Recovery data · 9 min read

Readiness with incomplete data: What the confidence percentage means

The readiness score is the result of a calculation using available signals. Data confidence shows how complete the five data groups used in that calculation are. It is not the probability that the estimate is correct.

Three panels against a mountain ridge: five similarly directed signals, four signals with one gap, and five signals pointing in different directions
On this page

Readiness is a derived estimate. Wearable products choose different metrics, time windows, baselines, and weights, so scores with similar names may summarize different things.

Sources: Doherty et al. (2025).

Two numbers, two questions

Read the readiness score first, then the confidence percentage. They describe different parts of the same calculation.

Readiness score

Question: What direction do the available signals suggest for today?

Limit: The score summarizes a model. It does not directly measure one physiological property or guarantee that a workout is safe.

Confidence

Question: How many of the five data groups were included in the calculation?

Limit: Each group contributes 20 percentage points. Four of five gives 80%, but that is not an 80% probability that the readiness score is correct.

Keep three scenarios separate

You judge whether the signals agree when you interpret them. SlopeReady does not measure agreement or assign an “agreed” status.

Complete and similarly directed

Available
All five data groups; the signals point broadly in the same direction
Missing
No group is marked missing
How to read it
The calculation uses every expected group. Their shared direction is your interpretation of the signals.
What does not follow
A diagnosis, freedom from injury risk, or a guarantee of a safe and successful workout

Partial

Available
A readiness score may appear even when at least one group is missing
Missing
The confidence view names the missing groups
How to read it
Account for the gap. Check the source and recency before deciding what it means.
What does not follow
Low confidence means low readiness or poor recovery

Complete but conflicting

Available
All five groups; for example, good sleep beside unusual HRV or a higher resting heart rate
Missing
No group is marked missing
How to read it
Confidence can stay high. The signals are complete, but the result is harder to explain.
What does not follow
High confidence proves a cause or that the estimate is correct

Sources: SlopeReady: How it works.

Four checks before you explain a mismatch

Do not explain a mismatch from the percentage alone. Check the data path in this order.

  1. Name the missing groups

    Record whether recovery, sleep, HRV, current resting heart rate, or training load is missing. “Incomplete” is too vague on its own.

  2. Check the date and source

    Open the category in Apple Health. Check the latest entry and “Data Sources & Access.” Permissions and available sources vary by category.

  3. Separate measurements from context

    Record the planned workout separately from facts the watch does not know, such as pain, unusual effort, travel, or equipment. That context does not prove a cause.

  4. Do not infer beyond the data

    Classify the case as complete, partial, or complete with conflicting signals. Leave the cause open and do not turn a missing value into a training instruction.

Sources: Ibrahim et al. (2024), Apple Support.

Why apps can disagree despite using the same watch

Even when data comes from the same Apple Watch, apps may choose different metrics, records, time windows, baselines, weights, and missing-data rules. A review of 14 composite health scores from ten wearable brands found substantial differences in those choices.

Sources: Doherty et al. (2025).

Completeness also does not remove measurement error. In a laboratory study of 53 healthy adults, agreement between six wearables and polysomnography or electrocardiography varied by device and metric. The study does not provide a personal error rate.

Sources: Miller et al. (2022).

How SlopeReady handles incomplete data

SlopeReady calculates confidence from five equally weighted groups: recovery, sleep, HRV, current resting heart rate, and training load. Each available group adds 20 percentage points, so four of five gives 80%. A missing value stays missing; the app does not substitute zero or an invented default.

When too little data is available, SlopeReady withholds the numerical readiness score. When a score appears, readiness and confidence are shown separately. Complete but conflicting signals can keep high confidence because confidence measures completeness only.

SlopeReady reads permitted Apple Health data. It does not record workouts, live heart rate, GPS, or ski runs. An empty HealthKit read cannot reliably distinguish a missing record from missing read access.

Sources: Apple Platform Security, Apple Developer, SlopeReady: How it works, SlopeReady: Privacy.

A short note for support

These details help support inspect the data path, but they may not identify the cause.

  • date, time, readiness score, and confidence
  • data groups marked as missing
  • latest timestamp and source in Apple Health
  • planned workout and the mismatch you noticed

Choose the next useful step

Keep data gaps visible when you discuss your plan.

With your permission, SlopeReady reads available Apple Health data and shows readiness separately from confidence. The AI coach can compare the available signals with your plan. It does not fill in missing values.

See how SlopeReady works

Readiness and confidence are not a diagnosis. They do not predict injury, illness, workout safety, or whether a session will succeed. Assess pain, acute symptoms, and persistent unusual changes separately from the score.