Search “best exercise for sleep,” and you’ll usually find the same recommendation repeated: do moderate cardio consistently and your sleep should improve.

That advice isn’t wrong. Aerobic exercise has a strong evidence base behind it, particularly for subjective sleep quality, mood regulation, and stress reduction. Walking, cycling, swimming, and jogging are all associated with better sleep outcomes across multiple reviews and meta-analyses.

The problem is that most mainstream sleep advice understates resistance training and oversimplifies the broader evidence.

Recent comparative studies suggest different exercise types may improve different dimensions of sleep. Aerobic exercise appears especially effective for stress regulation and perceived sleep quality, while resistance training often performs better on sleep efficiency, sleep duration, and sleep onset latency.

The largest recent paper on the subject — Li et al. 2024 — compared multiple exercise modalities against one another rather than simply asking whether exercise helps sleep at all. Its conclusions challenged several assumptions that dominate the current search landscape.

In this peer-reviewed study, combined aerobic-and-resistance training ranked highest overall for sleep quality improvement. Resistance training alone ranked surprisingly well, while Tai Chi performed comparably to aerobic exercise on some subjective sleep measures.

That’s not to suggest cardio “doesn’t work.” It just means the relationship between exercise and sleep is more nuanced than the standard advice suggests.

Different forms of training appear to influence different sleep mechanisms, including circadian regulation, autonomic balance, recovery pressure, and sleep architecture. But across nearly every study reviewed, one finding remained remarkably consistent: people who exercise regularly tend to sleep better than people who remain sedentary.

What 58 RCTs ranked first — and why

The Li 2024 network meta-analysis used SUCRA rankings, a statistical model estimating the probability that an intervention is the most effective among competing treatments.

Combined exercise — programs pairing aerobic and resistance training — ranked first overall for sleep quality improvement, with a standardized mean difference (SMD) of −0.99. Resistance training, Tai Chi, and aerobic exercise all improved sleep as well, but combined protocols produced the broadest overall effects across the dataset.

The rankings surprised many researchers because most public-facing sleep advice still centers on cardio.

Part of the explanation may be that different studies measure different sleep outcomes. Some focus heavily on subjective sleep-quality questionnaires like the Pittsburgh Sleep Quality Index (PSQI), while others examine sleep latency, fragmentation, efficiency, or objective polysomnography findings.

Those distinctions matter because exercise doesn’t appear to improve every aspect of sleep in the same way.

Aerobic exercise seems to influence the signaling side of sleep regulation more directly. Moderate cardio can affect melatonin timing, serotonin activity, mood regulation, and cortisol balance — all factors tied to stress levels and the body’s readiness for sleep. Resistance training may work differently. Strength-focused exercise appears to exert stronger effects on recovery pressure, sleep continuity, and slow-wave sleep physiology through mechanisms involving autonomic recovery and BDNF-related neuroplasticity.

The combined-training result makes intuitive sense through that lens. Cardio and resistance work appear to affect overlapping but non-identical pathways. When paired together, the physiological effects may complement one another.

At the same time, the Li 2024 paper has limitations worth acknowledging. Many studies relied on subjective PSQI scoring, some exercise categories were unevenly represented, and SUCRA rankings estimate probability rather than certainty.

The practical takeaway here is not that combined training “wins” in every situation. It’s that sleep appears to respond well to programs that challenge multiple physiological systems at once: cardiovascular fitness, muscular adaptation, stress regulation, and recovery capacity included.

The aerobic vs. resistance debate is closer than you think

Mainstream fitness advice still tends to frame aerobic exercise as the default recommendation for better sleep.

However, the comparative literature is more divided than mainstream fitness advice often suggests.

Aerobic exercise does not consistently outperform resistance training for sleep outcomes. In several comparative studies, strength-focused exercise matched — and sometimes exceeded — aerobic training on measures like sleep duration, sleep efficiency, and sleep onset latency, particularly in older adults and overweight populations.

A network meta-analysis focused on older adults with sleep disturbance found resistance training produced larger improvements in sleep efficiency and duration in several populations. Similar findings appeared in hypertensive and overweight adults, where strength-focused programs often improved sleep continuity more effectively than cardio alone.

Aerobic exercise still has one of the deepest evidence bases in the sleep literature. It consistently improves subjective sleep quality, daytime energy, and mood regulation.

Resistance training may affect sleep differently.

Strength work increases neuromuscular recovery demand and homeostatic sleep pressure — the biological drive for sleep that builds across the day. Some researchers also suspect resistance exercise influences slow-wave sleep more directly through autonomic recovery and BDNF-related pathways.

This may help explain why resistance exercise typically performs especially well on fragmentation-related sleep metrics.

Researchers are especially interested in how resistance training may affect slow-wave sleep, sometimes called deep sleep. Slow-wave sleep is tied to physical recovery, immune function, memory consolidation, and glymphatic clearance in the brain. Several comparative studies suggest resistance and combined training may influence deep sleep architecture more strongly than aerobic exercise alone, although the evidence remains preliminary.

The evidence does not support a simplistic “cardio versus weights” conclusion. Different exercise types appear to target different dimensions of sleep. Aerobic exercise may improve the signaling side of sleep regulation, while resistance training may exert stronger effects on sleep continuity and architecture.

For many people, that’s probably why combined training ranks highest overall.

When combined exercise is the answer

Combined exercise appears especially effective in populations where sleep problems overlap with broader metabolic or cardiovascular issues.

In adults with obesity and obstructive sleep apnea, combined aerobic-and-resistance programs improved sleep quality alongside body composition and cardiometabolic markers. Similar findings appeared in sedentary adults and overweight populations.

But combined training is not universally superior.

Some insomnia-focused studies found aerobic exercise alone performing better on subjective sleep-quality measures, while certain older-adult populations responded particularly well to resistance training.

That variability matters because population-level rankings don’t always translate neatly to individuals.

For example, someone struggling with stress-related insomnia may respond extremely well to moderate aerobic exercise or mind-body movement like yoga or Tai Chi. Another person dealing with fragmented sleep and sedentary muscle loss may benefit more from resistance work.

Preference also matters more than many optimization-focused discussions admit.

Any theoretically ideal training protocol loses much of its value if consistency collapses after three weeks. A simpler routine performed steadily for two months is far more likely to outperform a more ambitious combined program that never becomes sustainable.

Lower-barrier options also deserve serious attention. Walking, yoga, Tai Chi, and light resistance circuits may not produce the largest effect sizes across every dataset, but they remain accessible interventions with a lot of legitimate evidence behind them.

The intensity question: high vs. moderate

One of the more controversial findings in the Li 2024 analysis involved exercise intensity.

High-intensity exercise ranked first by SUCRA score for sleep improvement, ahead of moderate-intensity protocols.

The broader research landscape is less definitive.

Many comparative studies still find moderate-intensity exercise produces more reliable sleep improvements across wider populations, particularly among people with stress-related sleep problems or low baseline fitness. Vigorous late-evening training can also increase physiological arousal close to bedtime, sometimes lengthening sleep latency or reducing sleep efficiency in sensitive individuals.

Both findings can be true simultaneously.

High-intensity training may produce stronger improvements when programmed appropriately, but moderate-intensity exercise appears easier to tolerate and sustain across diverse populations. The clearer pattern in the broader literature is that regular exercise matters more than chasing the single “optimal” intensity.

The dose: frequency, duration, and weeks

The Li 2024 analysis also examined exercise dosage variables, including frequency, session length, and intervention duration.

The highest-ranked programs tended to follow a surprisingly manageable structure:

• about four sessions per week
• sessions lasting 30 minutes or less
• sustained for roughly 9–10 weeks

That final number may be the most important finding in the paper.

Many people expect exercise to improve sleep immediately. But such expectations can become a problem. In several intervention studies, the largest sleep-quality improvements did not appear during the first few weeks of training. The more meaningful changes often emerged closer to weeks 9 and 10, precisely when many people begin skipping workouts or abandoning routines entirely. Sleep adaptation appears to follow a slower curve than improvements in mood or daytime energy.

Some studies do show small early benefits, particularly for stress reduction and mood, but the larger sleep-quality gains generally appeared later in the intervention period.

The session-length findings were also counterintuitive. Programs lasting 40–55 minutes often performed worse than shorter sessions in the meta-analysis. One possible explanation involves cortisol and recovery demand. Longer workouts are not automatically better for sleep, particularly when recovery capacity is limited.

For most people, the evidence points toward a relatively sustainable target: moderate exercise volume, consistent weekly frequency, and enough recovery to avoid chronic physiological stress.

What this means for your workout this week

If you want the simplest evidence-based takeaway from the current research, it looks something like this:

Aim for a mix of aerobic and resistance training about four times per week, keep most sessions around 30 minutes, and maintain the routine for at least two months before judging the results.

That doesn’t require an elaborate fitness plan.

A practical week could include brisk walking or cycling on some days and basic resistance work on others using bodyweight exercises, machines, resistance bands, or free weights.

If combined training is unrealistic, the fallback hierarchy still looks encouraging:

• resistance training
• aerobic exercise
• mind-body movement like yoga or Tai Chi
• regular walking

All four approaches have demonstrated measurable sleep benefits in clinical studies, even though some produce larger effects than others.

Walking deserves special mention because it remains one of the most accessible interventions in the literature. Its effects are generally smaller than structured combined programs, but daily walking is consistently associated with better sleep quality and lower sleep disturbance in sedentary adults.

That makes walking especially valuable for people restarting exercise after long sedentary periods, recovering from injury, or managing low energy levels. A lower-intensity routine that actually becomes sustainable may produce larger long-term sleep benefits than an ambitious program that quickly collapses from fatigue or inconsistency.

Where the BetterSleep wind-down fits

Exercise helps create the physiological conditions for better sleep. The transition into sleep still matters.

For many people, the most effective approach combines regular movement with a consistent wind-down routine that helps shift the nervous system toward parasympathetic recovery. Breathing exercises, low-light routines, guided audio, and sleep tracking can make it easier to identify whether changes in training are actually improving sleep latency, efficiency, and deep sleep patterns.

Caveats — and the short version

The evidence here is strong, but it is not perfectly clean.

Many exercise-and-sleep studies rely heavily on subjective questionnaires like the PSQI rather than full polysomnography, the laboratory-based sleep testing used to measure brain activity, breathing, movement, and sleep stages objectively.

That distinction matters because people often report feeling more rested before large objective sleep changes appear in laboratory measurements. Subjective improvement is still meaningful, but it can make exercise effects appear larger in some studies than they do under polysomnography or actigraphy, a wearable-based method used to estimate sleep duration, movement, and nighttime awakenings objectively.

Individual responses also vary substantially. The population-level “best exercise for sleep” may not be the best option for your own schedule, recovery capacity, fitness level, or underlying health conditions.

The broader pattern, though, is remarkably consistent.

• Regular exercise improves sleep.
• Combined training appears to produce the strongest overall effects.
• Resistance training is probably more underrated for sleep than most mainstream advice suggests.

And consistency still matters more than optimization obsession.

These recommendations are aimed primarily at generally healthy adults. People with cardiovascular conditions, obstructive sleep apnea, diagnosed insomnia, recent injuries, or other medical concerns should speak with a healthcare professional before starting high-intensity exercise specifically to improve sleep.

FAQ

What’s the best type of exercise for sleep?

The strongest recent evidence suggests combined aerobic-and-resistance training produces the largest overall sleep improvements across broad populations. Resistance training alone also performs extremely well on sleep duration, sleep efficiency, and sleep latency in several comparative studies.

Is cardio or strength training better for sleep?

The answer is closer to a tie than many fitness articles suggest. Aerobic exercise has a larger historical evidence base, but resistance training often performs equally well — or better — on measures involving sleep continuity and efficiency.

Does yoga help you sleep?

Yes. Mind-body exercises like yoga, Tai Chi, and qigong consistently improve subjective sleep quality, particularly in older adults and people with stress-related sleep problems.

How long does it take for exercise to improve sleep?

Across many intervention studies, the most meaningful sleep improvements tend to appear after about 9–10 weeks of consistent training rather than immediately.

Can exercise increase deep sleep?

Possibly. Several studies suggest resistance training and combined aerobic-and-resistance exercise may have stronger effects on slow-wave sleep, also called deep sleep, than aerobic exercise alone. Slow-wave sleep is associated with physical recovery, immune function, and memory consolidation, although researchers are still studying exactly how different exercise types influence sleep architecture.

The Bottom Line

The current evidence suggests there is no single “perfect” exercise for sleep across every population or sleep problem. Combined aerobic-and-resistance training appears to produce the strongest overall results, but resistance training, aerobic exercise, yoga, Tai Chi, and even regular walking all show measurable sleep benefits in the research.

The more important pattern is consistency. The people who experience the largest improvements in sleep are usually not following the most optimized program. They’re the ones still moving regularly two months later.