Sleep · Research · 6 min read
Evening exercise and sleep: later, higher-strain workouts were linked to later, shorter sleep
In one year of wrist-worn device data from 14,689 physically active adults, exercise ending closer to habitual sleep onset, especially at higher exercise strain, was associated with later sleep onset, shorter sleep and a higher nocturnal resting heart rate.
Sleep advice has long warned against hard exercise close to bedtime, while recent meta-analyses found evening exercise in general was not linked to shorter or worse sleep. A 2025 study in Nature Communications looked at one year of wrist-worn device data from 14,689 physically active adults, 4,084,354 person-nights in total. Exercise that ended closer to each person's habitual sleep onset, and at higher exercise strain, was associated with later sleep onset, shorter sleep duration, lower sleep quality, a higher nocturnal resting heart rate (RHR) and a lower nocturnal heart rate variability (HRV). Light or moderate exercise ending more than about 2 hours before habitual sleep onset looked similar to non-exercise nights for sleep onset and sleep quality. This was an observational study of physically active device users, mostly men, so it shows associations, not cause and effect. [1]
What the researchers did
This was a large retrospective cohort study reported under the STROBE guidelines for observational studies. The data came from 14,689 subscribers to the WHOOP biometric device platform over one year. To be included, people had to be at least 18, wear the device on 350 or more days (95.6% of possible wear time) and log at least 50 exercise activities between 10 hours before and 2 hours after habitual sleep onset. Mean age was 37.89 years, and 10,845 (73.83%) were men. The final dataset held 4,084,354 person-nights. The authors cite validation of the device against electrocardiography for heart rate (99% agreement) and against polysomnography for 2-stage sleep categorisation (86–89% agreement). [1]
The study used these terms: [1]
- Habitual sleep onset: a person's mean sleep onset for that calendar month, counted separately for weekdays and weekends. Exercise timing was when an exercise session ended, relative to this time.
- Exercise strain: minutes in each heart rate zone, multiplied by the zone number and added up (the summated-heart-rate-zones score, SHRZS), then grouped as light, moderate, high or maximal. The paper's examples: a brisk walk or slow jog (light), a gym class or 5-mile run (moderate), hockey training or a 10-mile run (high), a hockey match or half marathon (maximal).
- Sleep quality: a device measure, the sleep percentage (the sleep period minus wake after sleep onset, divided by the sleep period). It is not a self-rating.
- Nocturnal RHR and HRV: heart rate and heart rate variability (RMSSD) averaged over the sleep period, with more weight on slow wave sleep.
The analysis used generalized additive mixed models. It adjusted for weekday vs. weekend, the prior night's value, general fitness, gender and age, and used p < 0.005 as the threshold for statistical significance. [1]
Key findings
| Measure | Exercise ended 2 h before habitual sleep onset | Exercise ended 2 h after habitual sleep onset |
|---|---|---|
| Sleep onset | 36.0 min later | Over 1 h later (the Results text says 80.0 min; the Table 1 contrasts add up to 70.5 min) |
| Sleep duration | 5.4% (22.2 min) shorter | 13.9% (42.6 min) shorter |
| Sleep quality | 0.87 percentage points lower | 5.6 percentage points lower |
| Nocturnal RHR | 6.8% (3.9 beats/min) higher | 15.0% (9.4 beats/min) higher |
| Nocturnal HRV | 14.1% (8.3 units) lower | 32.6% (14.6 units) lower |
Later and harder went with later sleep onset. When exercise ended 6 or more hours before habitual sleep onset, maximal exercise was associated with earlier sleep onset than lower strain levels and non-exercise days. Closer to habitual sleep onset, the pattern turned around. [1]
“When exercise concluded 4-h before to 2-h after habitual sleep onset, sleep onset became progressively later for all levels of strain and was more pronounced when strain was higher.”
Leota et al. · Results · full text
Sleep duration and sleep quality followed a similar pattern. When exercise ended 6 or more hours before habitual sleep onset, higher strain went with longer sleep. From 6 hours before to 2 hours after habitual sleep onset, sleep duration was progressively shorter for all strain levels, more so at higher strain. Sleep quality was progressively lower when exercise ended from 2 hours before to 2 hours after habitual sleep onset. [1]
Overnight heart measures moved with strain at every exercise time studied. All strain levels were associated with a higher nocturnal RHR and a lower nocturnal HRV than non-exercise days. Exercise ending 6 to 10 hours before habitual sleep onset went with a slightly lower HRV than non-exercise days (a 1.56–4.52 unit decrease), which the authors describe as minimal. [1]
“Regardless of when exercise concluded, higher strain was associated with higher RHR.”
Leota et al. · Results · full text
Lighter sessions earlier in the evening looked like rest nights. The authors say this fits recent meta-analyses that found no effect of evening exercise on sleep, and note that 82% of the studies in those meta-analyses would count as light or moderate strain. [1]
“When light or moderate strain exercise ended more than ~2-h before habitual sleep onset, sleep onset and sleep quality were similar to non-exercise nights.”
Leota et al. · Discussion · full text
The authors' suggestion. A secondary analysis that timed exercise against each night's actual sleep onset gave consistent results. Exploratory analyses found the associations did not differ by gender, age or BMI status. Based on this, the authors suggest a gap of four or more hours, and lighter strain exercise when evening exercise can't be avoided. [1]
“Our findings suggest that to optimize sleep timing, duration, and quality, when possible, individuals should stop exercising four or more hours before their habitual—or actual—sleep onset.”
Leota et al. · Discussion · full text
What this doesn't tell us
This is observational data. It shows how exercise timing and strain went together with sleep, not that the exercise changed the sleep. People who train hard late at night may differ on those days in ways the data did not capture. The authors could not control for some factors known to affect sleep, such as bright light exposure, and they mention food, caffeine and alcohol around exercise as possible pathways that were not part of the analysis. [1]
The sample was physically active device users, and 73.83% were men. The authors write that the sample is not representative across the spectrum of physical activity levels. The paper does not report separate results for people with insomnia, so the results may not apply to them. [1]
Sleep was measured by the device, and sleep quality here means the device's sleep percentage. The device could not tell when someone first tried to fall asleep. [1]
“The biometric device did not detect when sleep was first attempted; therefore, we were unable to estimate the time participants began to attempt sleep or calculate sleep onset latency.”
Leota et al. · Discussion · full text
The heart-rate-zone method may underestimate the strain of strength training with very brief high-effort movements, so strength training may be under-counted. Yoga, stretching, meditation and massage therapy were excluded. [1]
Two authors (E.R.C. and D.M.P.) are affiliated with WHOOP, Inc., which paid their salaries. The paper says the company did not otherwise play a role in the study design, data collection or analysis, decision to publish, or preparation of the manuscript. [1]
What to take away
Opinion, not a study result: for someone who can only exercise in the evening, these data fit with keeping the hardest sessions further from habitual sleep onset and choosing lighter ones close to it. The data come from physically active, mostly male device users and show associations only, so results may differ from person to person. If sleep problems persist, talk to a doctor.
Reference
- Nat Commun 2025[1] Leota J, Presby DM, Le F, Czeisler MÉ, Mascaro L, Capodilupo ER, Wiley JF, Drummond SPA, Rajaratnam SMW, Facer-Childs ER. Dose-response relationship between evening exercise and sleep. Nature Communications. 2025;16:3297.
DOI: 10.1038/s41467-025-58271-x · Full text
This is a research explainer for general education, not medical advice.
