Posture guide
Anterior pelvic tilt means the top of your pelvis tips forward, which deepens the curve in your lower back. Most people stand like this to some degree: in one study of 120 young adults, 85% of the men and 75% of the women did. Average readings are about 9 to 12 degrees, a little more in women, and part of the number is the shape of your pelvis.
Studies haven't found that a forward-tilted pelvis causes back pain, and the 'tight hip flexors, weak glutes' story doesn't hold up well when muscles are measured. Exercise changes the tilt only slightly, but it does build hip range, glute and trunk strength, and control of your pelvis.
This guide gives the normal range, what the trials found, what the Thomas test really measures, the moves with 3D demos and a 20-minute routine.
Most people stand with the front of the pelvis a little lower than the back: in 120 healthy young adults, 85% of the men and 75% of the women did. With the standard clinic tools, average readings are about 9 to 12 degrees, women about 2 to 3.5 degrees more than men, and readings from 2 to 19 degrees turned up in healthy people.
In 30 pelvises held in exactly the same position, the angle between the bony points at the front and back ranged from 0 to 23 degrees. You can't tell posture from bone shape by eye, and judging posture by looking is unreliable.
People with and without low back pain stand with the same pelvic tilt and the same lower-back curve. In 600 adults, tilt was not linked to back pain, while the endurance of the back muscles was. Reviews have found no agreement that any posture causes back pain.
In trials, exercise and stretching changed standing tilt by about 0.5 to 1.6 degrees, close to the error of the measuring tools, and no trial checked whether the change lasted.
They build what they train: how far your hip extends, glute and trunk strength, and control of your pelvis. You can already tilt your pelvis about 9 to 11 degrees either way on purpose, so you are not stuck in one position.
The popular self-test measures how far your hip extends, and it is misleading unless the pelvis is held still. Hip-flexor length doesn't predict how far your pelvis tilts when you stand.
The pelvis can rock forwards and backwards on the hip joints. In an anterior tilt the top of the pelvis tips forward, so the bony point at the front of the hip (the anterior superior iliac spine) sits lower than the one at the back (the posterior superior iliac spine). Clinics measure the angle between a line joining the two and the horizontal.
Tipping the pelvis forward deepens the curve in your lower back while you hold it: in a lab study, tilting as far forward as people could (11.4 degrees) deepened the curve by 10.8 degrees, and tilting back (8.7 degrees) flattened it by 9 degrees. In relaxed standing, though, how far someone's pelvis tilts and how deep their curve is are only loosely related.
Looking from the side is the most common method and the least reliable: when clinicians rated the lower-back curve from photographs, they agreed poorly with each other and only moderately with themselves. A caliper with its arms on the bony points at the front and back of the pelvis agreed closely with X-rays in one study and gives similar readings when repeated, and smartphone inclinometer apps were reliable too.
The 'pelvic tilt' on an X-ray report is a different angle, used for planning spinal surgery, so its numbers can't be compared with clinic readings. Because healthy people vary so much, a review advises comparing yourself with yourself over time rather than with a 'normal' number.
In 120 healthy young adults measured with a palpation meter (a caliper placed on those bony points), 85% of the men and 75% of the women stood with some anterior tilt; 6 to 7% tilted backwards and the rest were level.
In 100 college-age adults measured with a caliper, men averaged about 8.6 degrees and women about 12 degrees. In 118 active adults aged 18 to 32, men averaged 9.6 degrees (range 3 to 18) and women 11.7 degrees (range 2 to 19). A review used about 8 degrees as a reference value and found no agreed cut-off for 'excessive' tilt.
In both studies, women averaged about 2 to 3.5 degrees more forward tilt than men, with a lot of overlap. It is a normal difference, not a problem to correct.
In 30 pelvises fixed in the same anatomical position, the angle between the front and back bony points ranged from 0 to 23 degrees (average 13), with up to 11 degrees of difference between the two sides of the same pelvis. Two people standing the same way can get very different readings, so a high number on its own doesn't mean poor posture.
Source 4
Largely unknown. Researchers name muscle tone, pain, bone shape, movement at neighbouring joints, and sex. Sitting and high heels are often blamed, but no study we found shows that either causes a lasting tilt: the only study touching on heels was a taping study in 16 women, which says nothing about cause.
In a meta-analysis, standing pelvic tilt (3 studies) and the lower-back curve (8 studies) did not differ between people with and without low back pain. An umbrella review of 41 reviews found no agreement that spinal posture causes low back pain. The authors of the first review note that grouping people by which movements hurt might reveal links that averages hide.
In 600 adults, the size of the lower-back curve, pelvic tilt and the length of the hip flexors, hamstrings and abdominal muscles were not linked to back pain. The endurance of the back muscles was the factor most linked to it. This study looked at one point in time, so it can't show cause.
Source 7
A 2020 review found only 2 randomised trials (72 people) and 2 uncontrolled studies (23 people) of non-surgical treatment for excessive anterior tilt, with very low certainty and no overall evidence that it works. In the main trial, 8 weeks of abdominal strengthening with hip and lower-back stretching, 5 days a week, made the abdominal muscles stronger but changed tilt by only 0.5 degrees (8.7 to 8.2), which could have been chance.
Source 9
One session of hip-flexor stretching in 23 healthy men increased hip extension by 2.6 degrees and reduced standing tilt by 1.2 degrees, with no change in the lower-back curve and no link between the two changes. Over 6 weeks, active or passive home stretching both improved hip extension in 33 people with limited range.
In a crossover trial of 26 adults, tucking the pelvis under during the half-kneeling stretch lowered the hip flexors' resistance to stretch more than the usual hips-forward version, using 2 holds of 30 seconds.
An 8-week home programme of education, core, strength and stretching exercises in 42 young adults with excessive tilt and a backward-facing hip socket changed tilt by 1.6 degrees, which could have been chance, and hip pain didn't improve meaningfully, though 85% kept up the programme. In 35 active men, 18 sessions over 6 weeks reduced tilt while walking compared with men who didn't train, but the groups were not randomly allocated.
In 75 inactive men with deep curves, core stability exercise twice a week for 6 weeks (drawing in the stomach with the lower back pressed down, bridges, crunches and reverse crunches, cat-cow, leg lowering and planks) reduced the curve by about 2 degrees, or 3.4 degrees in an electrical stimulation suit, with almost no change in men who didn't exercise. In 30 inactive women, 24 hour-long sessions over 8 weeks reduced it by about 6 to 7 degrees more than in women who didn't train.
Neither trial measured pain or followed people up afterwards.
Twelve weeks of twice-weekly mat Pilates in 50 active adults improved abdominal endurance and hamstring flexibility but not posture. A 2024 review of 9 small studies (643 people) reported better spinal curves and posture, from studies of low to moderate quality.
The 2020 review found no trial of posture belts or correctors for pelvic tilt, and NICE (the National Institute for Health and Care Excellence) says not to offer belts or corsets for low back pain. Taping the pelvis into a backward tilt reduced tilt by about 5 degrees after a day in one study of 16 women, with no comparison group. Three 15-minute sessions of soft-tissue manipulation (Rolfing) reduced tilt by 1.7 degrees straight after and 1.4 degrees a day later, and nobody checked for longer.
Lower crossed syndrome is a clinical model from the 1980s. It says weak abdominal and glute muscles, combined with stiff, overactive lower-back muscles and hip flexors, tip the pelvis forward, deepen the lower-back curve and cause back pain.
Source 22
When researchers measured the muscles, the links were weak or missing. Hip extension range and pelvic tilt were not related in 25 adults, and abdominal muscle performance and tilt were barely related in 31 adults. In 103 teenage girls, no measure of muscle length was related to tilt, and in 90 healthy adults and 60 adults with back pain, muscle length and abdominal strength explained little of the tilt or the curve.
The authors of one study concluded that strengthening the abdominals, or stretching the back and hip flexors, to correct standing posture 'should be questioned'.
There are no validated criteria for the syndrome. A 2024 scoping review found too little information on how it affects movement and walking, and trials use it as a label without saying how it was diagnosed.
Programmes built on the model (hip-flexor stretches, glute and trunk strengthening) are reasonable exercise. They improve hip range, strength and trunk endurance, even though they barely move the pelvis. The iliopsoas guide covers the claim that a tight psoas pulls your back into an arch.
You lie on your back on a firm table, hug one knee to your chest and let the other leg relax. In the modified version your legs hang off the end of the table. A resting thigh that stays raised is taken as a sign of tight hip flexors, and the knee angle is used to judge the front thigh muscle (rectus femoris).
Source 15
The test measures how far your hip extends, in other words hip-flexor length. It says nothing direct about how your pelvis sits when you stand, and hip-flexor length doesn't predict standing tilt.
Compared with motion capture, the modified test picked up only 32% of people who really had limited hip extension and flagged 43% of people who didn't, because the pelvis tipping during the test caused most of the error. Pass-or-fail scoring of the knee part was unreliable between testers. Measured with an inclinometer and a standard pelvic position, readings were consistent.
Keep your lower back pressed down so your pelvis can't tip, and treat the result as a rough guide to hip range, not a verdict on your posture. This is our editorial advice, drawn from the study above.
Source 15
Two to three sessions a week of the 20-minute routine below: a hip-flexor stretch, pelvic tilts, glute bridges and trunk exercises. That is our editorial starting point, modelled on the trial programmes: core work twice a week for 6 weeks, three sessions a week for 8 weeks, and abdominal strengthening with stretching. The NHS asks adults to do muscle-strengthening activity on at least 2 days a week anyway.
Half-kneeling, tuck your tailbone under and squeeze the glute of the back leg before you lean forward. The crossover trial used 2 holds of 30 seconds with 30 seconds' rest; 2 to 3 holds of 30 to 45 seconds a side on most days is our editorial range. Expect more hip range over about 6 weeks, not a different pelvis.
The trials ran for 6 to 8 weeks. Track stronger bridges, longer planks and easier lunges rather than the mirror: changes in tilt were about 0.5 to 1.6 degrees, too small to see.
Break up long spells of sitting with some movement, as the NHS advises for everyone. There isn't one correct way to stand: you can tilt your pelvis forward and back at will, so change position often rather than holding a 'perfect' one (our editorial advice). The sitting guide covers desk set-up.
Start with the half-kneeling stretch, tailbone tucked: 2 to 3 holds of 30 to 45 seconds each side, most days (editorial, from the trial's 2 holds of 30 seconds). The couch stretch is the harder version; skip it with knee problems. Stretching improves hip range; it changes pelvic tilt only a little.
8 to 10 slow pelvic tilts or bridges, and 6 to 10 slow rounds of cat-cow (editorial). Use the standing tilt as a break from sitting. This is a skill to practise, not a way to reset your posture.
2 to 3 days a week. Our editorial starting points: bridges, 2 to 3 sets of 10 to 12 with a 2-second squeeze; dead bugs and bird dogs, 2 to 3 sets of 6 to 8 each side; planks, 3 holds of 20 to 45 seconds, ending each hold when your hips sag. Bridges, reverse crunches and planks were part of the 6-week core programme that reduced the lower-back curve.
3D demoGlute bridgeBeginner · No equipment
3D demoSingle-leg glute bridgeIntermediate · No equipment
3D demoBarbell hip thrustIntermediate · A barbell + A bench or box
3D demoDead bugBeginner · No equipment
3D demoDead bug heel tapsBeginner · No equipment
3D demoBird dogBeginner · No equipment
3D demoKnee plankBeginner · No equipment
3D demoForearm plankIntermediate · No equipment
3D demoReverse crunchIntermediate · No equipment
3D demoDumbbell goblet squatIntermediate · A dumbbell
3D demoReverse lungeIntermediate · No equipmentDo the anterior pelvic tilt routine 2 to 3 times a week for 6 to 8 weeks, and use the others for variety. The glute workout has no hip-flexor stretch, so add the half-kneeling stretch at the end. Pilates didn't change posture in a 12-week trial: treat it as strength and control.
A 20-minute routine for anterior pelvic tilt: hip-flexor stretches, pelvic tilts, glute bridges, dead bugs, bird dogs and planks, for hip range and strength.
20 minBeginner10 movesA wallA 15-minute beginner core workout on the floor: dead bugs, bird dogs, knee planks, side planks and curl-ups.
15 minBeginner6 movesNo equipmentA 25-minute glute workout with bodyweight and a mini band: bridges, clamshells, band walks, kickbacks and lunges.
25 minBeginner9 movesA wall, A bandA gentle 15-minute Pilates-based routine for ongoing low back pain, built from neutral-spine moves with options for backs that dislike bending or arching.
15 minBeginner10 movesNo equipmentA lower body workout at home: split squats, slider hamstring curls, single-leg bridges, step-ups, band walks and calf raises, with dumbbell options.
30 minIntermediate12 movesA wall, A stepFor a longer structure: a general beginner plan, or four weeks of Pilates, both built on the same kinds of moves.
Your first four weeks: two, then three short strength sessions a week at home plus walks, building gently to a full routine.
2 training days20 minBeginner3 days a weekA free 4-week, 3-day Pilates plan for complete beginners. Two mat days and one wall day, with clear rules for when to move each exercise to its harder version.
3 training days25 minBeginnerPosture exercises are not where to start if you have back pain with any of these. In the US, call 911.
Source 32
If you have back pain and:
Source 32
If back pain:
A pinching pain in the groin when you bend deeply, or ongoing hip pain, is a reason to use the hip impingement guide rather than posture exercises.
Most healthy people have some: 85% of young men and 75% of young women in one study, with averages of about 9 to 12 degrees. Part of the reading is bone shape.
People with and without back pain have the same standing tilt and lower-back curve, tilt wasn't linked to back pain in 600 adults, and reviews found no agreement that posture causes back pain.
Hip-flexor length, abdominal strength and pelvic tilt are barely related, and there are no agreed tests for lower crossed syndrome.
Exercise and stretching trials found changes of about 0.5 to 1.6 degrees, close to measurement error, with no follow-up.
It measures hip-flexor length, and without controlling the pelvis it is a poor test of even that.
No trial supports it, and NICE advises against belts and corsets for low back pain.
It means the top of your pelvis tips forward, so the bony point at the front of your hip sits lower than the one at the back. Most people stand like this to some degree: 85% of young men and 75% of young women in one study. Tipping further forward deepens the curve in your lower back while you hold it.
With the standard clinic tools, healthy young adults average about 9 to 12 degrees, women a little more than men, and readings from about 2 to 19 degrees are common. Part of the number is the shape of your pelvis: in pelvises held in exactly the same position, the angle ranged from 0 to 23 degrees. There is no agreed cut-off for 'too much'.
You can't judge it reliably by eye, and even clinicians agree poorly when they rate posture by looking. Clinics use a caliper or an inclinometer on the bony points of the pelvis, and the useful comparison is with yourself over time. The Thomas test, often suggested online, checks how far your hip extends, not your pelvic tilt.
Studies haven't found that it does. People with and without back pain stand with the same pelvic tilt and lower-back curve, and in 600 adults tilt wasn't linked to back pain while back-muscle endurance was. If your back hurts, exercise helps whatever your posture.
Only a little, as far as trials show: 6 to 8 weeks of exercise changed tilt by about 0.5 to 1.6 degrees, and one stretching session by 1.2 degrees. Exercise does reliably improve hip range, strength and control, and you can tilt your pelvis on purpose whenever you like. Since some tilt is normal, there may be nothing to fix.
A half-kneeling hip-flexor stretch with your tailbone tucked, pelvic tilts, glute bridges, dead bugs, bird dogs and planks, 2 to 3 times a week. That mix mirrors the programmes tested in trials, and tucking the pelvis made the hip-flexor stretch work better in one trial. The 20-minute routine on this page puts them together.
Trials ran for 6 to 8 weeks, and none checked whether changes lasted after people stopped. Hip range and strength improve over those weeks; expect only small changes in the tilt itself.
It's a clinical model, not a proven diagnosis. It says tight hip flexors and back muscles with weak abs and glutes tip the pelvis forward, but measurements show the length and strength of those muscles barely relate to tilt. There are no agreed tests for the syndrome.
There's no evidence they change pelvic tilt, and NICE advises against belts and corsets for low back pain. Taping changed tilt for a day in one small study without a comparison group.
Yes: on average women measure about 2 to 3.5 degrees more than men, with a lot of overlap. It's a normal difference, not a problem.
No study has tested sitting or sleeping positions for pelvic tilt, so choose what's comfortable. Breaking up long periods of sitting is good advice for everyone, and the sitting posture guide covers desk set-up.
People with hip impingement or groin pain move their pelvis differently in lab studies, but an 8-week exercise programme didn't change tilt or hip pain in young adults with a hip socket shape problem. If your hip or sacroiliac (SI) joint hurts, use those guides rather than posture exercises.
Exercise education, not a diagnosis or treatment plan, and clinical review of these guides is pending. Get a curve or hunch that is new or getting worse checked, above all in a growing child or after a fall, and get urgent help for numbness or weakness in the legs or a change in bladder or bowel control. Health disclaimer.
