Most of what is written about flexibility by age is written to sell something to people over 40, and it overstates the decline considerably. The best population data available puts the loss across five decades of adult life at a handful of degrees on the joints that have been measured properly.
That is a real decline. It is not the collapse the category implies, it is smaller than the difference between two healthy people of the same age, and it responds to training at every age that has been studied.
The short answer
- Range declines with age, modestly, and more at some joints than others.
- The best adult data is from hip and knee measurements in a national health survey: a few degrees across fifty years.
- Women score higher than men on trunk-flexion tests at essentially every age.
- Cross-sectional data overstates the effect, because it compares different people rather than following the same ones.
- Training works at every age studied. Over 65 the effective hold is longer.
Flexibility by age, in numbers
The most defensible adult figures come from Roach and Miles, writing in Physical Therapy in 1991. They used hip and knee range of motion data from the first National Health and Nutrition Examination Survey — a genuine population sample of American adults aged 25 to 74, which is a rarity in this field.
Their finding: the oldest group had less range than the youngest, and the difference across those five decades was small, on the order of a few degrees for most of the movements measured. They concluded that a single set of normal values could reasonably be used for adults under 74 for most hip and knee motions.
That is worth restating because it contradicts almost everything written on the subject. A national sample found the age effect small enough that one normal range covered most of adult life.
Grade: the strongest adult data available, and limited to hips and knees. Shoulders, spine and ankles have nothing comparable.
What the population data shows
A few patterns hold across the samples that exist.
Not every joint declines equally. Shoulder movements and spinal rotation tend to show larger age-related differences than hip and knee movements. This is consistent across smaller studies and it has never been measured in a large representative sample.
Sex is a bigger term than a decade of age. Women outscore men on trunk-flexion tests in essentially every published norm table, at every age band. Beighton and colleagues, in the 1973 paper that introduced the hypermobility score, found joint mobility higher in females and declining with age in both sexes.
Variation within an age band dwarfs the difference between bands. The spread among 50-year-olds is far wider than the gap between the average 30-year-old and the average 60-year-old. Your age tells you very little about your reading.
Norms for adults over 60
If you are over 60, better reference data exists than for anyone else, and it is worth naming.
Rikli and Jones published normative scores for community-dwelling adults aged 60 to 94 in the Journal of Aging and Physical Activity in 1999, drawn from a large multi-site sample. Their battery includes a chair sit-and-reach for lower-body range and a back-scratch reach for the shoulder — the same movement as the Apley-style reach in Limber's battery.
The tables are broken down by age band and by sex, and they show two things clearly: a steady decline across the decades from 60 to 94, and a persistent gap between men and women, with women reaching further on both tests.
These are the most defensible percentiles in the whole flexibility literature, and they still come with the usual caveat — the sample volunteered, and volunteers for a fitness study are not a random sample of 80-year-olds. The other floor test people meet at this age is the sitting-rising test, which scores getting down to the floor and back up out of ten; what it carries is an association between score and mortality in a cohort of volunteers aged 46 to 80 at one Rio de Janeiro clinic, not a prediction about a person, and it is scored on its authors' rubric rather than Limber's, so it stays out of the Range Score.
Children and the growth spurt
The pattern most often reported in school fitness surveillance is a dip: sit-and-reach scores plateau or fall during the adolescent growth spurt, particularly in boys, and recover afterwards. The usual explanation is that long bones lengthen faster than the soft tissue crossing them adapts.
Grade: thin. The pattern is repeatedly observed in cross-sectional school data; the mechanism is plausible and has not been demonstrated directly, and the effect is confounded by the enormous changes in activity, body composition and limb proportions happening at the same time.
The practical read for a young athlete: a temporary drop in a reach test during a growth spurt is common and is not a training failure.
Why cross-sections mislead
Nearly all of this data is cross-sectional — it measures different people of different ages at one moment — and cross-sections systematically confuse ageing with everything else that differs between generations.
An 80-year-old measured in 1999 grew up in a different world of work, transport, occupation and physical activity than a 30-year-old measured the same day. Some of the difference in their range is time on Earth. Some of it is a lifetime of different habits. Cross-sectional data cannot separate them, and longitudinal flexibility studies following the same people for decades essentially do not exist.
There is also a survivorship effect in the older bands: the 90-year-olds who turn up for a fitness assessment are not a representative sample of 90-year-olds.
So read the age curves as an upper bound on the true ageing effect, not as a forecast for you.
Training still works, at any age
This is the part that matters and it is well enough supported to say without hedging.
Stathokostas and colleagues reviewed flexibility training in older adults in the Journal of Aging Research in 2012 and found that range improved with training across the studies. They were considerably more cautious about carry-over — the evidence that improved range translates into better everyday function was much weaker than the evidence for the range gain itself.
That is an honest split and both halves are useful. If you are 70 and want a better reach behind your back, stretching will give you one. Whether that makes getting dressed easier is a separate claim with much thinner support behind it.
The dose changes after 65
One concrete, actionable difference, and it is the best-evidenced age-specific finding in the field.
Feland and colleagues, in Physical Therapy in 2001, ran the duration question in adults aged 65 and over: 15, 30 and 60 seconds of hamstring stretching, six weeks. In that population 60 seconds beat 30, and 30 beat 15. The plateau that sits at 30 seconds in younger adults sits at 60 in older ones.
The American College of Sports Medicine's 2011 position stand reflects this, recommending 10 to 30 seconds for most adults and 30 to 60 for older adults. How long to hold a stretch works through the whole duration literature.
Otherwise the advice does not change with age: a small dose, most days, on positions you care about, measured on a slow clock. The Range Score tool and the toe touch test run the same scale at 25 and at 75.
What age does not explain
If your reading is low, age is usually a small part of the reason. Bone shape, limb proportions, inherited laxity and how much time you spend near the ends of your range all contribute more — why am I so inflexible puts them in order.
Age also explains nothing about pain. Imaging findings that people assume are ageing — disc degeneration, for instance — are extremely common in people with no symptoms at all, and get more common with every decade. A stiff back at 60 is not a diagnosis. If you have pain that radiates into a limb, numbness, weakness, pain that followed a fall, or back pain with fever, unexplained weight loss or any change in bladder or bowel control, see a clinician who can examine you. Nothing on this page diagnoses or treats anything.
Where the evidence is weak
Longitudinal data. Almost nonexistent. Everything here compares different people rather than following the same ones.
Joints other than hip and knee in adults. No representative sample has ever been published.
Whether the rate of improvement changes with age. Older adults improve. Whether they improve as fast, per unit of work, has not been tested — no trial has compared age groups on the same protocol.
Carry-over to function. Range improves with training in older adults. That this improves daily function is the weaker half of the same review, and it is worth knowing which half of the claim you are being sold.
The rest of the cluster sits under mobility, measured, and what a reading actually means is in what is a good flexibility score.
Questions
Does flexibility decrease with age?
Yes, and by less than most people assume. The best adult population data — hip and knee measurements from a national health survey — found the difference across five decades of adult life was on the order of a few degrees, small enough that the authors suggested a single normal range covered most of adult life.
At what age does flexibility start to decline?
There is no clean threshold in the data. The population samples show a gradual difference across adult age bands rather than a drop at a particular birthday. Sit-and-reach scores also dip temporarily during the adolescent growth spurt, which is a separate phenomenon and recovers.
Can you get flexible after 50?
Yes. Reviews of flexibility training in older adults consistently find range improves with training. What is less well supported is whether that improvement carries over into everyday tasks — the review evidence for range gain is much stronger than the review evidence for functional benefit.
Should older adults stretch differently?
Longer holds, on direct evidence. A trial in adults aged 65 and over found 60 seconds clearly better than 30, whereas in younger adults 30 seconds matches 60. Everything else — frequency, position choice, the instruction to stop at a stretch and never at sharp pain — is the same.
Why are women more flexible than men?
Published norm tables show women scoring higher on trunk-flexion tests at every age band, and hypermobility scores are consistently higher in females. Differences in pelvic geometry, connective tissue and limb proportions are all proposed; none of them has been established as the main driver.
Is my flexibility normal for my age?
Norm tables exist for trunk flexion and, in far better form, for adults over 60 through the Senior Fitness Test. Treat them as an order-of-magnitude check: the variation between two healthy people of the same age is much larger than the difference between age bands, so a percentile tells you less than your own reading six weeks from now.