If you are worried about your testosterone, the first question worth answering is not about supplements or training programmes. It is how you slept this week.
That is not a wellness cliché. Testosterone is one of the few hormones whose production is tightly welded to the sleep-wake cycle, and the research on sleep loss and testosterone goes back decades. The honest version of the story is more nuanced than either the "sleep fixes your hormones" crowd or the sceptics admit. Some studies show large drops. Others show nothing. The differences turn out to be informative.
TL;DR: Most of the daily testosterone release in men happens during sleep, and testosterone follows a circadian rhythm that peaks in the early morning. A landmark laboratory study put 10 healthy young men on 5 hours of sleep per night for a week and found daytime testosterone fell by 10 to 15%, with the biggest gap in the afternoon and evening — a change the authors compared with roughly a decade of normal ageing, and one that was not explained by cortisol. A later meta-analysis of the whole literature found that total sleep deprivation (24 hours or more awake) reliably lowers male testosterone, while short-term partial sleep restriction does not have a consistent significant effect. In other words, duration matters less than the extreme end of sleep loss — and it is not the only input. Sleep timing matters too: in around 2,000 US men, night owls had lower morning testosterone than early risers, roughly 28 ng/dL lower, and the difference held even when the blood draws were shifted. Obstructive sleep apnea is associated with low testosterone, though a 2014 review argues the direct effect is smaller than commonly claimed once age and obesity are accounted for. Practical takeaway: aim for 7+ hours on a regular schedule, treat snoring seriously, do not panic over one bad night, and remember that no consumer wearable measures testosterone — it measures the sleep that shapes it.
YouTube: Related video
The basic biology, in plain language
Testosterone production is not spread evenly across 24 hours. In men, the largest share of the daily release happens during sleep, and circulating levels follow a rhythm that rises overnight and typically peaks in the early morning around waking. That is exactly why blood tests for testosterone are supposed to be taken in the morning: you are sampling the top of the daily curve.
Two things about that rhythm are worth holding onto.
First, it is a rhythm, not a reservoir. A single day of short sleep does not empty your testosterone stores. It blunts the overnight rise and shifts the daytime curve downward — most noticeably in the afternoon and evening, when the difference between a well-slept and a sleep-deprived day is easiest to detect.
Second, the system that controls it — the hypothalamus, pituitary and testes working together — is sensitive to more than just hours in bed. Sleep fragmentation, sleep apnea, shift work, obesity, opioids, alcohol and illness all interact with it. That is why the sleep-testosterone literature is full of results that seem to contradict each other.
What the laboratory studies found
The most cited study on this topic is a 2011 JAMA research letter by Rachel Leproult and Eve Van Cauter at the University of Chicago. Ten healthy young men (average age 24) were studied in a laboratory. After a week of 8-hour bedtimes at home, they spent three nights with 10 hours in bed (the rested condition), followed by eight nights restricted to 5 hours in bed.
The results were blunt. Total sleep time fell from about 8 hours 55 minutes to about 4 hours 48 minutes. During the waking hours common to both conditions, daytime testosterone was lower after sleep restriction — 16.5 versus 18.4 nmol/L — and the gap widened in the afternoon and evening, 15.5 versus 17.9 nmol/L. That is a decrease of roughly 10 to 15%. For context, the authors noted that normal ageing is associated with a decline of about 1 to 2% per year, so a week of 5-hour nights moved the number in the direction and magnitude of about a decade of ageing. They also reported something important for the mechanism: cortisol profiles were similar in both conditions, so this was not simply a stress response.
Participants' self-rated vigour also dropped steadily across the restricted week, from 28 after the first night to 19 after the seventh.
It is a small study — ten men — and a short one. It should not be over-read. But it was tightly controlled, with blood sampled every 15 to 30 minutes across 24 hours, which is why it keeps showing up in reviews.
The meta-analysis that complicates the story
Here is where the honest version diverges from the internet version.
A 2022 systematic review and meta-analysis in Sleep Medicine Reviews pooled the experimental evidence on partial and total sleep deprivation and serum testosterone in healthy men. Its conclusion: total sleep deprivation of 24 hours or more reduces male testosterone levels, while short-term partial sleep deprivation has no significant effect.
That is a genuinely useful finding, because it tells you which kind of sleep loss actually shows up in the blood.
- Extreme, acute sleep loss (a full night awake) shows up reliably. The overnight testosterone pulse is missed entirely, and the daytime curve drops.
- Several nights of moderately shortened sleep do not consistently move the needle in pooled analyses. Individual studies go both ways: a 2019 randomized crossover study in Sleep Health found no significant effect of experimental sleep restriction on testosterone in one protocol and only a trend in the other.
In fairness, the partial-restriction trials are usually shorter and milder than a full week of 5-hour nights, testosterone is a noisy measurement with wide between-person variability, and small trials have limited power — the change in the Sleep Health study came with standard deviations far larger than the effect itself. So "no consistent significant effect" is not the same as "sleep duration does not matter". It means the effect of short-term, moderate restriction is too small to detect reliably in the studies we have.
If your sleep has been poor for months, you are further from the experimental conditions than you think. There is also a long-run observational signal to consider.
The observational picture: duration, timing and consistency
Duration. In a cross-sectional study of 1,274 older men, testosterone rose with sleep duration up to roughly 9.9 hours and then fell — an inverted U shape. Another study in older men found longer sleep duration was associated with higher testosterone specifically in men with a BMI under 25. These are associations, not proof of cause, but they line up with the experimental direction of travel and they suggest that both short and very long sleep sit on a less favourable part of the curve.
Timing. In an analysis of nearly 2,000 US men presented at the American Urological Association annual meeting, night owls had significantly lower morning testosterone than early risers — about 28 ng/dL lower — after adjusting for age, weight, diabetes and exercise. The researchers then tested whether this was simply a matter of when the blood was drawn: comparing early risers tested in the morning with night owls tested in the afternoon (when their body clock would predict a peak), the association still held. If that finding replicates, it suggests that a delayed sleep phase is doing something independent of sleep duration.
For anyone who has read our breakdown of why sleep consistency often matters more than total duration, this is the same pattern in a different hormone. Regularity is not just a sleep-hygiene slogan.
Sleep apnea. Obstructive sleep apnea is consistently associated with low testosterone in men, and it is one of the reasons heavy snoring is worth investigating rather than ignoring. But the causal story is less clear than it is often presented. A 2014 review by Gary Wittert in the Asian Journal of Andrology concluded that obstructive sleep apnea appears to have little direct effect on testosterone once age and obesity are accounted for, and that much of the association is explained by the metabolic factors that travel with apnea — obesity in particular. Other work has found improvements in testosterone after treatment with CPAP in some men.
The practical reading: if you snore heavily, wake unrefreshed, or have been told you stop breathing at night, get it assessed. Not because apnea is a simple testosterone switch, but because it wrecks sleep quality, and because the same metabolic picture that produces apnea is the one that suppresses testosterone.
Our guide to what snoring actually means covers the red flags worth taking to a doctor.
What actually moves the needle
If you want to take this seriously without going down a supplement rabbit hole, the evidence points to a short list.
- Sleep 7 or more hours, regularly. The joint consensus statement from the American Academy of Sleep Medicine and the Sleep Research Society recommends adults sleep 7 or more hours per night on a regular basis, and flags less than 7 as associated with a long list of adverse outcomes. The "regularly" is doing real work in that sentence.
- Keep the timing stable. Bedtime and wake time within a narrow window, weekdays and weekends included. The chronotype data above suggests timing itself matters, independent of total duration.
- Do not treat one bad night as a hormonal event. The experimental evidence says moderate short-term restriction does not reliably change testosterone. Chronic restriction, total deprivation and untreated sleep disorders are the situations with real signal.
- Take snoring and daytime sleepiness seriously. A sleep study is cheap compared with years of unexplained fatigue. Ask about it specifically rather than hoping the topic comes up.
- Look at the whole picture before blaming sleep. Age is the biggest single factor. Weight, alcohol intake, opioids, thyroid disease, severe illness and very high volumes of endurance training all affect testosterone. A sleep intervention will not fix a picture driven mainly by something else.
- If you get tested, get tested properly. Morning draw, fasting, and ideally two separate measurements, because day-to-day variability in testosterone is substantial. If you have symptoms — persistent low libido, fatigue, erectile changes, loss of muscle — talk to a doctor rather than a supplement shop.
What your wearable can and cannot tell you
Let me be direct: no consumer watch, ring or band measures testosterone. Anyone claiming otherwise is extrapolating. There is no optical sensor for steroid hormones.
What your device does measure, and what is genuinely relevant here, is the input side of the equation:
- Sleep duration and time in bed. Inconsistent with your own report more often than people expect.
- Sleep consistency. Bedtime variability across a week is one of the most useful numbers your watch produces, and it maps directly onto the chronotype and circadian findings above.
- Sleeping heart rate and HRV. These reflect autonomic recovery, which is where sleep loss shows up first. If your overnight HRV trends down while your sleeping heart rate trends up, you are accumulating load — hormonal consequences usually follow that pattern rather than lead it.
- Respiratory rate and blood oxygen. Not diagnostic, but persistent changes are a reason to ask about sleep apnea.
- Wrist temperature. Useful for cycle tracking and for spotting illness, and a signal about how disrupted your nights really are.
The right way to use this is as a trend, not a verdict. One night tells you almost nothing; four weeks of consistently short, inconsistently timed sleep tells you something worth acting on. That is the same logic behind why a wearable trend beats a single daily score, and it is what Century AI is built around: a daily health score, recovery score and sleep insight built from the watch you already wear, so you can see whether your nights are actually changing rather than guessing from how you feel on a Tuesday.
The bottom line
Sleep and testosterone are connected, but the connection is not the simple "sleep more, boost your T" story that circulates online. What the evidence supports is narrower and more useful: a week of severe sleep restriction lowers daytime testosterone by 10 to 15% in young men, full nights of total sleep deprivation reliably lower it, and short-term moderate restriction does not show a consistent effect. Sleep timing matters on top of duration, with night owls showing lower morning testosterone in large observational data. Sleep apnea travels with low testosterone, though much of that relationship runs through age and body composition rather than breathing alone. And no wearable measures the hormone itself.
So the sensible approach is unglamorous. Get 7 or more hours, keep the schedule boringly consistent, take snoring seriously, get tested properly if you have symptoms, and use your overnight data to spot the weeks when sleep is quietly falling apart. That is not a hack. It is just the part of the picture you control.
Century AI helps you understand your body with a daily health score, recovery score, and sleep insights — using the watch you already wear.