strength you can use. Train with us in the studio, come to a KINSTRETCH class, or work through the progressions in KINSTRETCH Online.