If you want to know how to measure flexibility at home, the honest answer is that the hard part is not the test. It is taking the same test twice. Precision you cannot repeat is worth less than a crude landmark you can hit identically in six weeks, because the only comparison that ever means anything is your reading against your own earlier reading.
So this page publishes the whole method: five tests, the protocol for each, what each one genuinely reads, how much of the result is noise, and what none of it can tell you. Nothing here needs equipment beyond a floor, a wall and the edge of a bed.
The short answer
- Pick a small set of tests and never change them.
- Take them under the same conditions — same warmth, same hour, same order.
- Record left and right separately. Never average them.
- Write the date on every reading.
- Re-test in weeks, not days. Anything more frequent measures the weather.
How to measure flexibility at home
The five tests below are Limber's battery, which is drawn from the standard field tests rather than invented for an app: a forward fold, a weight-bearing ankle lunge, an Apley-style reach behind the back, a Thomas-style hip flexor test and a deep squat. Four of the five are per-position bands; one is a real measured length. A sixth floor test exists and is deliberately not in that battery: the sitting-rising test scores sitting down to the floor and standing back up out of ten, on its authors' anchors rather than Limber's, so it is read on its own and never enters the Range Score.
| Test | What it reads | Reported as |
|---|---|---|
| Toe touch | Posterior chain — hamstrings, glutes, lower back | 5 bands |
| Knee to wall | Ankle dorsiflexion, each side | Centimetres, 0–20 |
| Shoulder reach | Reach behind the back, each side | 5 bands |
| Hip flexor | Hip flexor length, each side | 3 bands |
| Deep squat | The whole lower body as one screen | 4 bands |
Each one takes under two minutes. The full battery is about ten, and you will do it perhaps twenty times a year.
The five readings, and their bands
Toe touch. Stand with your feet hip-width apart, knees soft. Fold forward slowly and reach for the floor. Stop at a stretch, never at sharp pain. Hold at the bottom, breathe out, and note where your hands are: more than 20 cm from the floor, fingertips to your shins, fingertips to the floor, knuckles to the floor, or palms flat. The toe touch test scores those five bands 10, 35, 60, 75 and 95.
Knee to wall. Face a wall in a half-kneeling stance with your front toe touching it. Drive that knee to the wall with the heel down, then walk the foot back until the knee only just touches. Keep the heel down, and measure from your big toe to the wall in centimetres. Do both sides. The knee to wall test converts the centimetres to a score.
Shoulder reach. Reach one arm over that shoulder and down your back. Bring the other arm up behind your back and reach the hands toward each other, without forcing. Note the gap: more than two hand-widths, one to two, less than one, fingertips touching, or fingers overlapping. Swap arms and repeat — the reading is named for the arm that goes over the top.
Hip flexor. Lie back on a bed or bench with your hips right at the edge. Hug one knee to your chest and let the other leg hang. Keep your lower back flat and watch the hanging thigh: it lifts above the bench line, rests level with it, or hangs below it. Three bands only. It is the coarsest test in the battery and it is worth saying so.
Deep squat. Stand with your feet shoulder-width apart, toes forward or turned out a little. Squat as low as you can with your arms forward for balance, slowly. Settle at the bottom, breathe, and note how deep you got and where your heels are: short of parallel, parallel with heels lifted, full squat with heels down, or a squat you could comfortably rest in.
What each test actually reads
This is the part the category leaves out, and it is where a self-test earns or loses your trust.
A forward fold is not a hamstring test. It is the whole posterior chain plus however much flexion your hips and lumbar spine will give, plus your arm-to-torso proportions. Mayorga-Vega and colleagues meta-analysed the sit-and-reach family in the Journal of Sports Science & Medicine in 2014 and found moderate criterion validity for hamstring extensibility and low validity for lumbar extensibility. It is a decent hamstring test and a poor spine test. That is exactly why Limber lets the toe touch supply the whole hamstring reading but only half the lower-back reading, with the deep squat supplying the other half.
The knee to wall is the best-evidenced test here. Bennell and colleagues reported good intra-rater and inter-rater agreement for the weight-bearing lunge in the Australian Journal of Physiotherapy in 1998 — better than most field tests can claim, and better than any of the banded tests on this page, none of which have that kind of published reliability.
The reach behind the back combines three movements into one number. Shoulder external rotation and abduction on the top arm, internal rotation and extension on the bottom one. Rikli and Jones built a version into the Senior Fitness Test in the Journal of Aging and Physical Activity in 1999 with normative data by age and sex, so it is not unfounded — but a short reach tells you the chain is short somewhere, not where.
The Thomas-style hip flexor test is reliable in a clinic and coarse at home. Published reliability studies use an inclinometer and a second person; three self-rated bands is a much blunter instrument, and one band of disagreement is worth 30 points on a 0–100 scale.
The deep squat is a screen, not a verdict. It reflects ankle dorsiflexion, hip shape, hip range and your femur-to-torso proportions at once. Moran and colleagues' 2017 systematic review in the British Journal of Sports Medicine found that Functional Movement Screen composite scores predict injury poorly, and Bahr argued in the same journal in 2016 that this is a general property of screening tests. A deep squat is a good goal and a useful reading. It is not a report card on how you move.
The rules that make a reading count
- Same protocol. Same setup, same knee position, same cue, same landmark. Write your own version down the first time and follow it afterwards, including the parts that feel too obvious to record.
- Same conditions. Always cold or always warm. Your range at 7am and your range after twenty minutes of movement are different numbers on the same body.
- Same order. Later tests in a battery are taken on a warmer body than earlier ones. Keep the order fixed and the bias is at least constant.
- Both sides, apart. An average hides an asymmetry, and the asymmetry is often the most useful thing on the sheet.
- The date. A reading without a date compares to nothing.
- Stop at a stretch, never at sharp pain. A reading you had to hurt for is not a reading worth having, and it is not repeatable either.
How much change is real change
Every measurement has noise, and on home tests the noise is larger than people assume.
On the knee-to-wall scale, the app scores 4 cm as 35 and 6 cm as 55 — ten points per centimetre through the middle of the range. So a centimetre of difference in where you placed your toe is worth ten points, and the threshold Limber uses to call a left–right gap an imbalance is twelve. One sloppy foot placement can manufacture nearly an entire imbalance.
The banded tests are worse in this specific sense. One band of the toe touch is worth between 15 and 25 points. One band of the hip flexor test is worth 30 or 35. There is no such thing as a small change on a band test: the smallest move you can record is already large.
The practical rule follows directly. Treat a single band's movement, or one or two centimetres, as noise until it repeats. Two readings six weeks apart that agree are worth more than six readings a week apart that disagree.
What these tests cannot tell you
They cannot tell you whether you are injured, and they cannot tell you why something hurts. Range and pain are different problems that often arrive together, which is precisely how people end up stretching something that needed examining.
If you have pain that radiates into a limb, numbness, weakness, pain that followed a fall or an injury, or back pain with fever, unexplained weight loss or any change in bladder or bowel control, stop testing and see a clinician who can examine you. Nothing on this page diagnoses or treats anything, and a test taken on the internet is not an assessment.
They also cannot tell you how much of a low reading is bone. Frank and colleagues, in Arthroscopy in 2015, found cam morphology — a structural feature that mechanically limits deep hip flexion — in about a third of people with no hip symptoms at all. A deep squat that will not deepen is sometimes a hip, not a habit.
Composing the five into one number
Five separate readings are hard to hold in your head, which is why Limber rolls them into a single 0–100 Range Score: each test scores an area, the areas group into five regions of the body, and the score is the mean of the groups you have actually measured. Ground you never surveyed stays unsurveyed — it is not filled in with an average.
You can run the same arithmetic here for nothing. The Range Score tool takes whatever readings you have, composes them, and states how many of the five groups your survey covers. Today the battery reaches four of five, because the spine group is fed by a seated rotation test that is not built — in the app either — and the page says so instead of quietly re-weighting around the hole.
What the number means once you have it is the subject of what is a good flexibility score. What a left–right gap means is in one side more flexible than the other. The definitions sit in mobility, measured.
Where the evidence is weak
Reasonably solid. The weight-bearing lunge has published reliability. The sit-and-reach family has published validity for hamstrings. Repeated measurement by the same person is more consistent than measurement by two different people — Gajdosik and Bohannon set that pattern out in Physical Therapy in 1987 and it has held.
Mixed. What the sit-and-reach says about the lumbar spine. What a deep squat says about anything other than a deep squat.
Thin. Self-rated bands taken alone, by a person who is folded over and cannot see themselves, have essentially no published reliability. We use them because they are repeatable and honest about their coarseness, not because a study validated them. That is a real limitation and it belongs on this page rather than in a footnote.
Questions
Do I need any equipment to measure flexibility at home?
No. A floor, a wall and the edge of a bed cover all five tests. A tape measure helps for the knee-to-wall reading, which is the only one reported as a length; everything else is a named landmark you can check without a second person.
How often should I re-test?
Every two to four weeks. Limber defaults to 21 days and lets you set 14 or 28. Testing more often than that mostly measures how warm you were, how well you slept and what time of day it is — all of which move your range by more than a fortnight of training typically does.
Should I warm up before testing?
Either always or never, and then never change. Warm readings are higher than cold ones, so a warm test compared against a cold one manufactures progress that is not there. If you want the higher number for your own morale, take it warm every single time.
Can I take these tests alone?
Yes, all five. That is the constraint they were chosen under. It costs precision: you cannot see your own hanging thigh clearly, and a self-rated band is coarser than an inclinometer held by a second person. The trade is worth it, because a test that needs a partner is a test you will not repeat.
Are home flexibility tests accurate?
Accurate enough to track yourself, not accurate enough to compare against anyone else. The same person testing themselves the same way gets reasonably consistent readings; two different people testing the same body get less consistent ones. Compare your numbers only to your own.
What if a test hurts?
Stop. The instruction on every test is to stop at a stretch and never at sharp pain, and a reading taken past that point is neither safe nor repeatable. Pain that radiates, numbness, weakness, or symptoms that followed an injury are matters for a clinician who can examine you, not for a self-test.