Historical interim report / Original release 17 September 2026 / Corrected 24 September 2026

Food-to-Sleep Pilot — first ten outcomes

This dated view covers dinner dates 4–15 September and next-morning scores through 16 September. The current 22-observation release supersedes it.

Dated scope and corrections

The original table transcribed the September 6 score as 76. The source master records 72, with HRV 34 ms. Correcting that record changes the ten-score mean from 67.8 to 67.4 and the nine non-alcohol score mean to 69.8. The September 16 late-mooncake dinner was outside this report’s completed sleep-outcome window; its result appears in the current release.

Ten observed scores

Rows are keyed to the following morning’s sleep date. Dashes mean not recorded.

Sleep date	Prior day activity	Score	Sleep h:mm	HRV ms	RHR bpm
09-05	—	69	—	—	—
09-06	Hike	72	6:06	34	58
09-07	Rest	70	7:05	42	54
09-08	Rest	81	6:59	44	56
09-09	Rucking	46	5:55	25	60
09-10	Run	77	6:07	40	54
09-11	Run	54	5:16	46	51
09-12	Hike	63	6:04	44	53
09-15	Rest	68	7:50	56	55
09-16	Ruck	74	7:45	35	60

Early signals and limits

The single sake night scored 46, with HRV 25 ms and resting heart rate 60 bpm; the next night scored 77, with HRV 40 ms and resting heart rate 54 bpm. This is one natural perturbation and rebound, not a causal effect estimate. Some nights show divergence between total sleep, stage estimates and overnight autonomic measures. Food and activity often changed together, and most reported final-intake times cluster in a narrow early-evening range. These first ten outcomes cannot rank foods or settle a timing effect.

Current dataset

The active September observation set contains 22 linked dates, 21 detailed sleep outcomes, 21 scored nights and reported bedtimes where available. Its methods are in the research framework. This page remains as a corrected, dated interim snapshot.