Lower back stretches are among the most searched movements on the internet, and most of the writing about them promises something the movements have never been shown to deliver. This page gives the positions, the holds, and then the part the category leaves out: what the trials found, how small the effects were, and the symptoms that mean you should be examined by a person rather than working through a list.
The short answer
- Movement helps chronic low back pain modestly, and stretching classes did about as well as yoga classes in the best trial of the question.
- No type of exercise clearly beats another. The one you will keep doing is the one to pick.
- A stretch is not treatment, and this site never offers one as treatment.
- "Tight" is a sensation, not a measurement. Felt stiffness and measured spinal stiffness do not match.
- Limber files three of its fifty-six movements to the lower back. Everything else in its back track works the hips, the hamstrings and the upper back.
Read this first
Some symptoms are not a stretching question at all. If any of these apply, stop and get examined rather than working through a routine.
Loss of bladder or bowel control, difficulty passing urine, or numbness in the saddle area — between the legs, around the back passage — alongside back pain, especially with weakness or numbness in both legs, needs assessment the same day. It is rare and it cannot wait.
Also: leg weakness that is getting worse rather than fluctuating; back pain after a fall, a crash or an impact, or after a minor one if you have osteoporosis or take long-term steroids; back pain with fever, feeling generally unwell, or unexplained weight loss; a history of cancer, or new pain that is constant, worse at night and unrelieved by changing position; and a first significant episode before roughly 20 or after roughly 55.
That is not a diagnostic list and it is not exhaustive. It is a list of reasons to see a clinician who can examine you. Pain that is not settling over several weeks, pain that keeps returning, or pain stopping you doing things you need to do are all good reasons to see a physio or a doctor. When to see a physio covers what that appointment is actually like.
Nothing here diagnoses anything and nothing here is treatment.
What lower back stretches do
The best single trial is Sherman and colleagues in the Archives of Internal Medicine in 2011. They randomised 228 adults with chronic low back pain to twelve weeks of yoga classes, twelve weeks of stretching classes, or a self-care book. Both class groups improved on function more than the book, and the two class groups did not differ from each other.
Read that twice, because it cuts both ways. Stretching classes genuinely helped, matching a much more heavily marketed intervention. And whatever helped was probably not the specific poses, because two quite different sets of movements produced the same result.
Hayden and colleagues' 2021 Cochrane review of exercise therapy for chronic low back pain reaches the same shape of answer at scale: exercise probably reduces pain and improves function modestly against no treatment or usual care, and the differences between exercise types are small.
Grade: moderate, with small effects. Real randomised findings, not large ones. The trials are varied, blinding is impossible, and the placebo component of any class-based intervention is substantial.
So the honest working rule is that time spent moving is the active ingredient, and the choice of movement matters much less than the category's confident lists imply. The full evidence article sets out every trial with its numbers.
Six positions and their holds
These are the positions people actually use. The holds are Limber's own defaults, which sit above the tested floor of 30 seconds without pretending that longer buys more.
Cat-cow. On hands and knees, alternating a gentle arch and a gentle round, following the breath. Dynamic — one repetition per breath cycle. Limber runs it for 45 seconds and treats it as a warm-up movement, not a finisher. The full guide has the setup and the three ways it goes wrong.
Child's pose. Kneel, sit the hips back toward the heels, fold forward, forehead down. A long quiet hold at 60 seconds — the longest default hold in the whole catalogue. The guide is here.
Supine twist. On your back, knees drawn up and dropped to one side, opposite arm out along the floor, both shoulders staying down. 45 seconds a side.
Happy baby. On your back, knees toward the chest, holding the outside of the feet or behind the thighs, knees drawn wide and down. 50 seconds, with the lower back staying heavy on the floor.
Figure-4 glute stretch. On your back, one ankle crossed over the opposite thigh, reaching through to draw the lower thigh toward the chest. 40 seconds a side. Filed to the glutes rather than the back, and it is in the back track anyway, for the reason below.
Half-kneeling hip flexor stretch. One knee down, one foot forward, hips pressed gently forward with the ribs stacked over the pelvis. 40 seconds a side.
Two of those six are not back stretches at all, and that is deliberate. The hips and the hamstrings decide how much the lumbar spine has to do in any forward movement, which is why hip flexor stretches and hamstring stretches sit inside a back routine.
Across all of them the rule is the same: a mild stretch, never a sharp one, and no position that reproduces pain travelling down a leg.
The three the app files here
Limber's catalogue holds 56 moves across ten body areas, and exactly three are filed to the lower back as their primary area: cat-cow, child's pose and the supine twist. Seven more carry the lower back as a secondary area — the standing forward fold, happy baby, toe-touch rocks, the lying open-book, the sphinx-to-child flow, the seated tall rotation and a supported backbend hold.
Ten movements out of fifty-six, and only three of them named for the region. That is a small number and it is worth printing rather than hiding, because it tells you what a back routine mostly is: work on everything around the back.
The app's four-week Lower Back Care track makes the point in its own composition. Twelve days, sixty movement slots, and fifteen distinct movements drawn from the catalogue. Child's pose appears on ten of the twelve days, the supine twist on eight, cat-cow on seven and happy baby on seven. The rest is hips, glutes, hamstrings and mid-back. The track's own description says it is mobility support and not treatment, which is the only claim it is entitled to make.
What a reading can tell you
This site's argument is that you should measure. The lower back is where that argument needs its own caveats, and here they are.
The toe touch test is the instrument to start with, and it comes with a specific warning: it is a loaded forward fold. If bending forward is what provokes your symptoms, today is not the day to take it. A reading you had to hurt for is not a reading worth having.
It is also, honestly, a poor measure of the lower back. Mayorga-Vega and colleagues' 2014 meta-analysis in the Journal of Sports Science & Medicine found the sit-and-reach family has moderate validity for hamstring extensibility and only low validity for lumbar extensibility.
That is why the app scores the lower-back area as half the toe touch and half the deep squat — two proxies, neither of which measures the lumbar spine — and labels it as inferred. Run the arithmetic and something sharper falls out: the whole Posterior group score, which averages hamstrings, glutes and lower back, collapses to exactly the mean of your toe touch and your deep squat. Three areas, two tests. The Range Score prints the working rather than hiding it.
And the last caveat is the biggest. A range reading is not a pain score. Improving one does not mean the other followed, and nobody should sell you a number that implies it did. Is your back tight or weak takes that apart, and the rest of the cluster — the red flags, the evidence, the referral — sits under lower back.
Questions
Do lower back stretches actually work?
Modestly, and no better than other movement. In the clearest trial — Sherman and colleagues, Archives of Internal Medicine, 2011 — twelve weeks of stretching classes improved function in chronic low back pain about as much as twelve weeks of yoga, and both beat a self-care book. Cochrane's 2021 review of exercise for chronic low back pain found the same shape: a real but modest benefit, with little to separate one type of exercise from another.
Should I stretch if my back hurts right now?
Gentle movement inside a comfortable range is usually reasonable and is what guidelines recommend over rest, but forcing range while a back is irritable is not. If you have any of the symptoms in the section above — saddle numbness, bladder or bowel changes, progressive leg weakness, pain after significant trauma, or pain with fever or weight loss — do not stretch, get examined. If the pain travels below the knee or comes with numbness or weakness, be assessed before working through any routine.
How long should I hold a lower back stretch?
Around 30 seconds is the supported floor from the general stretching literature, and Limber's back positions default to 40 to 60 seconds. The more useful unit is weeks rather than seconds: the trials that showed benefit ran class programmes over eight to twelve weeks, not single sessions. Consistency across weeks matters more than the stopwatch inside one.
Which stretch is best for the lower back?
No trial ranks one against another, so any confident ordering is invention. Two different sets of movements produced the same result in the one good comparison, which suggests the position matters less than the time spent. Pick what you can do without provoking symptoms, and notice that a back routine is mostly hips and hamstrings anyway.
Can stretching stop back pain coming back?
Exercise reduces the risk of a future episode — Steffens and colleagues showed that in JAMA Internal Medicine in 2016, where education alone, back belts and shoe insoles did not. What that evidence supports is exercise in general, not a particular stretch, and not stretching over strengthening. Pick something you will still be doing in three months.
Is my back tight if it feels tight?
Not reliably. Stanton and colleagues, in Scientific Reports in 2017, measured spinal stiffness objectively in people who reported feeling stiff and found that the feeling did not match the measurement. Tightness is a sensation a guarding region produces just as readily as a short one, which is why months of stretching sometimes changes nothing at all.