---
date: 2026-09-10
type: sleep
tags: [sleep, late-bedtime, hrv-high-normal, deep-heavy, rem-low]
---

# Sleep — 2026-09-10

**8h 26min** · 00:10→08:56 · REM: 9% · Deep: 39% · HRV: **123** · Resting HR: 49 · Woke: 2× · Score: **91/100**

> Score **91 — new best in the log** (beats 90 from Jul 20), better than 99% of users. No narrative given for the night. Watch advice: woke 2×, slightly above average; suggested meditating before bed.

**Interpretation:** The highest-scoring night on record, and it was built on **deep sleep, not on duration or timing**. Deep **39% (3h 15min)** at **95-point continuity** is the heaviest repair share since Sep 5, light 52% sits normal for the first time in three nights, and deep continuity 95 is the cleanest of the week. The shape is unmistakably *repair-first*: the body spent the night on tissue/immune work and shorted dreaming to pay for it — **REM 9% (48 min) is the second consecutive night under 10%** (Sep 8: 7%), which is still the Sep 7 crash-night (32ms, ~4h) being worked off rather than a new problem.

HRV **123ms** lands high-normal, third night of the settle: **145 → 127 → 123**. That is not a recovery failure — it's the *bedtime* ceiling. The two nights at 145 (Sep 2, Sep 8) both went to bed pre-midnight (23:50 / 22:59); Sep 9 and Sep 10 both went to bed at ~00:10–00:25 and both landed in the 120s. Same story the scores don't capture: a late bedtime costs ~20 ms of HRV but, with 8h26 in bed, it doesn't dent the score at all. Resting HR 49 and SpO₂ 98% close it out — this is a genuinely recovered base.

*(Curiosity, not signal: 00:10→08:56 is the exact clock of [[Sleep/2026-07-27]] — same bedtime, same rise time, eight weeks apart.)*

| Metric | Value | Reference |
|--------|-------|-----------|
| Night sleep | **8h 26min** | 6-10h (Normal) |
| Bed → Rise | 00:10 → 08:56 | — |
| Deep sleep | **39% (3h 15min)** | 20-60% (Normal) |
| Light sleep | 52% (4h 23min) | <55% (Normal) |
| REM sleep | 9% (48min) | 10-30% (**Low**) |
| Deep sleep continuity | 95 points | 70-100 (Normal) |
| Times woke | 2 times | 0-1 (**High**) |
| Breathing quality | 98 points | 70-100 (Normal) |
| HRV | **123ms** | 77-138ms (Within range, upper) |
| Resting HR | **49 bpm** | 48-58 bpm (Within range, low) |
| SpO2 | 98% | 97-99% (Within range) |
| Respiratory rate | 14 brpm | 12-16 brpm (Within range) |
| Score | **91 / 100** | Better than 99% — log best |

### HRV progression

| Date | HRV | Context |
|------|-----|---------|
| Sep 6 | 74ms | 0% beers ×4, home 01:11, residual illness |
| Sep 7 | **32ms** | Heavy night, ~4h sleep — near all-time low (chat-reported) |
| Sep 8 | **145ms** 🚀 | Sober, bed 22:59 (pre-midnight), 8h40 — ties all-time high |
| Sep 9 | **127ms** | Sober, bed 00:25, 7h37 — late-bedtime settle |
| **Sep 10** | **123ms** | Sober, bed 00:10, 8h26 — best score yet, HRV ceiling = bedtime |

## Connected

- [[Sleep/2026-09-09]] (the 127ms this settles from — same late-bedtime pattern)
- [[Sleep/2026-09-08]] (the 145ms pre-midnight record this is measured against)
- [[Runs/2026-09-10]] (the 4.05 km this night powered)