4.2 The Joint-by-Joint Approach to Movement

The short version

If your hamstrings feel tight every single day no matter how much you stretch them, it is worth asking whether the hamstrings are the whole story. They may simply be the part complaining loudest.

The idea behind this article is simple, and it explains an enormous number of nagging aches. Working up the body from the floor, your joints take turns in what they mainly need. One is built to move a lot. The next is built to stay put. The next moves a lot again. When one of them stops doing its job, the joints either side quietly cover for it — and the ones doing the covering are usually the ones that start to hurt.

Before any of this lands you need the distinction from article 4.1: how far a tissue can be stretched versus how much of that range you can actually control under load. Everything here assumes you have that straight.

Part 1 — Beginner Section: Why Your Tight Hamstrings May Not Be the Whole Story

Your knee hurts, so you treat your knee. Obvious, right? Except that the knee is sometimes only where the complaint surfaces — and this section is about why it is worth looking further.

One honest disclaimer before we start: this is a map, not a law. Bodies are messier than any tidy alternating pattern, and plenty of people break the rule without breaking anything else.

But a map still beats the default plan, which is "stretch whatever hurts". Think of it as a search strategy: not a diagnosis, just a far better first guess.

The whole idea in one sentence

Start at the floor and work upwards: your joints take turns in what they mainly need. One is built to move a lot, the next is built to stay put, the next moves a lot again.

Move, hold, move, hold, all the way up the body. Break that chain anywhere and the neighbours quietly inherit the missing job — whether they're built for it or not.

One-line recap: pain can turn up in the joint that's doing somebody else's work — a hypothesis worth testing, not a diagnosis you can make on your own.

Doors, hinges and frames

Picture a row of doors. Hinges are meant to swing; frames are meant to stay square.

Now seize one hinge. The door still has to open, so the frame bends a little instead — and for a while, nothing looks wrong at all.

Then the frame cracks. The crack is where the pain is. The seized hinge may well be part of why it happened — though frames crack for their own reasons too.

In your body, the seized hinge is very often an ankle or a mid-back. The cracking frame is a knee or a lower back.

Three patterns you'll see in every gym

Stiff ankle, sore knee. If your ankle won't let your shin travel forward — picture the bottom of a squat — that range still has to come from somewhere. Your knee obliges, usually by drifting inward.

Trouble is, your knee is basically a hinge. Side-to-side is not a direction it's designed to offer, so it pays for every degree it lends.

And nobody blames the ankle, because the ankle doesn't hurt. It's just quietly refusing to participate.

It's also why you can stretch your calves for months and see nothing change: the block is often in the ankle joint itself, not the muscle. You keep stretching what's tight instead of freeing what's stuck. The same borrowing shows up when you land from a jump — knees caving inward is often an ankle story as much as a knee one.

Stiff mid-back, sore lower back or shoulder. Your upper back is supposed to rotate and extend. To reach fully overhead, it has to extend underneath the shoulder — the shoulder can't finish the move alone.

So when the mid-back won't play, your lower back arches to make up the shortfall on every overhead press, and your shoulder scrapes for the last few degrees it never had. Meanwhile the actual culprit feels completely fine.

Stiff hip, sore lower back. Same story, one floor up. If your hip runs out of range partway into a squat or a stride, your spine bends to finish the job.

And it does that a few hundred times a session, without anyone noticing a thing.

Spot the thread. In all three, the symptom sits about one joint away from the restriction. That single observation is basically the entire practical value of this model — and it's why chasing the sore spot so often gets you nowhere.

Try it on a real problem

Your lower back feels tight after squat day. The instinct is to stretch the lower back and pile on some core work.

The joint-by-joint reading says something different: check the ankle and the hip first. If either one is short, your lower back has been quietly filling the gap on every rep.

Stretch the sore bit and it'll feel wonderful for twenty minutes — and often change very little. The reading to test is that the back was covering a shortfall rather than creating one; that is a hypothesis to check, not a verdict on the back.

Restore the ankle and the hip, then teach the trunk to stay quiet, and the symptom usually disappears without ever being treated directly. Nothing mystical about it — you just gave the borrowed job back.

Steal this habit. When something aches, look one joint up and one joint down before you treat anything. Often the ache is doing somebody else's work, and it'll keep doing it until you hand that job back. This applies to ordinary training niggles; pain that persists, follows an injury, or comes with swelling, giving way, numbness or weakness needs assessing properly rather than screening around.

Here's your cheat sheet for where to look first:

• Sore knee? Check the ankle below it and the hip above it.
• Sore lower back? Check the hips and the mid-back.
• Sore shoulder? Check the mid-back and the shoulder blade.

Part 2 has the machinery: where this model came from, its honest limitations, and a segment-by-segment walk up the body from the ankle. It's more technical, so head there when you want the mechanism rather than the map.

Part 2 — Advanced Section: What the Model Gets Right, and Where It Breaks

Origins and honest limitations

The joint-by-joint concept was popularised in strength and conditioning by Michael Boyle and physical therapist Gray Cook as a clinical heuristic. It was never presented as anatomy; it was presented as a decision shortcut. That distinction matters, because the model is frequently repeated as though each joint has only one job. Every joint requires both mobility and stability. What alternates is the primary training emphasis when a movement problem appears.

Two honest caveats belong alongside the model:

• It is a hypothesis, not a cause. Regional interdependence — the idea that a restriction at one segment shows up as a symptom at the next — is one of several explanations worth testing when a movement problem appears. It does not establish that the neighbouring joint caused the pain, and it never rules out local pathology at the site that actually hurts.
• It is not a substitute for assessment. Persistent or worsening pain, pain following a fall or collision, swelling, a joint that gives way or locks, numbness, pins and needles or weakness, night pain, or a genuine loss of function all need a qualified assessment. Screen the neighbours by all means, but not instead of getting those looked at.

Segment-by-segment mechanics

Ankle — mobility priority (dorsiflexion)

Closed-chain dorsiflexion requires the talus to glide posteriorly within the mortise while the tibia translates anteriorly. Restriction is commonly posterior capsule and talocrural joint mechanics rather than gastrocnemius or soleus length, which is why athletes can stretch calves for months with no change in weight-bearing dorsiflexion. Functionally, insufficient dorsiflexion reduces knee flexion range in the squat, shifts load anteriorly, and is associated with increased dynamic knee valgus during landing.

Knee — stability priority

The tibiofemoral joint is a modified hinge with small, obligatory rotation coupled to flexion and extension. It is not designed to create frontal-plane range. Frontal-plane knee motion under load is commonly attributed to a hip that is not controlling femoral adduction and internal rotation, or to a foot and ankle that cannot organise the arch — an attribution to test rather than assume, since local knee pathology can produce the same picture.

Hip — mobility priority

A ball-and-socket joint with the largest available motion in the lower body, but with substantial individual bony variation. Acetabular depth, version and femoral neck angle differ enormously between people; two athletes with identical soft tissue can have very different end ranges. Programming that assumes a universal “correct” squat depth ignores this.

Lumbar spine — stability priority

Facet orientation in the lumbar spine restricts axial rotation to a small range per segment. Rotation demanded of the lumbar spine because the thorax will not rotate is therefore concentrated across few segments and repeated thousands of times in rotational sports.

Thoracic spine — mobility priority

Facet orientation and rib attachments permit substantial rotation and extension. The thorax is the primary rotational engine for throwing, striking and swinging. Loss of thoracic extension also directly limits overhead reach, because full shoulder flexion requires the thorax to extend underneath it.

Scapulothoracic region — stability priority (with controlled motion)

The scapula must move — upward rotation, posterior tilt, external rotation — but it must move on a stable base and in proportion to humeral elevation. Uncontrolled scapular motion is one proposed contributor to impingement-type symptoms at the joint above, rather than an established cause of them.

Glenohumeral joint — mobility priority

The least bony constraint in the body and correspondingly the most dependent on active control. Range without control at the shoulder is the single most common precursor to overhead-athlete pathology.

Force-transfer diagram

Why the nervous system is the fastest lever

Range of motion changes across a session are dominated by stretch tolerance and motor control rather than structural adaptation (Sharman et al., 2006). Reflexive mechanisms — autogenic inhibition following an isometric contraction, reciprocal inhibition of the antagonist, and reduced protective guarding once the brain registers a position as controllable — explain most acute gains (Sharman et al., 2006). Structural change in muscle and connective tissue is real but slow, and requires loaded end-range work over weeks to months. That is a mechanistic argument for loaded, controlled end-range training rather than direct evidence that it beats passive stretching: the underlying reflex mechanisms are reasonably well described for contract-relax and PNF-style work, but mechanism evidence is indirect evidence, and no branded end-range system has been shown superior to another without head-to-head trials.

Part 3 — Testing: Find the Quiet Joint

Test in weight-bearing wherever possible. Passive table tests tell you what tissue allows; weight-bearing tests tell you what the athlete can actually use.

Ankle: Knee-to-wall

Hip: Supine flexion, prone internal and external rotation

Thoracic spine: Seated rotation

Shoulder: Elevation with the wall

The screening logic

Part 4 — Practical Section: Building the Missing Range

The three-step sequence

Skipping step two is why mobility work so often fails. Range that is never loaded is range the nervous system will not release under load.

‌

Programming variables

A sample week layered onto an existing programme

What each element in the week is doing

‌

• Ankle mobilisation with band distraction, 2 x 60 s per side (Day 1, pre-lift). The band pulls the talus posteriorly, addressing the joint restriction rather than the calf, which is why it is placed before squatting rather than after.
• Heel-elevated ankle rocks, 2 x 8 slow. Take the newly available dorsiflexion into a loaded squat-like position, so the joint learns to use the range immediately.
• 90/90 hip switches, 3 x 5 per side (post-lift). Restore rotational hip range after a session of largely sagittal loading, which is the pattern most likely to leave the hip feeling stiff.
• Thoracic extension over a foam roller, 2 x 8 (Day 2, pre-lift). Opens thoracic extension so the shoulder does not have to borrow range from the lumbar spine during overhead pressing.
• Quadruped thoracic rotation, 2 x 6 per side. Trains thoracic rotation specifically, which the joint-by-joint model treats as a mobility joint and which is commonly the limiting factor overhead.
• Prone scapular Y-raise, 3 x 8 with a 2 s hold (post-lift). Builds active control of the upward-rotation position the mobility work created, converting passive range into usable range.
• Hip flexor and adductor loaded end-range holds, 2 x 20 s per side (Day 3). Prepare the tissues that reach their longest lengths during sprinting and cutting, which is what the session that follows will demand.
• Repeat Day 1 with a deeper squat variation (Day 4). The deeper variation exists to spend the range that was gained. Mobility work that is never used in a loaded pattern does not persist.
• Repeat Day 2 with an overhead carry (Day 5). The carry integrates thoracic position, scapular control, and trunk stiffness into one loaded task, which is the point at which the separate pieces become a movement.
• Placing mobility before or after the lift. Restrictions that limit the day’s main lift are addressed before it. Range that needs consolidating is trained after it, once the joint is warm and the main work is banked.

‌

Fatigue and recovery notes

End-range loaded work is low in systemic cost but can produce noticeable local soreness for 24–48 hours the first two weeks. If soreness is limiting the main training session, reduce set count before reducing frequency — frequency is doing the useful work here.

Part 5 — Sport Applications

Exercise Library

Band-Distracted Ankle Mobilisation

90/90 Hip Switch

Quadruped Thoracic Rotation

Prone Y-Raise with Hold

Recommended Viewing

Channel links are provided rather than specific video URLs so the reference does not break as catalogues change.

‌

Frequently Asked Questions

Is the joint-by-joint model scientifically proven?

It is a clinical heuristic, not a validated theory. It is useful because it widens attention from the painful site to the neighbouring segments, but it does not identify a cause and it does not replace an assessment when pain is persistent, followed an injury, or comes with swelling, instability, numbness or weakness. Treat it as a search strategy, not a diagnosis.

If my lumbar spine needs stability, should I avoid loaded spinal flexion?

No. The spine adapts to load like any other tissue, and progressive, controlled loaded flexion has a legitimate place in training and rehabilitation. The argument is about sequencing: The lumbar spine should not be the first segment to move when the hips or thorax should have supplied the range.

How long until I see change?

Acute within-session change is normal and mostly neural. Durable change in a weight-bearing test typically takes four to eight weeks of consistent, loaded end-range work. If nothing has moved in four weeks, reassess whether the limit is bony.

Should I stretch before lifting?

Prolonged static stretching immediately before maximal strength or power work can transiently reduce force output. Brief mobilisation followed by active loading of the new range does not carry the same penalty and is the preferred pre-session structure.

My hips are “tight” but the end range feels hard and blocky. What now?

A hard, abrupt end-feel usually indicates bony contact rather than tissue restriction. Adjust the task — stance width, foot angle, squat depth, bar position — instead of trying to force range that anatomy will not give.

Key Takeaways

Next in this series: 4.3 — Movement Screening and What It Can and Cannot Predict.

References

Sharman, M. J., Cresswell, A. G., & Riek, S. (2006). Proprioceptive neuromuscular facilitation stretching: Mechanisms and clinical implications. Sports Medicine, 36(11), 929–939. Read on PubMed