In Gray Cook’s Functional Movement Screen (FMS) framework, “adding extension to the back” doesn’t fix movement problems because it usually targets the symptom (lack of motion or perceived weakness) rather than the movement strategy causing the problem.
The core idea is this: most dysfunctional movement patterns are not problems of “not enough extension,” but problems of where extension is happening and what is being forced to compensate.
When someone “fixes” a squat, hinge, or overhead pattern by arching the low back, what’s actually happening is:
The lumbar spine is taking motion that should come from the hips or thoracic spine
The pelvis may be tipping anteriorly to “fake” range
The trunk loses its ability to stabilize while moving
So you may see more motion, but not better movement.
Cook often frames this as:
“We don’t want more motion in the spine if the hips are missing it.”
In FMS thinking, healthy movement requires balanced distribution across joints:
Hips = primary source of extension and rotation
Thoracic spine = contributes mobility
Lumbar spine = stability zone, not a prime mover
If someone lacks hip extension, the body will often “borrow” it from the lumbar spine. Training more back extension just strengthens the compensation pattern.
If you train “back extension” (like supermans, aggressive bridging, or excessive arching) without correcting the underlying pattern, you may:
Increase compression in the lumbar spine
Reinforce anterior pelvic tilt dominance
Reduce ability to stabilize the ribcage over the pelvis
Make squatting, running, and hinging less efficient
So the body becomes more mobile in the wrong places.
Cook’s approach is not “get stronger in extension,” but:
Restore hip mobility (especially hip extension and internal rotation balance)
Improve trunk control (ribcage stacked over pelvis)
Re-train sequencing (hips → trunk → limbs, not spine-driven motion)
In other words:
Extension is a result of good mechanics, not a driver of them.
A useful Cook-style takeaway:
If you can’t control extension, adding more extension just expands the error.
In FMS, extension is not the problem — misplaced extension is.
If the body can’t create extension from the hips or thoracic spine, it will steal it from the lumbar spine. Training more lumbar extension just makes the thief stronger.
The nervous system is lazy in a very predictable way.
When a movement demands:
hip extension
thoracic extension
ribcage control
pelvic control
…the body will default to the path of least resistance:
Why?
Because the lumbar spine has:
large vertebral bodies
long moment arms
strong erector leverage
fewer rotational constraints
It’s the “big engine” that can fake almost anything — until it can’t.
So when someone arches their back to “fix” a squat, hinge, or overhead reach, they’re not improving the pattern. They’re shifting the work to the wrong segment.
Cook’s principle:
“Don’t load a compensation.”
FMS is obsessed with where movement comes from, not how much movement exists.
Healthy extension should follow this hierarchy:
Hips — primary extension driver
Thoracic spine — secondary contributor
Lumbar spine — stable anchor
When the hips don’t extend well, the lumbar spine becomes the substitute teacher who keeps getting called in.
But substitute teachers don’t fix the curriculum.
poor hip extension
poor hip internal rotation
poor thoracic mobility
poor ribcage–pelvis stacking
poor timing of core engagement
So adding more lumbar extension is like giving a megaphone to the wrong person.
When you train lumbar extension (supermans, hyperextensions, aggressive bridging), you reinforce:
anterior pelvic tilt
rib flare
lumbar dominance
loss of abdominal control
reduced hip contribution
increased compressive load
This is the classic Cook warning:
“You’re getting better at the wrong thing.”
The athlete feels stronger.
The pattern becomes more stable.
But the strategy becomes more dysfunctional.
This is how people become “strong but fragile.”
Cook’s model always follows this order:
Hip extension
Hip IR/ER balance
Thoracic extension/rotation
Ankle dorsiflexion
Lumbar spine
Pelvis
Ribcage
Half-kneeling
Tall-kneeling
Rolling patterns
Chop/lift
Assisted squat/hinge regressions
The magic is in the order.
If you restore hip extension first, lumbar extension becomes unnecessary.
If you restore thoracic extension, overhead patterns stop stealing from the low back.
If you restore ribcage–pelvis stacking, the core can actually do its job.
Extension becomes a result of good mechanics, not a forced input.
In FMS, extension is not the problem — misplaced extension is.
If the body can’t create extension from the hips or thoracic spine, it will steal it from the lumbar spine. Training more lumbar extension just makes the thief stronger.
The nervous system is lazy in a very predictable way.
When a movement demands:
hip extension
thoracic extension
ribcage control
pelvic control
…the body will default to the path of least resistance:
Why?
Because the lumbar spine has:
large vertebral bodies
long moment arms
strong erector leverage
fewer rotational constraints
It’s the “big engine” that can fake almost anything — until it can’t.
So when someone arches their back to “fix” a squat, hinge, or overhead reach, they’re not improving the pattern. They’re shifting the work to the wrong segment.
Cook’s principle:
“Don’t load a compensation.”
FMS is obsessed with where movement comes from, not how much movement exists.
Healthy extension should follow this hierarchy:
Hips — primary extension driver
Thoracic spine — secondary contributor
Lumbar spine — stable anchor
When the hips don’t extend well, the lumbar spine becomes the substitute teacher who keeps getting called in.
But substitute teachers don’t fix the curriculum.
poor hip extension
poor hip internal rotation
poor thoracic mobility
poor ribcage–pelvis stacking
poor timing of core engagement
So adding more lumbar extension is like giving a megaphone to the wrong person.
When you train lumbar extension (supermans, hyperextensions, aggressive bridging), you reinforce:
anterior pelvic tilt
rib flare
lumbar dominance
loss of abdominal control
reduced hip contribution
increased compressive load
This is the classic Cook warning:
“You’re getting better at the wrong thing.”
The athlete feels stronger.
The pattern becomes more stable.
But the strategy becomes more dysfunctional.
This is how people become “strong but fragile.”
Cook’s model always follows this order:
Hip extension
Hip IR/ER balance
Thoracic extension/rotation
Ankle dorsiflexion
Lumbar spine
Pelvis
Ribcage
Half-kneeling
Tall-kneeling
Rolling patterns
Chop/lift
Assisted squat/hinge regressions
The magic is in the order.
If you restore hip extension first, lumbar extension becomes unnecessary.
If you restore thoracic extension, overhead patterns stop stealing from the low back.
If you restore ribcage–pelvis stacking, the core can actually do its job.
Extension becomes a result of good mechanics, not a forced input.
Your line is dead-on:
If you can’t control extension, adding more extension just expands the error.
Let me sharpen it even further in Cook’s voice: