How does sleep affect athletic performance? 1.7× injuries, 5% speed

Athletes track load to the kilometre and to the kilogram, then treat sleep as whatever is left over after the day. The evidence says sleep is the largest recovery input you have, and the one most training plans never see. Two studies put numbers on it, and the numbers are the same size as a training block.

How much does short sleep raise injury risk?

Milewski and colleagues followed 112 adolescent athletes for 21 months. Those sleeping under eight hours were 1.7 times more likely to be injured than those sleeping eight or more, and hours of sleep was one of the two best independent predictors of injury in the whole dataset, ahead of training volume.

Adults are not adolescents, but the mechanism, slower reaction time and poorer motor control on short sleep, does not retire at twenty. The exposure measured was habitual sleep, not one bad night; what raised the risk was the pattern.

Does more sleep improve performance?

The Stanford basketball study went the other way: instead of measuring the cost of short sleep, Cheri Mah asked eleven varsity players to sleep as much as they could for five to seven weeks. Free-throw accuracy rose 9%, three-point accuracy 9.5%, and sprint times fell about 5%. No change in training. Just more nights of the thing training is supposed to be converted by.

Five percent on a sprint is the difference between a block of speed work and none. It is a training effect, delivered by a pillow. Eleven players is a small study and they were sleeping more than most adults can, so treat the exact figures as a demonstration of direction and size rather than a promise.

PeakU does this from your own data. A coach that reads everything you already track. Join the free beta →

What does sleep loss cost first?

Not everything degrades at the same rate. Maximal strength holds up surprisingly well for one night. What goes first is everything that depends on the central nervous system staying sharp: reaction time, time to exhaustion, perceived effort, technical skill and the judgement to stop a set before form fails. That is why a short night costs you an interval session or a technical lift more than it costs you a heavy single at a familiar weight.

The practical answer to the next morning's question lives in a separate guide: how much to train after a bad night of sleep. The short version is that one night is noise and three is a trend, and the right move is usually to swap intensity for easy volume rather than to cancel.

What counts as a short night?

Not "under eight hours". Under your normal. Someone whose average is six and a half hours has a different short night from someone whose average is eight, and the body responds to the deviation, not the absolute. An hour below your average is noise. Two hours below, or under about six hours in absolute terms, is a short night worth noting.

Timing matters as much as length. A short night before an easy day costs almost nothing. The same night before the week's one quality session costs the session, because the capacities that go first, reaction, form under fatigue, perceived effort, are the ones a hard session depends on. That is why the rule in the next section protects the nights before the sessions that matter rather than chasing a nightly target.

Why do training plans ignore sleep?

Because plans are written in advance and sleep happens afterwards. A block that prescribes threshold on Tuesday cannot know that Monday night was five hours. The runner either does the session badly, or skips it and feels they failed the plan. Neither is what the plan intended.

The fix is a plan that reads last night before it commits to today. That means the coach needs the sleep data, needs a baseline for what your normal night is, and needs the authority to move the session. A readiness score is the usual way that gets packaged; how a training readiness score is built shows where sleep sits in it.

How do you periodise sleep?

You do not periodise sleep the way you periodise mileage, because you cannot bank it in advance. What you can do is treat it as a variable the plan responds to, with rules written down before the week starts.

• Know your normal. A rolling average over two weeks, from whatever device you wear. The target is your average, not eight hours.
• One short night changes nothing. Train as planned; cap the hard part if you feel it.
• Two or three in a row move quality work. Swap the session for easy volume of the same length; put the hard session on the first day after a normal night.
• Protect the nights before the sessions that matter. If the long run is Sunday, Saturday night is part of the long run.
• Watch sleep when training gets harder. Sleep that gets worse as load climbs is one of the signs worth watching for overreaching.

Sleep also moves the numbers people care about outside training. It is one of the inputs to any fitness-age estimate, which is why lowering your fitness age is partly a sleep project.

How PeakU periodises it

PeakU pulls sleep duration and sleep score from whatever wrote them to Apple Health or Health Connect, scores them against your own baseline, and lets them move the day's call. Two short nights in a row pull quality work later in the week and swap in easy volume; a run of good nights lets the plan hold its ramp. The weekly review names sleep explicitly when it was the thing that slipped, because a plan that absorbs your bad nights without saying so teaches you nothing.

Hen Eytan · Founder, UPeak

Builds the UPeak apps and writes up the research behind them.