Nobody can — the brands don't publish the formula, and anyone who says they've cracked it is guessing. So Oriya doesn't fake a conversion. Here's the honest way we put every wearable on one fair board: we compare each person to themselves.
Comparing scores from two different wearables directly is like comparing Celsius to Fahrenheit — same digit, completely different meaning. The numbers aren't built the same way: an independent review of 14 composite scores across 10 brands found they use undisclosed, inconsistently weighted formulas and are rarely validated. Even the raw inputs diverge — measuring the same heartbeats, HRV agreement ranged from near-perfect on one device to poor on another. Put these on a shared leaderboard untouched and the most generous algorithm wins, not the best recovery.
Two “83”s on different scales — that's noise, not a tie.
That rough −8 is only a day-one placeholder. There's no honest formula to convert one brand into another — so the real fix, below, is to rank each of you against your own baseline.
There's no public constant that turns an Oura score into a Whoop score: the brands don't publish their weights, and anyone who claims to have the formula is guessing. So Oriya doesn't convert between brands. It reads your own recent history and finds your personal floor, par and ceiling — then today's score becomes a percentile inside your distribution, and that percentile maps onto the shared 0–100 index.
A Whoop wearer and an Oura wearer on the same board — where rank means “who's closest to their own best,” not “whose algorithm is more generous.”
You need about 14 days of history before your number is honest — the percentile reads your last ~90 days, and until roughly day 14 there isn't enough of it. We show a provisional score before then — and we label it provisional.
On day one there's no history to rank against — so a single screenshot can't be a real baseline percentile yet. To get you on the board immediately, we apply a rough, per-device correction and mark it clearly. It's a starting estimate, not the science.
These are Oriya's own working constants for the single-screenshot estimate — rough approximations, not published or independently validated. They exist only to place a first score before your real baseline does, and they're replaced by your personal percentile the moment you have enough history.
| Device | Native metric | Factor ≈ | Offset ≈ |
|---|---|---|---|
| Oura | Readiness | 0.97 | +2.5 |
| Whoop | Recovery % | 0.92 | −1.0 |
| Garmin | Body Battery* | 0.95 | 0.0 |
| Apple Watch† | HRV + sleep (computed) | 0.90 | +1.0 |
* Garmin's Body Battery is a depletion gauge that drains through the day, so a single reading is the roughest of the four — we treat it as directional only until we can read the underlying signals. † Apple ships no native recovery score; Oriya computes one from Apple Health's HRV, resting heart rate and sleep. Recovery apps like Athlytic, Bevel or Gentler Streak read that same Apple Health data — so they run through this one Apple path, not a separate per-app calibration.
Pulled straight from your screenshot and OCR-verified against the device's own metric label, or read directly through a read-only connection.
With ~14 days of history, your score becomes a percentile inside your own floor-to-ceiling range. On a first screenshot, a provisional per-device estimate stands in — clearly labelled until your baseline exists.
Connected directly, Oriya's own weighting blends sleep, recovery (HRV + resting HR) and activity — the same contributors the research finds in every brand's score — instead of trusting one opaque composite.
Your normalized number joins the squad leaderboard — ranked against your own baseline, never a raw public score. Par 71 marks a typical morning on the shared scale; rank is the gap you hold above your own par.
The only honest way to publish a cross-device conversion is data from people wearing two devices at once — same body, two numbers, every morning. We're recruiting the first dual-wearers now. At 500, we publish the empirical mapping: the actual factors, their confidence intervals, and the sample size behind each one — and you get to tell us we're wrong.
The research doesn't hand us a conversion formula — it explains why one can't exist. That's the whole reason we compare you to your own baseline instead of converting between brands.
1 · You can't derive a conversion from these scores. Doherty C, Baldwin M, Lambe R, Burke D, Altini M. "Readiness, recovery, and strain: an evaluation of composite health scores in consumer wearables." Translational Exercise Biomedicine, 2025;2(2):128–144.14 scores across 10 brands: the common inputs are HRV (86%) and resting HR (79%), but the weightings and formulas are undisclosed and the scores are rarely validated against a reference. There is no published constant to look up — so we don't pretend to have one.
2 · Devices don't even agree on the raw signal. Miller DJ, Sargent C, Roach GD. "A Validation of Six Wearable Devices for Estimating Sleep, Heart Rate and Heart Rate Variability in Healthy Adults." Sensors, 2022;22(16):6317.Measuring the same heartbeats, HRV agreement ran from excellent (ICC 0.99) to poor (ICC 0.24), and sleep-stage accuracy spanned 50–65%. When devices disagree on the underlying physiology, converting one brand's composite into another's only compounds the error — one more reason to rank you against yourself.
3 · HRV metrics aren't interchangeable. Shaffer F, Ginsberg JP. "An Overview of Heart Rate Variability Metrics and Norms." Frontiers in Public Health, 2017;5:258.RMSSD and SDNN are distinct measures (healthy SDNN ≈ 50 ms) — a value on one scale can't be read off as the other.
4 · Against orthosomnia. Baron KG, Abbott S, Jao N, Manalo N, Mullen R. "Orthosomnia: Are Some Patients Taking the Quantified Self Too Far?" Journal of Clinical Sleep Medicine, 2017;13(2):351–354.Fixating on a tracker number can worsen the very thing it measures — so rank is against your own baseline, and a red morning is never a public failure.
The whole point is to share a number with people you choose — which only works if you stay in control of everything underneath it.
When you connect a device, Oriya requests read-only scopes. We never get your account password, and we can't write or change anything in your wearable account.
Granular toggles control exactly what each squad or challenge sees. Nothing is shared until you turn it on, and you can export or delete your data at any time.
For workplace challenges, an admin only ever sees aggregate stats — and only when at least five people are participating, so no individual is ever singled out.
Your recovery data is not a product. We don't sell it, we don't run ads against it, and we design data minimization in from the start.
Informed by FTC Docket C-4747 (Flo Health, 2021)