The short version
- A meta-analysis of 227 outcome measures from 69 publications found sleep loss reduced exercise performance by a mean of 7.56%, with significant effects across every exercise category.
- Consistent negative effects appeared only with total sleep deprivation and with late restriction, meaning waking earlier than normal, while a later bedtime with a normal wake time was less damaging.
- Performance fell roughly 0.4% for every hour spent awake before the task, and morning exercise was largely unaffected while afternoon and evening exercise suffered.
- In 24 men restricted to 4 hours in bed for five nights, myofibrillar protein synthesis fell, but it was maintained at control levels in the restricted group that performed three high-intensity interval sessions.
- The best-known sleep extension study in athletes involved 11 basketball players with no control group, who gained 110.9 minutes of sleep per night and improved sprint time from 16.2 to 15.5 seconds.
Sleep is the recovery intervention people are most confident about and least precise about. The confidence is largely earned. The precision is missing for a specific and slightly awkward reason: it is far easier to run a study that takes sleep away than one that gives it back.
What taking it away costs
Craven and colleagues pooled 227 outcome measures from 69 publications in Sports Medicine, covering anaerobic power, speed endurance, high-intensity intervals, strength, endurance, strength-endurance and skill. Sleep loss reduced exercise performance by a mean of 7.56% (95% CI −11.9 to −3.13, p = 0.001), and the effect was significant for every exercise category.
The detail inside is more useful than the headline. The pattern of sleep loss mattered: consistent negative effects appeared only with total deprivation and with late restriction — waking earlier than normal. Going to bed later than usual, with a normal wake time, did not produce the same consistent damage. Performance dropped roughly 0.4% for every hour awake before the task. And tasks performed in the morning were largely unaffected, while afternoon and evening tasks took the hit.
Two caveats belong with that number. Heterogeneity was 98.1%, which is enormous, and 89% of participants were male.
What is happening in the muscle
Saner and colleagues gave us a mechanism, and an unexpected escape route. Twenty-four healthy young men spent five nights with either 8 hours or 4 hours in bed, with one restricted group also performing three sessions of high-intensity interval exercise. Myofibrillar protein synthesis — the rate at which contractile muscle protein is rebuilt — was lower in the sleep-restricted group.
But in the group that was equally sleep-restricted and exercised, rates were maintained at control levels. Curiously, none of the usual regulators moved: no changes in p-AKT, p-mTOR, or the degradation markers they measured, in any group.
Read practically, that is permission to keep training through a bad week. Twenty-four men is small and five nights at four hours is more extreme than most people's bad week. But the direction is the opposite of the common advice to skip the gym when you are underslept, and it is the only controlled human data pointing either way.
The study everyone quotes about adding sleep
Mah and colleagues asked eleven Stanford varsity basketball players to get as much sleep as possible, aiming for at least ten hours in bed, for five to seven weeks after a baseline period. Nightly sleep rose by 110.9 ± 79.7 minutes. Sprint times improved from 16.2 to 15.5 seconds. Free throw percentage rose 9%, three-point percentage 9.2%. Reaction time, sleepiness and mood all improved.
Those numbers are quoted constantly, usually as though they came from a trial. They did not. Eleven players, no control group, no randomisation, measured across a competitive season in which practice alone would be expected to improve shooting. It is a genuinely interesting observation and it is the weakest study design in this article.
That is the asymmetry. The cost of losing sleep is measured across 69 publications. The benefit of gaining it rests substantially on eleven basketball players.
So what should you take from this?
Protect the wake time above all. Late restriction was one of the two damaging patterns; a later bedtime with a normal wake time was not consistently harmful. If a night is going to be short, let it be short at the front end.
Move hard sessions to the morning after a bad night. Morning tasks were largely unaffected in the pooled data while afternoon and evening tasks suffered, and every hour awake beforehand cost about another 0.4%.
Train anyway. The one mechanistic study available suggests exercise preserved muscle protein synthesis during severe restriction. Lower the ambition, keep the session.
And keep the claim honest when you talk about it. Sleep is the largest lever in recovery mostly because of what its absence costs, not because anyone has demonstrated that adding an hour to an already-adequate night makes you recover better. That is a real distinction and most of the category ignores it.
Where a supplement honestly fits
Narrowly. We sell magnesium and the case for it is about ordinary nutritional adequacy, not about buying performance — the general argument on magnesium and sleep is one we have made elsewhere and we are not going to relitigate or inflate it here. If your intake is genuinely low, correcting that is sensible. If you are sleeping six hours because of your calendar, no capsule fixes a scheduling problem.
The rule-out matters more than the purchase. Loud snoring with daytime sleepiness, waking unrefreshed after adequate hours, or months of early waking are questions for a clinician — sleep apnoea, thyroid disease, iron deficiency and depression all present this way, and a supplement aimed at any of them is a delay dressed as a solution. If a deload and a fixed wake time do not restore you, that is the point where overtraining syndrome and the medical causes that imitate it deserve a proper look.
Good questions
How much does bad sleep affect training performance?
About 7.56% on average, according to a meta-analysis of 227 outcome measures, with significant reductions across anaerobic power, strength, endurance and skill. The variability between studies was very high, so treat that as an order of magnitude rather than a precise figure. It is still one of the largest single effects anywhere in the recovery literature, and it costs nothing to avoid.
Does one bad night actually cost me muscle?
Probably not. The only controlled human study on the mechanism found that muscle protein synthesis dropped during five nights of severe sleep restriction but was preserved in the group that did high-intensity intervals during the same period. Lower the ambition and keep the session, ideally in the morning, since morning performance was largely unaffected by sleep loss in the pooled data.
Is it better to go to bed late or wake up early?
Go to bed late. In the pooled analysis, consistent negative effects on performance appeared with total deprivation and with late restriction — waking earlier than normal — while delayed bedtimes with a normal wake time were not consistently damaging. If a night has to be short, protect the morning end of it. That is a genuinely actionable finding and it contradicts the usual advice.
Does sleeping more than usual improve performance?
Possibly, but the evidence is much weaker than for the reverse. The most quoted sleep extension study had 11 basketball players, no control group and no randomisation, across a season where practice alone would improve shooting. Adding sleep when you are short is clearly sensible. Adding an hour to an already adequate night has not been shown to make you recover better.
Will magnesium help me recover by improving my sleep?
That is a bigger claim than we would make. Magnesium's case rests on nutritional adequacy — correcting a low intake — rather than on buying performance, and treating a supplement as a fix for a short night misses where the problem is. If you are sleeping six hours because of your schedule, nothing in a capsule changes that arithmetic, and we would rather say so than sell into it.
When should I see a doctor about my sleep?
When the hours are adequate and you still wake unrefreshed, when you snore loudly and are sleepy in the day, or when early waking has run for months. Sleep apnoea, thyroid disease, iron deficiency and depression all present this way and all are treatable. Reaching for a sleep supplement in those situations delays a diagnosis, which is the more expensive mistake.
Sources
- Craven J, McCartney D, Desbrow B, Sabapathy S, Bellinger P, Roberts L, Irwin C. Effects of Acute Sleep Loss on Physical Performance: A Systematic and Meta-Analytical Review. Sports Med, 2022. View study
- Saner NJ, Lee MJ, Pitchford NW, Kuang J, Roach GD, Garnham A, et al. The effect of sleep restriction, with or without high-intensity interval exercise, on myofibrillar protein synthesis in healthy young men. J Physiol, 2020. View study
- Mah CD, Mah KE, Kezirian EJ, Dement WC. The effects of sleep extension on the athletic performance of collegiate basketball players. Sleep, 2011. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
