Release forecasts

Apple OS beta release forecasts

History-based public-release and next-milestone ranges, with the comparison cohort, sample size, spread, confidence, and stale-data safeguards available for every estimate.

Calculated Sep 15, 2026 · Refreshes after Sanity updates

Tracked Cycles: 8
Active Forecasts: 0
Paused / Stale: 3
Backtest Median Error: 4d

Current cycles

Active forecasts

Forecasts pause automatically when the latest recorded milestone is more than 60 days old or the entire historical window has elapsed.

Not enough evidence

Forecast unavailable because there is not enough comparable history for this release stage.

Methodology & evidence

Evidence note. Only 0 comparable cycles found; at least 3 are required.

Not enough evidence

Forecast unavailable because there is not enough comparable history for this release stage.

Methodology & evidence

Evidence note. Only 0 comparable cycles found; at least 3 are required.

Not enough evidence

Forecast unavailable because there is not enough comparable history for this release stage.

Methodology & evidence

Evidence note. Only 0 comparable cycles found; at least 3 are required.

Forecast paused

Forecast paused: the expected next milestone window passed without a recorded milestone. This is anomalous against recent comparable Apple release cycles, though rare historical exceptions exist; it does not mean the release is cancelled or that the public release is late. A newer milestone is needed before publishing another estimate.

Why this is marked anomalous →

Elapsed next-milestone window

Aug 27, 2026 – Aug 28, 2026

Methodology & evidence

Evidence note. Low: limited history (3 cycles) or a wide 1-day middle-50% spread.

Comparable cohort

macOS minor releases at RC 2

15.4 15.2 15.1

Range definition

25th–75th percentile of observed days from RC 2 to public release.

Forecast paused

Why this is marked anomalous →

Elapsed next-milestone window

Aug 27, 2026 – Aug 28, 2026

Methodology & evidence

Evidence note. Low: limited history (3 cycles) or a wide 1-day middle-50% spread.

Comparable cohort

macOS minor releases at RC 2

15.4 15.2 15.1

Range definition

25th–75th percentile of observed days from RC 2 to public release.

Forecast paused

Why this is marked anomalous →

Elapsed next-milestone window

Sep 7, 2026 – Sep 14, 2026

Methodology & evidence

Evidence note. High: 7 same-position cycles with a 5-day middle-50% spread.

Comparable cohort

watchOS .0 cycles at Beta 8

26.0 10.0 8.0 7.0 6.0 5.0 4.0

Range definition

25th–75th percentile of observed days from Beta 8 to public release.

Prior-only backtest. 4-day median error; 75% inside the historical range across 4 tests.

Not enough evidence

Forecast unavailable because there is not enough comparable history for this release stage.

Methodology & evidence

Evidence note. Only 0 comparable cycles found; at least 3 are required.

Not enough evidence

Forecast unavailable because there is not enough comparable history for this release stage.

Methodology & evidence

Evidence note. Only 2 comparable cycles found; at least 3 are required.

Comparable cohort

visionOS .0 cycles at Beta 8

26.0 2.0

Range definition

25th–75th percentile of observed days from Beta 8 to public release.

Historical backtest

In 4 rolling tests, each release was estimated using only cycles already public at that point—never later data. The median absolute error was 4 days, and the actual date fell inside the historical 25th–75th percentile window in 75% of tests. Historical performance is not a promise of future accuracy.