✓ Practiceby Lab 10 Get started ›
Crossfire

Should you train your back to round under load, or keep a loaded spine stiff?

What the experts say

Stuart McGillPhD · spine biomechanist, Distinguished Professor Emeritus, University of Waterloo

Keep a loaded spine stiff if you have a pain history; save spinal movement for unloaded conditions. If you have never had back pain, keep your style.

What it rests onRepeated flexion under compression delaminates disc collagen cumulatively, and the nucleus works through the delaminations to form a bulge, faster with more load and more repetitions. Demonstrated in his own porcine work (Callaghan and McGill 2001, Clin Biomech 16(1):28-37, up to 86,400 flexion-extension cycles) and a body of tissue studies he states is about 10 percent of hundreds of experiments over 30 years.

Source: backfitpro.com ↗
Kelly StarrettDPT · physical therapist and mobility coach

Reintroduce spinal flexion gradually, including light loaded flexion, instead of treating a rounded back as dangerous

What it rests onReal life is flexed; capacity you never train is capacity you do not have, and the rigid neutral-spine dogma he helped spread overshot. Graded exposure builds tolerance.

Source: tim.blog ↗

Overview

The positions are narrower than they look and they partly talk past each other on two variables at once, load magnitude and pain history. McGill's rule is aimed at people with a pain history and at meaningful load, and he explicitly exempts pain-free lifters from it. Starrett's softening is aimed at the general population's fear of everyday flexion, with load introduced lightly and progressively. Neither is talking about the same person.

Where they agree

More than either reputation suggests. Both endorse unloaded spinal movement: McGill recommends the cat-camel by name and writes that "When loads on the spine are small, movement is healthy" (mcg-cat-camel). Both brace heavy maximal lifts. Both treat capacity as trainable.

Both reject the caricature that McGill says never bend, and McGill has complained about that misquote in print for a decade. The dispute is genuinely narrow: whether deliberately loading a flexed spine is a training input or an injury mechanism, and for whom.

What would settle it

Prospective trials of graded loaded-flexion training (Jefferson curl progressions, for example) against neutral-spine-only training, on back injury incidence and pain, in healthy and previously injured populations separately, since the positions disagree mainly about the previously injured.

Practice does not pick a winner here

You do, and your call is dated, kept, and yours to change. Follow the people you already trust and what they publish turns into a short daily practice.

Get started

Free to sign up and use. We do not ask for a card.