Read your night the way a sleep clinic reads a PSG.

Two people. Same AHI. Very different nights.

Most sleep apps end the night with a single number. "Last night: AHI 12, moderate." Maybe a colored ring. Maybe a smoothed graph. That's it.

Here's what that single number hides: 30 events spread evenly across 8 hours and 30 events clustered between 3 and 5 a.m. produce the same AHI. They are not the same night.

One is mild snoring with the occasional pause. The other is your body in real distress every morning during the most restorative window of sleep.

What every line in the timeline gives you

For each night, a list. Each entry is one real event:

• Time — start and end, to the second.
• Duration — how long the event lasted.
• Type — snore (palatal / tongue-base / epiglottic) / breathing pause / reduced breathing.
• Audio — tap to hear it.

Above the list: a per-hour density visualization. You see at a glance whether the night was steady or whether something went wrong specifically at 3 a.m.

Below the list: severity buckets per hour, color-coded against the same AHI thresholds clinicians use. Per hour, not per night. Normal under 5, mild 5–15, moderate 15–30, severe over 30. The hour 4 a.m. on a "normal-AHI" night can still bucket "severe" if events cluster there.

Two real cases the timeline showed and the average hid

Nick, 64 — the clustered REM apnea

Average AHI: about 5. Statistically "normal." His timeline showed 80% of events happening between 3:30 and 5:00 a.m. — the late-REM stretch when restorative sleep should be doing its work.

His PCP saw the timeline. Sleep-specialist referral. In-lab study. REM-AHI was 18. He's now on REM-only CPAP support; his morning energy came back over six weeks.

None of that happens from a single AHI number. The timeline made the pattern visible.

Mike, 40 — the CPAP mask seal

Already on CPAP. AHI dropped to 8 on his first week of treatment. Win, right? The timeline showed all 8 events were happening in the first hour of sleep. His mask seal wasn't stabilizing for 60–90 minutes. He adjusted his strap pattern based on a CPAP forum post; the next week, hour-one events disappeared.

The AHI alone wouldn't have told him the cause. The timeline did.

What this isn't

• Not sleep stage analysis. We don't claim to detect REM vs deep vs light sleep. Nick's "REM-clustered" pattern was inferred from time-of-night, not from EEG. Real staging needs a head sensor.
• Not a diagnosis. "80% of events between 3:30 and 5 a.m." is a pattern. What it means medically is for a clinician. We surface the pattern. We don't interpret it.
• Limited by detector accuracy. We catch about 80% of breathing-pause events with 91% precision. The timeline shows what we catch — it doesn't reconstruct what we miss.