Experiment guides → Timing → Exercise timing
Is exercise before bed bad for sleep?
Quick answer: less bad than the old rule claimed, and more nuanced than the new one. Two meta-analyses of lab studies found evening exercise didn't harm sleep — if anything it slightly increased deep sleep — with one exception: vigorous exercise ending within about an hour of bedtime delayed sleep onset and lowered efficiency. Then a 2025 analysis of 4 million nights from 14,689 wearable users found a dose-response: the harder and the closer to bedtime you trained, the later you fell asleep, the less you slept, and the higher your night-time heart rate — with effects visible out to four hours before bed. Your own cutoff sits somewhere between those findings, and a 14-day test will locate it.
"I will finish any workout at least 4 hours before bed for 14 days."
What the research says
Lab meta-analysis: evening exercise is fine, with one exception. Stutz and colleagues pooled 23 studies of exercise in the evening. Overall, evening exercise didn't impair sleep and slightly increased slow-wave sleep. The exception was vigorous exercise ending less than an hour before bed, which lengthened sleep onset and reduced sleep efficiency (Stutz et al., 2019).
High-intensity evening exercise: same conclusion. Frimpong's meta-analysis focused on high-intensity sessions. Performed in the early evening — two to four hours before bed — they didn't disrupt sleep and increased deep sleep, at the cost of slightly less REM. Sessions ending closer to bedtime were where sleep onset suffered (Frimpong et al., 2021).
Real life, 4 million nights: the closer and harder, the worse. Leota and colleagues analysed 4 million nights from 14,689 people wearing a fitness tracker. Exercise within four hours of bedtime was associated, dose-dependently, with later sleep onset, shorter sleep, lower sleep quality, higher resting heart rate and lower heart-rate variability overnight — and higher strain amplified every effect. Lab studies use young, fit volunteers on a schedule; this is what happens at home (Leota et al., 2025).
What this evidence doesn't say
The lab meta-analyses come from small studies in young, fit people. The wearable data are observational — people who train late may also eat late, drink, or be stressed — and tracker-derived sleep is imperfect. "Vigorous" is relative to your fitness. Nobody has run a randomised trial of training-time cutoffs in ordinary adults, so the four-hour number is a hypothesis for you to test, not a rule.
Run it as an experiment
- Baseline five nights with your usual training time. Log when you finished, how hard it was (easy / moderate / hard), and the morning ratings: sleep quality, restedness, time-to-sleep.
- Move the session earlier, keep it the same. Same workout, same intensity, ending at least four hours before bed — or morning, if that's the only way. Don't also make it easier; that's a second variable.
- Make the pact. "I will finish any workout at least 4 hours before bed for 14 days." If a late session is unavoidable, do it easy and note it — an honest exception is data.
- Morning check-ins. The three ratings, plus resting heart rate if your watch gives it — the wearable study's clearest signal was elevated overnight heart rate after late training.
- Compare, then narrow the window if you like. Better sleep onset and restedness? Late training was costing you. No difference? You're one of the people the lab studies describe — train when you want. Curious about two hours? Run a second block.
What to expect
For people who train hard in the last couple of hours before bed, the change is often quick: falling asleep sooner, a calmer heart rate on the tracker, less of that wired-but-tired feeling. For those who train gently or earlier, expect nothing — the meta-analyses predict exactly that. Don't be surprised if morning sessions leave you sleepier at night in a good way; that's the other half of the exercise–sleep relationship.
Who should skip this
If evening is the only time you can train, keep training — regular exercise improves sleep overall, and the meta-analyses say moderate evening sessions are fine. Test the cutoff only if you can actually move the session. If you have heart disease or take beta-blockers, the heart-rate notes will be harder to interpret; focus on how you feel.
Frequently asked questions
How many hours before bed should you stop exercising?
Lab studies flag the last hour, for vigorous work. Wearable data suggest effects out to four hours, scaling with intensity. A four-hour cutoff is the conservative test; if it changes nothing for you, you can move it closer in a second block.
Is it OK to work out at night?
For most people, yes — evening exercise didn't harm sleep in pooled lab studies and slightly increased deep sleep. The caveat is hard sessions close to bedtime. Easy or moderate evening exercise is well within the evidence.
Does exercising before bed help or hurt sleep?
Both, depending on timing and intensity: earlier in the evening it tends to help (more deep sleep), very late and hard it tends to hurt (later onset, higher heart rate). Which side you're on is what the two-week test tells you.
Related experiments
Sources
- Stutz J, Eiholzer R, Spengler CM. Effects of evening exercise on sleep in healthy participants: a systematic review and meta-analysis. Sports Med. 2019;49(2):269–287. PubMed 30374942
- Frimpong E, Mograss M, Zvionow T, Dang-Vu TT. The effects of evening high-intensity exercise on sleep in healthy adults: a systematic review and meta-analysis. Sleep Med Rev. 2021;60:101535. PubMed 34416428
- Leota J, Presby DM, Le F, et al. Dose-response relationship between evening exercise and sleep. Nat Commun. 2025;16:3297. PubMed 40234380
Track this experiment in N of 1
A few baseline nights, then the change — and your own mornings deliver the result. No streaks, no guilt. Your first experiment is free.
Downloadon the App StoreThis page is general information about published research, not medical advice. If you have a sleep disorder, are pregnant, or take medication, talk to your doctor before changing your routine.