A visual field report

What 30 seconds of standing still reveals

Hold a phone to your chest and stand quietly, and it records the tiny, constant corrections your body makes to stay upright. We turned those wobbles — from Wobble's first users — into pictures. Here is what the data shows so far.

  • ~15 min read

Background

How the check works

Standing perfectly still is impossible. Your body is always making tiny adjustments — the same small motions you'd make balancing a broom on your palm. The phone sits flat against your chest, and because your chest tilts as you sway, the phone's motion sensor traces the path of your wobble for about half a minute.

Steady stand small, quiet trace Wobbly stand large, restless trace
Looking straight down at your feet — held close together, the way the check is taken — the line traces where your body drifts as you try to stand still. Every number in this report is squeezed from this one wandering trail: a smaller, calmer, smoother line generally means steadier balance.

Your balance leans on three senses working together: your eyes, the pressure and stretch sensors in your feet and joints, and your inner ear. The check deliberately switches some off, one at a time, to see how well the others cope.

👁 EyesSeeing the room around you helps you stay upright.
🦶 Feet & jointsPressure and stretch tell you which way you're leaning.
👂 Inner earA tiny motion sensor in your head feels tilt and turn.

So the check runs four times, flipping two switches: eyes open or closed (vision on/off) and firm floor or soft foam (reliable foot feedback on/off). Foam scrambles the signal from your feet, much as closing your eyes removes sight.

Eyes OPEN Eyes CLOSED FIRM floor FOAM all 3 senses on → easiest eyes off (firm) → a bit wider foam fools the feet → more sway inner ear only → hardest
Each oval is the rough shape of a typical sway, seen from above. With firm ground and eyes open (top-left) it stays small and narrow — what little sway there is runs mostly front-to-back. As a round gets harder the oval grows and, tellingly, spreads sideways: you start to wobble side-to-side as well, not just front-to-back. Foam with eyes shut (bottom-right) — down to the inner ear alone — is the widest and largest of all.

1 · The dataset

Our data so far

Before any finding, the honest part: this is a small, young dataset. Wobble's iPhone app first reached the App Store on 13 May 2026, and the Android version is still in closed testing on Google Play — so at the time of writing the app has been public for only a few weeks. Uploads are opt-in, so what we have is a handful of curious early users, not a representative sample of anybody.

iOS
Live on the App Storesince 13 May 2026
Android
In closed testingon Google Play
20
people who opted in
~1–2
checks per person (typical)
1
power user with dozens of checks
024 681012 the one power user 012–3 4–78–1516–31 32–6364–127 checks completed, per person
How many checks each person has completed. The buckets double as they widen rightward — 1, then 2–3, 4–7, and so on — so the lone enthusiast past sixty still fits beside everyone else. Most people did just one to three; then a long empty gap; then that single outlier. Wide and shallow — rich for one-person stories, thin for population averages.
Read everything here as early signals, not conclusions. We show what's already visible and we say plainly when there isn't enough data yet. The dataset will keep growing over time, so treat this as an early report — the picture should sharpen as more people take the check.

2 · What the data already shows

Four things you can already see

The deeper "for the curious" layers are tucked into expandable panels, so a normal read flows straight past them.

2.1 · The shape of a single check

One round of standing isn't a single number — it's five at once. We measure how busy your sway is (mean velocity), how far it drifts (sway amplitude), its rhythm (centroidal frequency), how jerky it is (jerk), and the size and shape of the area it covers (the sway ellipse). Drawn on five spokes, they form a little shape — a fingerprint of that round.

Not sure what "jerk" or "centroidal frequency" mean? Open any measure to see what a low and a high value look like — and which one is steadier.

Mean velocity — how busysee how

Adds up the whole length of your wobble trail and divides by the time — how far your sway travels each second. A high number means lots of busy little corrections, non-stop.

Low — short trail, few corrections
✓ steadier
High — long, busy trail
Sway amplitude — how farsee how

How far from your average centre you usually stray — the radius of your typical wandering, ignoring how fast you got there. The dashed ring is that radius, like a dartboard around the bullseye — a tight one, or a wide one.

Small — stays parked over one spot
✓ steadier
Large — drifts further from centre
Centroidal frequency — slow drift vs fast jittersee how

The pace of your sway — a few slow, lazy drifts, or lots of quick little jitters. Like big, slow ocean swells versus choppy ripples; it's about how often the sway reverses, not how big it is.

Low — slow, rolling drifts
High — fast, frequent jitters

Neither extreme is simply "better" — rhythm describes the style of your sway, not how much you sway.

Jerk — smooth vs suddensee how

How abrupt your corrections are — gliding smoothly back to centre, or snapping back with sudden jolts. Like a driver who eases on and off the pedals versus one who stamps them and lurches you about.

Low — smooth, gliding corrections
✓ steadier
High — sharp, sudden jolts
Sway ellipse — size & shapesee how

The smallest oval that covers 95% of where your sway went. A bigger oval covers more ground; its shape also hints whether you wobble more one way than the other (it's usually taller than wide — people drift more front-to-back). Like a sprinkler's wet patch on the lawn.

Small — covers little ground
✓ steadier
Large — covers a lot of ground

So how does that fingerprint change across the four rounds? Here are all four on one radar:

A five-spoke radar with all four rounds overlaid. The three easier rounds bunch together near the centre with heavily overlapping bands; the hardest — foam with eyes closed — pushes well outside on every spoke in orange, farthest on mean velocity and jerk.
All four rounds at once, each in its own colour — firm ground, eyes open (C1) in blue, firm ground, eyes closed (C2) in yellow, foam, eyes open (C3) in green, and foam, eyes closed (C4) in orange. The three easier rounds bunch together near the middle, their spreads overlapping so much they're hard to separate; only the hardest, C4, clearly breaks away, pushing outward on every spoke.

New to this kind of chart? Here's how to read it

  • Each spoke is one measure. The five arms are the five numbers above. A point near the centre means a little of that measure — steadier; out toward the edge means a lot — more sway.
  • The shape is the story. Join the five points into a shape: a bigger shape means more sway overall, while a lopsided one tells you how a person wobbles, not just how much.
  • The dark pentagon is your fixed yardstick. It's the cohort's overall typical check, and it's the same on every chart below — so you can compare the charts at a glance. Watch where each condition sits relative to it.
  • The soft band is the normal spread. It shows how much people differ on each spoke. When two bands overlap heavily, the two conditions are genuinely hard to tell apart.

Each chart below zooms in on one pair, always against the easy baseline — firm ground, eyes open (C1), the blue outline.

Firm ground: eyes open vs eyes closed (C1 vs C2)

Radar comparing firm ground with eyes open (C1, blue) and eyes closed (C2, orange). The two outlines almost trace each other and their bands overlap heavily; the eyes-closed round is slightly smaller on sway ellipse and sway amplitude and slightly larger on mean velocity and jerk.
Blue is firm ground, eyes open (C1); orange is firm ground, eyes closed (C2). The two shapes almost trace each other and their bands overlap — on solid ground, closing your eyes barely moves the fingerprint. The only nudges: the eyes-closed round is a touch smaller on sway ellipse and sway amplitude, and a touch larger on mean velocity and jerk.

Eyes open: firm ground vs foam (C1 vs C3)

Radar comparing firm ground (C1, blue) and foam (C3, green), both with eyes open. The foam round sits a small, even step outside the firm round on every spoke, with the largest gaps on jerk and centroidal frequency and the smallest on sway amplitude.
Blue is firm ground, eyes open (C1); green is foam, eyes open (C3). The shape grows, but evenly — the foam round steps out a modest, uniform amount on every spoke (only about half the normal person-to-person spread, which is why they still overlap). The step is largest on jerk and centroidal frequency, smallest on sway amplitude, with sway ellipse and mean velocity in between.
Why firm ground with eyes closed (C2) looks almost free — a caveat. That eyes-shut round follows the eyes-open round immediately, with no pause. By then the person has already settled into a steady stance, so they tend to do well the moment their eyes close. Starting cold, after a rest, it would very likely look noticeably worse. We chose this order on purpose — it's simpler, and beginning from an already-stable stance makes closing your eyes far less of a fall risk. So read the tiny gap between eyes open and eyes closed on firm ground as a best case, not the full cost of losing sight.

Firm ground, eyes open vs foam, eyes closed — the easy round vs the hardest (C1 vs C4)

Radar comparing firm ground with eyes open (C1, blue) and foam with eyes closed (C4, orange). The foam, eyes-closed round sits well outside the baseline all the way round, with only partial band overlap; the gap is widest on jerk and mean velocity.
Blue is firm ground, eyes open (C1); orange is foam, eyes closed (C4). This is the one pairing that clearly separates — the foam, eyes-closed round sits well outside the baseline all the way round, and the bands only partly overlap. The gap is widest on jerk and mean velocity: strip away both vision and reliable foot-feel and the sway turns faster and jerkier.

Put together: the three easier rounds are near-neighbours that share a lot of overlap, while foam with eyes closed (C4) is the one round that stands apart — and there's a tidy reason. Balance runs on three senses, but any two of them carry more than enough signal to keep you upright; the third is spare capacity. Knock out just one — close your eyes (C2), or stand on foam (C3) — and the other two cover for it, which is why those rounds barely budge from firm ground with eyes open (C1). Knock out two at once — eyes shut and foam, which is foam with eyes closed (C4) — and you're down to the inner ear alone, with no backup. That's the round where balance finally gets hard.

2.2 · Foam pillow vs. the standard pad

Clinics use a firm, standardized foam pad (an Airex-style block). At home, people stand on whatever soft thing they have — often a couch cushion or pillow. A natural worry: does a softer pillow even count as the same test?

Our early read — now with a first look at the data — is that a pillow is a milder challenge than the clinical pad, but it disturbs balance in a similar way — the sway fingerprint keeps its shape, just turned down a notch. Here's the hardest round (foam, eyes closed) on each surface, against the all-conditions average:

On the Airex pad

Five-spoke radar: the foam, eyes-closed round on the standardized Airex pad (orange, with a shaded band) sits well outside the all-conditions average (dark) on every spoke — a wide, clear gap.
On the standardized Airex pad, the hardest round (orange) pushes well past the average (dark line) on every spoke — a big, clear challenge.

On a pillow

Five-spoke radar: the foam, eyes-closed round on a pillow (orange, with a shaded band) still sits outside the all-conditions average (dark), but by a smaller margin than on the Airex pad.
On a pillow, the same round (orange) still sits outside the average (dark line) — so it works — but the gap is narrower: a gentler challenge.

Both surfaces clearly stress balance more than firm ground, so a pillow is a valid stand-in. The difference is how much — and it isn't a single number: each measure shifts by its own amount. Here's that gap for one of them, mean velocity, lining up the people who stood on both surfaces:

Scatter of mean sway velocity (deg/s) on the Airex pad (x-axis) versus a pillow (y-axis), one bubble per person who did both, bubble size = number of sessions. A dashed line marks 'pillow equals pad'; almost every bubble sits below it, and the navy best-fit line runs about half a unit under the dashed line — a pillow reads lower than the pad.
Each bubble is one person who did the hardest round on both surfaces (bigger bubble = more sessions). The dashed line is "pillow = pad"; almost every bubble sits below it, and the navy best-fit line runs roughly 0.4 deg/s under it — a pillow reads, on average, about 0.4 deg/s lower than the standardized pad.

How Wobble will handle it. The shift isn't one number — on a pillow, each measure drops by a different amount. So the planned fix corrects every metric on its own, by its typical pad-to-pillow gap — the per-spoke difference between the two radars above. The scatter zooms in on just one of those measures, mean velocity (where the gap is about 0.4 deg/s), because a single metric is easy to picture in 2-D; the other four are handled the same way. The part a correction can't fix: no two pillows are alike, so a pillow adds more scatter than a standardized pad — shifting each average back is one thing, removing the extra noise is another. This is early, thin data and will firm up as more people upload both surfaces. Bottom line: a pillow is fine for tracking yourself over time, but for the cleanest, most comparable numbers, stand on a real Airex-style pad.

2.3 · Which sense is carrying you

Here's where the four rounds pay off. Sway almost always grows as we remove a sense — but how much it grows at each step points a finger at which sense you were leaning on. We measure that with simple ratios: "How many times worse did I get when we took something away?"

amount of sway → eyes open firm eyes closed firm eyes open foam eyes closed foam eyes-shut jump soft-floor jump (eyes shut)
Sway grows as senses are removed. A jump that's much bigger than usual flags the sense that was just taken away — a large eyes-shut jump means you rely on your eyes; a large soft-floor jump (eyes already shut) leans the story toward the inner ear.
The comparisonA big value means…
Eyes-shut jump
(firm floor)
you rely heavily on your eyes to stay steady.
Soft-floor jump
(eyes shut)
with sight off and feet scrambled, you're down to your inner ear — and it may be working hard.
Soft-floor jump
(eyes open)
how well you tolerate an unstable surface while you can still see.

Turn each jump into a single number and you get a sensory ratio: the easy round divided by a harder one. A value of 1.0 means "no cost" — taking that sense away added no sway at all. The bigger the jump, the further the ratio falls below 1.0. Here is what the cohort looks like so far, split into four age groups:

Three sensory-ratio lines (geometric mean, log scale) across four age buckets 16–28, 28–33, 33–41, 41–95, each with a shaded quantile band. Eyes off on a firm floor (C1/C2, blue) stays near 1.0 then dips to about 0.82 in the oldest group. Foam with eyes open (C1/C3, purple) hovers around three-quarters. Both removed — foam with eyes closed (C1/C4, pink) — sits lowest, around 0.55–0.66, falling to about 0.37 in the 41–95 band.
Sensory ratios by age group — about five people per group, with shaded bands showing their spread. Eyes off on a firm floor (blue, C1/C2) hugs 1.0: on solid ground, closing your eyes barely costs anything — your feet cover it. A soft floor with eyes open (purple, C1/C3) falls a bit. Both at once (pink, C1/C4) sits lowest, and falls furthest in the oldest group. Every condition carries its own natural variation, so these values spread wider still; with only ~five people per bucket the bands are broad — read the lines as early signals, not settled facts.

Another way to look at the same question — which sense carries you — maps it straight onto the three senses themselves:

A three-spoke radar with corners Vision (top), Proprioception (bottom-right) and Vestibular (bottom-left), and age-group tabs across the top (Overall, 16–28, 28–33, 33–41, 41–95) with 41–95 selected. A crimson outline with a shaded band sits outside a dark grey outline on all three corners — the oldest age group reaches further out than the all-ages average on every sense.
One corner per sense — vision, proprioception (your feet & joints) and the vestibular (inner-ear) system. The further a corner reaches outward, the worse: more sway, heavier strain on that sense. The dark line is the all-ages average — a fixed yardstick — and the crimson line is the age band you pick (here the oldest, 41–95), with its person-to-person spread shaded around it. This group sits outside the average on all three corners: the older cohort fares a little worse across the board.

A note on availability: the condition radars earlier in this report are part of Wobble's coach dashboard. This sensory triangle isn't in the app yet — it's planned for a future update, where it will be available to any user who opts in.

Why "early signals": small samples don't just miss differences — they tend to exaggerate them, and the ratios themselves wobble from one stand to the next, even in healthy people. The cohort grows with every upload; and to tame that stand-to-stand noise, the plan is to wait until you've done at least six checks and average across them before showing your comparison — a handful of stands smooths out the jitter a single check can't.

2.4 · How it feels vs. what we measure

After each round, people can rate how steady they felt, 1 to 10. Lining those ratings up against the measured sway shows something very human: people rate on a single absolute scale. A foam-eyes-closed round feels shaky and gets rated low — even when it's that person's personal best for the hardest condition.

Scatter of self-rating (1–10) against mean sway velocity (deg/s, log scale), every recording coloured by round. The cloud slopes down to the right: firm ground, eyes open (C1) clusters top-left at high ratings and low sway; foam, eyes closed (C4) sits bottom-right at low ratings and high sway.
Every recording, coloured by round — from firm ground, eyes open (C1) to foam, eyes closed (C4). More measured sway → lower felt-steady rating: a clean downward trend, with the four rounds stepping from top-left (easy) to bottom-right (hard).
The four round means with error bars on the same axes: firm ground, eyes open (C1) top-left (rating ~10, low sway), firm/eyes-closed (C2) and foam/eyes-open (C3) in the middle (rating ~8), foam, eyes closed (C4) lower-right (rating ~5.5, higher sway).
The same data summarised — one mean per round with spread bars. The four rounds separate along both axes at once: as the task gets harder, sway rises and the felt rating falls together.

There's a telling detail in that easy corner. On the simplest round — firm ground, eyes open — almost everyone rates themselves a 9 or 10, no matter what the sensors record. The data backs it up: those top scores pile against the ceiling and barely move with measured sway, so the rating can't tell steady from rock-steady. On easy footing we simply don't notice our own balance — which is part of why a quick, repeatable check is worth doing: it shows what the feeling alone tends to miss.

It's a small but persistent pattern, and it holds within a single person's four rounds — so it doesn't need a big cohort to show up.

2.5 · Why one check is a snapshot, not a verdict

Stand twice and you won't get identical numbers — real bodies vary minute to minute. Some measures are rock-steady on a repeat; others are noisier. Here is roughly how reproducible each one is, from published test–retest studies1 (most dependable first):

MeasureHow reproducible on a repeat
Jerk — smoothnessHighest
Speed — busy-nessVery high
Zone — areaGood
Spread — driftGood
Rhythm — frequencyModerate

So why does Wobble ask for just one short check? Because the best test is the one people actually do. A single two-minute check is quick, needs no equipment, and is easy to repeat — and repeating it is exactly what turns a noisy snapshot into a trustworthy trend. One check tells you where you are today; a handful over weeks tells you which way you're heading.


3 · What's next

Balance and age — what we see, and why we won't call it yet

The headline question almost everyone asks is: does balance get worse with age? The expectation, well established in the literature, is yes — and in a specific, telling shape that we unpack below.

Our data so far hints at exactly that: the firm-ground rounds — eyes open (C1) and eyes closed (C2) — stay low and flat across every age band, while the hardest round, foam with eyes closed (C4), rears up in the oldest group. It looks like a textbook age effect — and that is precisely why we won't claim it.

Median mean sway velocity by age band (0–20 to 80–100) for the four rounds. Firm ground with eyes open (C1) and eyes closed (C2) stay near 1 deg/s across all ages; foam with eyes open (C3) rises gently; foam with eyes closed (C4) jumps to about 7 deg/s in the 80–100 band.
Median sway by age band, per round — from firm ground, eyes open (C1) to foam, eyes closed (C4). The easy rounds stay flat; the hardest spikes in the oldest band — the shape prior research predicts. But read that spike with care — see the note below.
That dramatic spike is one person. The 80–100 band holds a single participant: a 94-year-old who farmed all his life, still walks without trouble, and is — if anything — fitter than typical for his age, with only his hearing and eyesight faded. One remarkable individual is an anecdote, not a trend, and most of the other bands are still thin too. The curve is suggestive, not conclusive.

What the published literature reports — and the shape our own curve should eventually match:

  • As you get older, sway speeds up. Standing still, an older body drifts a little faster than a younger one. It doesn't happen overnight — it starts quietly in mid-life and picks up after about sixty.23 Of everything the check looks at, this speed is the clearest signal of age.4
  • The easy rounds stay easy; the hard ones give way first. The revealing part isn't that everything declines — it's that the hard rounds decline faster, and the more senses you take away, the earlier age shows. The two foam rounds — foam, eyes open (C3) and foam, eyes closed (C4), the hardest of all — start to separate older from younger people in the fifties; firm ground holds out until the eyes close (C2), in the sixties.5 Firm ground with your eyes open (C1) stays steady well into old age. The easy lines stay flat for years; the hard lines are the ones that climb.
  • A little wobble is normal — a lot, less so. Closing your eyes always adds a bit of sway, even when you're young and steady; that's expected, not a warning sign.6 But the hardest round — foam with your eyes closed (C4) — is the real test: in a large national study, more than a third of adults over forty couldn't hold it, and that share rose sharply with age.7 It's the same shape our own oldest-age spike hints at — though with a single 94-year-old in that band, it echoes the research rather than proving anything of our own.

That's the picture large studies of healthy people paint. We'll hold our own numbers up against it — and revisit the spike above — once enough older users have taken the check.

There's a deeper reason to measure rather than guess: people are often poor judges of their own balance. In one study, one in five older adults at high measured risk of falling didn't see themselves as at risk at all8 — the slide is gradual and easy to miss, which is exactly why a quick, repeatable check earns its keep.


Help the picture sharpen

Every honest gap above closes the same way: more checks, from more people. If you'd like the faint lines to become real findings, the best thing you can do is take the check yourself and opt in to share it — anonymously, no account, age only.

Your uploads are anonymous and opt-in. See the privacy policy for exactly what is and isn't shared.

Want a closer look at your own balance?

If you'd like a deeper, personal read of your results, get in touch and we'll connect you with a personal coach. They can walk through your data with you one-on-one and put together a plan tailored to you.

Notes & references

  1. Reproducibility tiers simplified from test–retest ICCs in Mancini M, Salarian A, Carlson-Kuhta P, et al., “ISway: a sensitive, valid and reliable measure of postural control,” Journal of NeuroEngineering and Rehabilitation 9:59 (2012), doi:10.1186/1743-0003-9-59; front–back / side-to-side convention from Prieto TE, Myklebust JB, Hoffmann RG, Lovett EG, Myklebust BM, “Measures of postural steadiness: differences between healthy young and elderly adults,” IEEE Transactions on Biomedical Engineering 43(9):956–966 (1996), doi:10.1109/10.532130.
  2. Wiesmeier IK, Dalin D, Maurer C, “Elderly Use Proprioception Rather than Visual and Vestibular Cues for Postural Motor Control,” Frontiers in Aging Neuroscience 7:97 (2015), doi:10.3389/fnagi.2015.00097.
  3. Era P, Sainio P, Koskinen S, et al., “Postural balance in a random sample of 7,979 subjects aged 30 years and over,” Gerontology 52(4):204–213 (2006), doi:10.1159/000093652; Abrahámová D, Hlavačka F, “Age-related changes of human balance during quiet stance,” Physiological Research 57(6):957–964 (2008), doi:10.33549/physiolres.931238.
  4. Prieto et al. (1996, n. 1); Roman-Liu D, “Age-related changes in the range and velocity of postural sway,” Archives of Gerontology and Geriatrics 77:68–80 (2018), doi:10.1016/j.archger.2018.04.007.
  5. Choy NL, Brauer S, Nitz J, “Changes in postural stability in women aged 20 to 80 years,” The Journals of Gerontology Series A 58(6):M525–M530 (2003), doi:10.1093/gerona/58.6.m525.
  6. Brech GC, Bobbio TG, Cabral KdN, et al., “Changes in postural balance associated with a woman’s aging process,” Clinics (São Paulo) 77:100041 (2022), doi:10.1016/j.clinsp.2022.100041.
  7. Agrawal Y, Carey JP, Della Santina CC, Schubert MC, Minor LB, “Disorders of balance and vestibular function in US adults: NHANES 2001–2004,” Archives of Internal Medicine 169(10):938–944 (2009), doi:10.1001/archinternmed.2009.66.
  8. Delbaere K, Close JCT, Brodaty H, Sachdev P, Lord SR, “Determinants of disparities between perceived and physiological risk of falling among elderly people: cohort study,” BMJ 341:c4165 (2010), doi:10.1136/bmj.c4165.