Sleep and recovery in fight camp

The variable nobody writes into the plan

Start with the document that ought to contain it. The ISSN position stand on nutrition and weight-cut strategies for combat sports is the most complete evidence synthesis the sport has, and its structure is sixteen numbered recommendations. Point 3 sets off-camp weight at 12–15% above the division. Point 6 puts floors under the macronutrients. Point 7 gives the graded 6.7 / 5.7 / 4.4% figures at 72, 48 and 24 hours. Point 10 covers acute water loss with supervision. Points 12 and 13 cover rehydration and carbohydrate after the scale.

There is no point about sleep. Not a shorter one, not a hedged one — none. In the body text the word occurs three times: once noting that exposure to low humidity "particularly during sleep" may increase water loss through respiration, and twice acknowledging that individual sleep and recovery needs make rigid fight-day scheduling impractical.

This is not a criticism of the authors. It is a nutrition position stand, and sleep is not nutrition. But it tells you something about the shape of the field. A fighter following the best available guidance to the letter receives fourteen distinct instructions about fluid and carbohydrate and none at all about the eight hours that occupy a third of every day of the camp.

A fighter following the best available nutritional guidance to the letter therefore receives sixteen numbered recommendations covering everything they eat and drink from off-camp through to the walk to the cage, and no guidance at all about the third of every day of that camp spent asleep.

The consequence is predictable. When performance drops in week six, the deficit gets blamed, or the volume, or the athlete's attitude. Sleep is not on the list of suspects because it was never on the list of variables.

What the sleep-restriction studies actually measured

The best single answer to "does losing sleep hurt physical performance" is a 2022 systematic review and meta-analysis in Sports Medicine by Craven and colleagues. It pooled 227 outcome measures from 69 publications, sorting tasks into anaerobic power, speed and power endurance, high-intensity interval exercise, strength, endurance, strength-endurance, and skill.

The pooled result was a mean change of −7.56% in performance (95% CI −11.9 to −3.13, p = 0.001), with effects significant across all exercise categories. That is the number worth carrying. Here is the number that should travel beside it: I² = 98.1%. In meta-analysis, I² estimates how much of the variation between studies exceeds chance. At 98.1%, the included studies are not measuring one stable effect with noise around it. They disagree almost completely, which is what you would expect when the protocols range from a full night without sleep to a slightly later bedtime, and the tasks range from a vertical jump to a bench press to a tennis serve.

The subgroup findings are more useful than the headline. Consistent negative effects appeared only under deprivation and late-restriction protocols — that is, total sleep loss, or being woken earlier than normal. Delayed sleep onset with a normal wake time did not produce a consistent effect. Time of day mattered enormously: tasks performed in the afternoon or evening showed consistent decrements, while tasks performed in the morning were largely unaffected. And there was a dose relationship with time awake: roughly 0.4% performance decrease for every additional hour awake before the task, under deprivation and late-restriction protocols.

Two practical readings follow. First, the fight-camp problem — a 5 a.m. strength session after a late finish — is the late-restriction pattern, which is the one with the consistent effect. Second, an athlete who trains at seven in the morning on short sleep and feels fine is not necessarily wrong; the morning tasks in this literature were largely unaffected. It is the evening session that pays.

Note also the composition of the evidence. Of the pooled participants, 959 — 89% — were male. The skill category, the one closest to what a fighter actually does under pressure, contributed 13 outcome measures out of 227.

Reaction time is not skill, and skill is not a result

This is the distinction the whole subject turns on, and almost every article about sleep and sport collapses it.

The most-cited study in this field is Mah and colleagues, 2011, in Sleep. Eleven students on the Stanford men's varsity basketball team, mean age 19.4, kept their habitual schedule for two to four weeks and then extended sleep for five to seven weeks with a target of ten hours in bed. Objective nightly sleep rose by 110.9 ± 79.7 minutes. Timed sprint improved from 16.2 ± 0.61 s to 15.5 ± 0.54 s. Free-throw percentage rose by 9% and three-point percentage by 9.2%. Psychomotor Vigilance Task reaction time and Epworth Sleepiness Scale scores fell. Profile of Mood States vigor rose and fatigue fell.

Every one of those results is real and none of them is "sleep makes you a better fighter."

Consider what each measure is. The Psychomotor Vigilance Task is a sustained-attention test: a stimulus appears at random intervals, you press a button, and it records your latency and your lapses. It is an excellent index of sleep pressure. It is not a measure of whether you slip a jab. A timed sprint on a basketball court is a closed, self-paced, maximal effort against a stopwatch. Free-throw percentage is a self-paced skill executed from a fixed position with no opponent. None of those is an open-skill task under an opponent's pressure, which is the entire content of a fight.

The design matters too. There was no control group — eleven athletes acted as their own baseline across a season in which they were also, unavoidably, practising basketball for seven to eleven weeks. A learning effect and a sleep effect are not separable in that design, and the authors do not claim they are; their conclusion is the appropriately hedged "optimal sleep is likely beneficial in reaching peak athletic performance."

The chain runs: sleep → sustained attention → closed-skill execution → open-skill execution under pressure → competitive outcome. The literature has good evidence for the first two links, thin and heterogeneous evidence for the third, and for the fourth — whether sleeping better changes who wins a fight — it has nothing. No study has measured it. Anyone who tells you a percentage for that is inventing it.

What the combat-sport evidence actually looks like

Strip out everything measured on footballers and basketball players and the combat-sport literature is small, mostly cross-sectional, and mostly conducted in Olympic-style striking and grappling rather than professional MMA or boxing.

The largest cohort. A 2026 cross-sectional analysis in Frontiers in Psychology surveyed 498 athletes in Olympic and non-Olympic combat sports — boxing, wrestling, judo, taekwondo, karate, kickboxing and Muay Thai — with a mean age around 21. Using the Pittsburgh Sleep Quality Index, 65% were classified as poor sleepers (PSQI > 5). In multivariable logistic regression, athletes reporting rapid weight loss above 4% of body mass had 2.70 times the odds of poor sleep quality (95% CI 1.80–4.00). Olympic-sport participation carried odds of 1.75 (1.25–2.45) and male sex 1.65 (1.20–2.30).

Three qualifications travel with that. It is cross-sectional, so it cannot establish that the weight cutting caused the poor sleep rather than both arising from the same demanding period. It is self-report — the PSQI is a questionnaire, not a polysomnogram. And the paper reports that while male sex carried higher odds of crossing the poor-sleeper threshold, female athletes had higher mean PSQI total scores, which is a tension worth noticing before anyone builds a sex difference out of it.

The experimental studies. Souissi and colleagues, 2013, in the Journal of Strength and Conditioning Research, tested 12 judokas on maximal voluntary contraction, handgrip and Wingate before and after a judo combat, at 09:00 and 16:00, after a normal night and after two four-hour partial sleep deprivation conditions — one imposed at the start of the night, one at the end. Sleep deprivation at the end of the night significantly decreased short-term maximal performance in the afternoon. The authors' conclusion: early rising is more detrimental than a late bedtime for judo athletes competing in the afternoon. That mirrors the late-restriction finding in the Craven meta-analysis exactly.

Mejri and colleagues, 2016, in the Journal of Exercise Rehabilitation, ran ten male taekwondo players through the Yo-Yo intermittent recovery test in three sleep conditions. Total distance covered fell significantly after sleep restriction, and fell more after deprivation at the end of the night than at the beginning. The detail worth the whole study: peak heart rate and rating of perceived exertion did not change. The athletes worked less and it did not feel different. Whatever internal signal a fighter uses to judge whether they are recovered, in that cohort it failed to detect a real performance loss.

The weight-cut study. There is exactly one published account of sleep monitored through an actual combat-sport weight cut, and it is a case report: Thomas and colleagues, 2022, in Frontiers in Sports and Active Living. One male athlete, eight weeks, wrist actigraphy. Seven weeks at an intake equal to resting metabolic rate (1700 kcal/day), then 915 down to 300 kcal/day in the five days before weighing in. Energy availability fell from 20 to 3 kcal per kg fat-free mass per day. Total sleep time fell modestly in the final phase — 386 minutes versus 429 and 430 in the earlier phases — while sleep efficiency rose across the period, and the correlations between sleep characteristics, energy availability and training load were trivial to small and non-significant.

The authors' conclusion is that these practices "may be implemented without compromising the sleep of combat sport athletes." Read the sample size before you carry that anywhere. It is one man, tracked with a wrist-worn device, over one camp. It is a case report doing what a case report does — describing a case. It is not evidence that a cut is neutral for sleep, and it is not evidence that it is not.

The deficit gets to sleep before it gets to the scale

The mechanism people expect — you are hungry, so you sleep badly — is the least interesting part of this. What the better-controlled work shows is a measurable change in sleep architecture, not just sleep satisfaction.

Saidi and colleagues, 2024, in Nutrients, studied 42 male rugby players with a mean age of 16.2 across a seven-day period of intensive training with fixed sleep schedules. Energy intake was weighed and recorded, exercise expenditure estimated by accelerometry, and sleep captured by portable polysomnography on the final night. Mean energy availability was 29.3 ± 9.14 kcal per kg fat-free mass per day, with 47.6% of the squad in low energy availability and only 16.7% in the optimal range.

Athletes in low energy availability showed lower sleep efficiency, a lower proportion of N3 (deep) sleep, and more wake after sleep onset than those in the optimal range, with large effect sizes. Segmented regression put a breakpoint between 21 and 33 kcal per kg fat-free mass per day, below which sleep quality declined considerably.

Its limits are worth being exact about: adolescent males, one sport, one night of polysomnography, and a cross-sectional comparison between energy-availability groups rather than a manipulation of energy availability within athletes. It does not tell you that dropping your intake tonight degrades your N3 tomorrow. It tells you that in this squad, at this point in a hard training block, the athletes eating least relative to their expenditure had measurably worse sleep on the night it was recorded.

The body-composition side comes from outside sport entirely. Nedeltcheva and colleagues, 2010, in Annals of Internal Medicine, ran a randomised crossover in 10 overweight adults with a mean age of 41 — three women, seven men, not athletes — through 14 days of moderate caloric restriction with either 8.5 or 5.5 hours of sleep opportunity. Sleep curtailment cut the proportion of weight lost as fat by 55% (1.4 vs 0.6 kg) and increased the loss of fat-free mass by 60% (1.5 vs 2.4 kg, p = 0.002), alongside increased hunger and a shift in substrate utilisation away from fat oxidation.

That study is the origin of every "you lose sixty percent more muscle when you diet on bad sleep" post you have ever read. The finding is real and the cohort is ten middle-aged overweight non-athletes over two weeks. It is a plausible mechanism for a fighter and it is not a measurement of one. If a fighter carrying a fourteen-week longitudinal descent is sleeping five and a half hours, the concern that the composition of the loss shifts is reasonable; the specific 60% is not transferable, and nobody has run the equivalent study in combat athletes.

Endocrine effects are quoted with the same looseness. Leproult and Van Cauter, 2011, in JAMA, restricted 10 healthy young men, mean age 24.3, to five-hour bedtimes for eight nights. Daytime testosterone from 8 a.m. to 10 p.m. fell 10–15%, from 18.4 to 16.5 nmol/L (p = .049). The authors noted for comparison that normal ageing reduces testosterone by roughly 1–2% per year. That comparison is how the claim "one bad week ages you ten years" got into circulation. The study measured a 10–15% decline in ten young men over eight nights of five-hour bedtimes. It did not measure ageing.

Dehydration, the descent, and what nobody has measured

Here the honest answer is short: there is no published study of sleep during an acute fight-week water cut. Not a small one, not a bad one. It does not exist.

What exists is adjacent. Rosinger and colleagues, 2019, in Sleep, analysed self-reported sleep duration against urinary hydration biomarkers in three adult samples — two NHANES waves and the Chinese Kailuan cohort, with healthier subsets of 11,353 and 8,766 people. Compared with eight hours, six hours of sleep was associated with higher odds of inadequate hydration by urine specific gravity: adjusted OR 1.59 (95% CI 1.25–2.03) in the US sample and 1.42 (1.26–1.60) in the Chinese sample.

Read the direction carefully, because it is the opposite of the one people assume. This is short sleep associated with more concentrated urine — a plausible mechanism being that vasopressin release late in the sleep period is truncated by waking early. It is cross-sectional, it is in general adults rather than athletes, and it does not establish that being dehydrated makes you sleep badly. That inference is the one a fighter cares about and it is the one the study cannot support.

So what can be said about fight week and sleep? Only this: an athlete in the final days of a descent is simultaneously in deep energy deficit, deliberately fluid-depleted, frequently in an unfamiliar hotel, often in a different time zone, and carrying the psychological load of a scheduled fight. Each of those has independent evidence associating it with disturbed sleep. Their combination has never been studied as a combination, and anyone offering a figure for the compound effect is extrapolating.

The one figure that is directly relevant is the cross-sectional one from the 498-athlete combat cohort: rapid weight loss above 4% of body mass, odds ratio 2.70 for poor sleep quality. It is self-report, it is correlational, and it is still the best combat-sport-specific number in existence on this question. That tells you how thin the field is. The rest of the fight-week protocol is covered in the weight cut and fight week material, where the same rule applies: the condition attached to a number is part of the number.

Sleep, immune function, and getting sick in week eight

The most reliable practical consequence of short sleep in a training block may not be performance at all. It is illness, and it is the one place the evidence gets close to causal.

Prather and colleagues, 2015, in Sleep, ran the study that matters here. 164 healthy adults aged 18 to 55 wore wrist actigraphs for seven consecutive days, were then quarantined and administered nasal drops containing rhinovirus, and were monitored for five days for a clinical cold. Compared with those sleeping more than seven hours, those sleeping under five hours had OR 4.50 for developing a clinical cold (95% CI 1.08–18.69) and those sleeping five to six hours OR 4.24 (1.08–16.71). Those sleeping 6.01 to 7 hours were at no greater risk (OR 1.66, CI 0.40–6.95). The association held independent of pre-challenge antibody levels, demographics, season, BMI, psychological variables and health practices. Sleep fragmentation was unrelated.

This is an experimental viral challenge with objectively measured sleep, which is about as strong as this kind of research gets. Look at the confidence intervals anyway: 1.08 to 18.69 is an interval whose lower bound is barely above one and whose upper bound is nearly nineteen. The direction is well supported. The magnitude is not pinned down.

In athletes specifically, Fitzgerald and colleagues, 2019, in the Journal of Science and Medicine in Sport, followed 44 nationally competitive male Australian football athletes across a 46-week season with illness recorded daily by doctors. In multivariate analysis, acute sleep quantity was the only variable significantly associated with illness incidence (OR 0.49, 95% CI 0.25–0.94) once training-load variables were controlled — and there was no relationship between external training load and illness once sleep metrics were in the model.

The 2021 BJSM expert consensus summarises the population-level position in one sentence: "Studies in the wider population show that habitually sleeping <7 hours/night increases susceptibility to respiratory infection."

Which brings us to the number you have definitely heard. "One night of bad sleep drops your immune system by seventy percent" traces to Irwin and colleagues, 1994, in Psychosomatic Medicine: 23 healthy male volunteers, deprived of sleep between 3 and 7 a.m. Natural killer cell activity fell in 18 of the 23, and average lytic activity was reduced to a level 72% of the mean of three baseline values. Reduced to 72% — which is a reduction of about 28%, not 70%. The seventy-percent version is a misreading of that sentence that has been repeated for three decades. The same paper also reports that after one night of resumed nocturnal sleep, NK activity had returned to baseline.

Six weeks out from a fight, "I got sick and lost ten days" is a bigger threat to a camp than a 2% decrement in a Wingate test. This is the part of the sleep literature with the most direct claim on a fighter's attention.

The 1.7× injury figure, and what it is actually made of

Every article on this subject cites it, so here is what is underneath it.

Milewski and colleagues, 2014, in the Journal of Pediatric Orthopaedics: 160 student athletes at a combined middle and high school were consented, 112 completed an online survey (70%) — 54 male, 58 female, mean age 15, range 12 to 18 — and their responses were correlated with a retrospective review of injury records kept by the school athletic department. Multivariate analysis found hours of sleep per night and grade in school were the best independent predictors of injury. Athletes averaging under eight hours were 1.7 times more likely to have had an injury (95% CI 1.0–3.0, p = 0.04). For each additional grade in school, athletes were 1.4 times more likely (95% CI 1.2–1.6, p < 0.001).

Four things about that, none of which appear when the figure is quoted:

The confidence interval's lower bound is 1.0. That is the boundary of no effect. The p-value is 0.04. This is a result at the edge of conventional significance in a sample of 112.

The design is retrospective and cross-sectional. Sleep was reported at one moment; injuries were read out of records covering a preceding period. An athlete who was injured may have slept badly because they were injured.

Grade in school was the stronger and far more precisely estimated predictor. In the paper's own model, being a year older in that population was more reliably associated with injury than sleeping under eight hours.

And the population is adolescents in school sports, in a country with a specific school-sport structure. It is not adult professional combat athletes, and the paper does not claim to be about them.

A 2019 conference abstract and, more recently, a preprint meta-analysis pooling nine observational studies of 1,078 athletes report a smaller pooled association — an odds ratio around 1.34 — with substantial heterogeneity. Preprints are not peer-reviewed, this article does not treat that figure as established, and it is deliberately not listed among the sources below for that reason. The direction of the literature is consistent. The magnitude, in adults, in combat sports, is unmeasured.

What can be said without stretching: less sleep is repeatedly associated with more injury across athlete populations; the mechanism is plausible and probably multi-causal, running through attention, motor control, tissue recovery and training-load tolerance simultaneously; and no one has produced a number that applies specifically to a professional fighter in a hard sparring block.

Travel, time zones and the disruption fight week is built from

This is where the evidence is best characterised and the interventions are worst.

Roberts and colleagues, 2019, in the British Journal of Sports Medicine, systematically reviewed 54 studies objectively reporting sleep in elite athletes. During normal training, most studies found athletes unable to meet total sleep time or sleep efficiency recommendations. On the night of competition, total sleep time was about 60 minutes shorter than on preceding nights. For evening competitions starting at or after 18:00 — which is every fight card — the reduction was about 80 minutes, versus about 20 minutes for daytime competition, with sleep efficiency 3–4% lower after night events. The review also identified early morning training starts before 07:00, increases in training load above 25%, late-night or early-morning travel departures, eastward air travel and altitude ascent as impairing sleep.

Sargent and colleagues, 2014, in Chronobiology International, is where the "elite athletes sleep about six and a half hours" figure comes from: 70 nationally ranked athletes from seven sports, wrist activity monitors and diaries for two weeks. They spent 08:18 ± 01:12 in bed and obtained 06:30 ± 01:24 of sleep, with significantly less sleep on nights before training days than before rest days. That figure is real, it is specific to that cohort, and it is not a fact about fighters.

Now the intervention side. Janse van Rensburg and colleagues, 2020, also in BJSM, systematically reviewed interventions for travel fatigue and jet lag in athletes — exercise, sleep, light, nutrition, melatonin, sedatives, stimulants, melatonin analogues, glucocorticoids, antihistamines. Twenty-two articles met criteria. Their two conclusions deserve to be quoted rather than summarised: "We found no literature pertaining to the management of travel fatigue", and "Evidence for the successful management of jet lag in athletes was of low quality." Only 12 of the studies were athlete-specific.

That is a remarkable state of affairs for a sport where the athlete routinely flies across four to ten time zones, weighs in on arrival day plus one, and competes in the evening. The strategies that circulate — shift your schedule an hour a day for a week beforehand, light exposure on a fixed timetable, melatonin at a specific dose and hour — are not baseless, but they are not backed by athlete evidence of any strength, and the review says so. The honest framing for fight week is that the disruption is well documented and the countermeasures are not.

One implication is concrete. If evening competition costs roughly 80 minutes of sleep on the night of the event, and the weigh-in is the day before, then the two nights most likely to be short are the two nights immediately either side of the scale — the same window in which rehydration and glycogen restoration have to happen. Whatever sleep is going to be banked in a fight week is banked earlier than that.

Sparring load, central fatigue, and the thing coaches call CNS fatigue

"CNS fatigue" is used in gyms to mean a specific feeling: heavy, flat, slow, unwilling, typically after hard sparring or heavy singles, and distinct from sore muscles. The feeling is real. The term, as used, is not a measured quantity, and it is worth separating what exists from what does not.

Central fatigue is a defined construct in exercise physiology: a progressive failure to voluntarily activate the muscle, distinguishable from peripheral fatigue occurring at or beyond the neuromuscular junction. It is measured — by superimposing an electrical or magnetic stimulus on a maximal voluntary contraction and quantifying voluntary activation, or by transcranial magnetic stimulation measures such as motor-evoked potential and silent period. Studies using those methods can and do show reductions in voluntary activation after fatiguing protocols, and can show them independent of peripheral markers.

What does not exist is a validated field measure of that construct, a dose-response between rounds of sparring and it, or a threshold at which a coach should pull a fighter out of a session. When someone prescribes deload volumes based on "CNS fatigue," they are using a real physiological term as a label for a subjective state, and the label carries an authority the measurement behind it has not earned in that context.

What is measurable about sparring is load. Kirk and colleagues, 2024, in the Journal of Sports Sciences, instrumented ten MMA participants with accelerometers through 3 × 5-minute sparring bouts and related external load to laboratory aerobic variables, finding VO2max strongly related to accumulated PlayerLoad and PlayerLoad per minute, with the lower-aerobic half of the cohort drifting from moderate to high session RPE across the bout. That is ten participants, and it is about intensity, not recovery. Nobody has published a sparring-load-to-sleep dose-response in combat sports.

The defensible version of the coaching intuition is narrower and better supported. Hard sparring is a high-arousal, late-in-the-day, occasionally head-contact event, and the competition-sleep literature shows evening high-arousal events cost sleep on the night they occur. Sparring at 8 p.m. and then trying to sleep at 10 is a scheduling problem with real evidence behind it. That is a different, more actionable claim than "your CNS is fried."

Monitoring: what a wearable can and cannot tell you

The device on your wrist is measuring movement and, increasingly, heart rate and skin temperature. It is inferring sleep. The gap between the measurement and the inference is where the caution belongs.

Chinoy and colleagues, 2021, in Sleep, tested four wearable and three non-wearable consumer devices against polysomnography in 34 healthy young adults across three consecutive nights including a disrupted-sleep condition. Most devices performed comparably to or better than research actigraphy for total sleep time, though two devices overestimated it by 43.7 to 46.8 minutes and underestimated wake after sleep onset by 47.6 to 49.5 minutes. Sleep-stage output was, in the authors' words, inconsistent; all devices overestimated light sleep, and deep and REM results were mixed. The finding that governs everything else: specificity for detecting wake ranged from 0.18 to 0.54. These devices are good at knowing you are asleep and poor at knowing you are awake.

A 2025 meta-analysis in the Journal of Clinical Sleep Medicine pooled 24 studies covering 798 people across Fitbit, Garmin, WHOOP, Apple Watch, Readiband and others. It found significant differences from polysomnography in total sleep time (about 17 minutes), sleep efficiency (about 4.7 percentage points), sleep latency (about 2.6 minutes) and wake after sleep onset (about 13 minutes). Its conclusion is the operative one: wrist-worn consumer devices "are not as reliable as polysomnography," while remaining "useful for tracking general sleep patterns."

Now the part that needs labelling clearly. Recovery scores, readiness scores and strain scores are vendor-published constructs. ⚠️ Each manufacturer defines its own composite from its own inputs with its own undisclosed weighting, and validation of the underlying sleep staging — the thing those scores are largely built from — is exactly what the studies above found inconsistent. A recovery score is not a measurement of recovery. It is a proprietary index correlated with some measurements of sleep, and it should be treated as marketing-adjacent until a manufacturer publishes an independent validation of the score itself, which as of this writing none has in a form that would satisfy the papers above.

The 2021 BJSM consensus flags this directly, listing among the field's problems "questions concerning the validity of the chosen sleep assessment tools."

There is also a failure mode specific to tracking. Orthosomnia — an excessive preoccupation with achieving perfect sleep, often driven by device data — is now a described construct with validated screening instruments and prevalence estimates in the general population. A fighter who lies awake at 1 a.m. worrying about a deep-sleep percentage has been made worse off by the device. Use the weekly trend, ignore the nightly score.

The head-trauma question, and exactly how far the evidence goes

This section is short on purpose, because the evidence is.

Sleep disturbance is a well-documented consequence of concussion. The 6th International Consensus Statement on Concussion in Sport, from the Amsterdam conference of October 2022, covers assessment and management including symptom domains in which sleep features, and scoping reviews report that a substantial proportion of concussed individuals develop post-injury sleep disturbance, which is associated with prolonged recovery. That direction — injury causing sleep disturbance — has real support.

The reverse direction is where the claims outrun the evidence. "Poor sleep makes you more likely to get concussed" is biologically plausible via attention and motor control, and it has not been demonstrated prospectively in combat sports.

The strongest version circulating is that deep sleep clears the proteins associated with chronic traumatic encephalopathy, so a fighter who sleeps well is protecting their brain from sparring. Trace it. It comes from Xie and colleagues, 2013, in Science: real-time tetramethylammonium diffusion measurement and two-photon imaging in live mice, showing that natural sleep or anaesthesia was associated with roughly a 60% increase in interstitial space and increased convective exchange of cerebrospinal fluid, which increased the rate of β-amyloid clearance. That is a mouse study, it concerns amyloid rather than the tau pathology characteristic of CTE, and the human evidence for glymphatic function relies on indirect imaging surrogates in largely cross-sectional designs. No study has measured brain clearance in combat athletes as a function of sleep. The mechanism is genuinely interesting and it is not a finding about fighters.

Even the direct sparring evidence resists the simple story. Esagoff and colleagues, 2023, in Sports Medicine, analysed 94 active professional MMA fighters from the Professional Fighters Brain Health Study and found that more weekly sparring rounds was significantly associated with larger left and right caudate volumes (β = 13.5 and 14.9 µL per round), with no significant association for thalamus, putamen, hippocampus or amygdala. Cross-sectional, so causally uninterpretable in either direction — but it is not the result anyone expected, and it should temper confidence in any tidy narrative about sparring, sleep and the brain.

The defensible position: reduce head-impact exposure because head-impact exposure is the risk factor, and sleep well because sleep is associated with better outcomes across almost everything measured. Do not connect those two claims with a mechanism the evidence has not established in humans, let alone in fighters.

What breaks first, and how you would know

In roughly the order it happens in a camp, with the signal that precedes each.

Sleep opportunity goes before sleep quality does. The first thing a busy camp takes is time in bed — an added morning session, a longer commute to a second gym, an evening sparring block that ends at nine. The signal is a calendar, not a sensation. If the schedule cannot accommodate eight hours in bed, no sleep practice fixes it.

Then the deficit reaches architecture. As energy availability falls into the range where the rugby polysomnography data showed a breakpoint, the loss is not in hours but in efficiency and deep sleep — more awakenings, less N3, the same time in bed producing less restoration. The signal is waking unrefreshed at an unchanged sleep duration, and it will not show up as a shorter night on a tracker.

Then perceived exertion stops tracking output. This is the most useful early warning in the whole picture and it comes from the taekwondo data: distance covered fell while heart rate and RPE did not change. When a fighter's honest self-report says the session felt normal and the objective output has drifted, the self-report is the thing that has failed. The detection method is objective: round-by-round output, timed work, a recorded lift, anything that produces a number the athlete does not have to feel.

Then illness takes a week. Not a fight-ending problem in isolation, but ten days of interrupted camp six weeks out is precisely the disruption that pushes the descent behind schedule, which deepens the deficit, which is where this sequence started.

Then the fight week compounds it. Deficit at its deepest, fluid restricted, hotel bed, time zone, evening competition worth about 80 minutes of lost sleep on the night, and a rehydration window that depends on doing several things correctly while at the lowest cognitive resource of the whole camp.

The intervention that addresses most of this is unglamorous and it is scheduling. Fixed wake time, sessions placed to protect the sleep window rather than filling the day, and the descent planned far enough back that the deficit never has to deepen to compensate for lost weeks — which is most of what the rest of the fight camp material is about. Building the plan backwards from the weigh-in rather than forwards from today is what makes that possible — it is the logic Fighter Cut is built on, and it is worth doing on paper if you never open the app.

What we could not verify

Stated plainly, because an article this specific about other people's numbers owes you its own gaps.

• The "30% drop in time to physical exhaustion below six hours of sleep" figure. It circulates widely as a coaching quote attributed to Matthew Walker's trade book Why We Sleep. It is not a peer-reviewed finding, and we could not trace it to a primary study with a stated cohort and protocol. The book has been the subject of published criticism regarding the representation of underlying data. We do not print the figure as fact. The closest defensible number is the Craven pooled −7.56% with I² = 98.1%.
• "Sleep is the greatest legal performance-enhancing drug." Attributable to the same trade book. It is a slogan, not a finding, and no study has compared sleep to any pharmacological agent on any performance outcome.
• "Athletes need nine to ten hours." We found no primary measurement establishing a sleep requirement for adult athletes. The ten-hour figure is the intervention target — hours in bed — in Mah's eleven-athlete pilot, and the nine-to-ten range appears in paediatric guidance for adolescents. The 2021 BJSM consensus explicitly states that a one-size-fits-all recommendation "is unlikely ideal" and recommends an individualised approach.
• "You can never repay sleep debt." We could not trace this absolute claim to a primary source. What we found points the other way in specific outcomes: Irwin's NK-cell activity returned to baseline after one night of recovery sleep, and Arnal's sleep-extension benefits persisted after a single recovery night. That is not a licence to under-sleep; it is a statement that the absolute version of the claim is unsupported by what we located.
• Whether the 2023 IOC REDs consensus formally lists sleep disturbance among REDs outcomes. We verified the consensus statement's citation and abstract, but the full text is paywalled and our attempts to retrieve the public PDF timed out. We therefore cite it only for the low-energy-availability framing its abstract states, and make no claim about its sleep content.
• The direction of the sign convention in the 2025 wearable meta-analysis. We report the magnitude of the device-versus-polysomnography differences, not which way each one errs, because the abstract does not state the convention unambiguously.
• Any sleep figure specific to professional MMA or boxing during a camp. There is none. The combat-sport evidence is one cross-sectional survey of 498 athletes, two small experimental studies in judo and taekwondo with 12 and 10 participants, and one case report. Everything else in this article is borrowed from other sports or from non-athletes and labelled as such.
• Nothing here is specific to you — not your training age, your schedule, your medical history, your medication, or any sleep disorder you may have and not know about. Persistent unrefreshing sleep is a reason to see a physician, not to read another article.

Sources

Sourced to

1. Sleep and the athlete: narrative review and 2021 expert consensus recommendations — Walsh NP, Halson SL, Sargent C, et al., British Journal of Sports Medicine, 2021;55(7):356–368. DOI 10.1136/bjsports-2020-102025, PMID 33144349
2. Effects of Acute Sleep Loss on Physical Performance: A Systematic and Meta-Analytical Review — Craven J, McCartney D, Desbrow B, et al., Sports Medicine, 2022;52(11):2669–2690. DOI 10.1007/s40279-022-01706-y, PMID 35708888
3. The effects of sleep extension on the athletic performance of collegiate basketball players — Mah CD, Mah KE, Kezirian EJ, Dement WC, Sleep, 2011;34(7):943–950. DOI 10.5665/sleep.1132, PMID 21731144
4. Chronic lack of sleep is associated with increased sports injuries in adolescent athletes — Milewski MD, Skaggs DL, Bishop GA, et al., Journal of Pediatric Orthopaedics, 2014;34(2):129–133. DOI 10.1097/BPO.0000000000000151, PMID 25028798
5. Sleep quality in Olympic and non-Olympic combat sports: associations with sex, rapid weight loss, and sport age (a PSQI-based analysis) — Eroğlu Ş, Ceylan T, Türkmen M, et al., Frontiers in Psychology, 2026;17:1847782. DOI 10.3389/fpsyg.2026.1847782, PMID 42317690
6. Association between Low Energy Availability (LEA) and Impaired Sleep Quality in Young Rugby Players — Saidi O, Souabni M, Del Sordo GC, et al., Nutrients, 2024;16(5):609. DOI 10.3390/nu16050609, PMID 38474738
7. Case Report: Effect of low energy availability and training load on sleep in a male combat sport athlete — Thomas C, Langan-Evans C, Germaine M, et al., Frontiers in Sports and Active Living, 2022;4:981755. DOI 10.3389/fspor.2022.981755, PMID 36733958
8. Behaviorally Assessed Sleep and Susceptibility to the Common Cold — Prather AA, Janicki-Deverts D, Hall MH, Cohen S, Sleep, 2015;38(9):1353–1359. DOI 10.5665/sleep.4968, PMID 26118561
9. Insufficient sleep undermines dietary efforts to reduce adiposity — Nedeltcheva AV, Kilkus JM, Imperial J, Schoeller DA, Penev PD, Annals of Internal Medicine, 2010;153(7):435–441. DOI 10.7326/0003-4819-153-7-201010050-00006, PMID 20921542
10. Effect of 1 week of sleep restriction on testosterone levels in young healthy men — Leproult R, Van Cauter E, JAMA, 2011;305(21):2173–2174. DOI 10.1001/jama.2011.710, PMID 21632481
11. Effects of training and competition on the sleep of elite athletes: a systematic review and meta-analysis — Roberts SSH, Teo WP, Warmington SA, British Journal of Sports Medicine, 2019;53(8):513–522. DOI 10.1136/bjsports-2018-099322, PMID 30217831
12. How to manage travel fatigue and jet lag in athletes? A systematic review of interventions — Janse van Rensburg DCC, Jansen van Rensburg A, Fowler P, et al., British Journal of Sports Medicine, 2020;54(16):960–968. DOI 10.1136/bjsports-2019-101635, PMID 32303523
13. The impact of training schedules on the sleep and fatigue of elite athletes — Sargent C, Lastella M, Halson SL, Roach GD, Chronobiology International, 2014;31(10):1160–1168. DOI 10.3109/07420528.2014.957306, PMID 25222347
14. Performance of seven consumer sleep-tracking devices compared with polysomnography — Chinoy ED, Cuellar JA, Huwa KE, et al., Sleep, 2021;44(5):zsaa291. DOI 10.1093/sleep/zsaa291, PMID 33378539
15. Performance of consumer wrist-worn sleep tracking devices compared to polysomnography: a meta-analysis — Lee YJ, Lee JY, Cho JH, Kang YJ, Choi JH, Journal of Clinical Sleep Medicine, 2025;21(3):573–582. DOI 10.5664/jcsm.11460, PMID 39484805
16. Partial sleep deprivation reduces natural killer cell activity in humans — Irwin M, Mascovich A, Gillin JC, et al., Psychosomatic Medicine, 1994;56(6):493–498. DOI 10.1097/00006842-199411000-00004, PMID 7871104
17. International society of sports nutrition position stand: nutrition and weight cut strategies for mixed martial arts and other combat sports — Ricci AA, Evans C, Stull C, et al., Journal of the International Society of Sports Nutrition, 2025;22(1):2467909. DOI 10.1080/15502783.2025.2467909, PMID 40059405
18. Effects of time-of-day and partial sleep deprivation on short-term maximal performances of judo competitors — Souissi N, Chtourou H, Aloui A, et al., Journal of Strength and Conditioning Research, 2013;27(9):2473–2480. DOI 10.1519/JSC.0b013e31827f4792, PMID 23974210
19. Does one night of partial sleep deprivation affect the evening performance during intermittent exercise in Taekwondo players? — Mejri MA, Yousfi N, Mhenni T, et al., Journal of Exercise Rehabilitation, 2016;12(1):47–53. DOI 10.12965/jer.150256, PMID 26933660
20. Sleep extension in athletes: what we know so far — A systematic review — Silva AC, Silva A, Edwards BJ, et al., Sleep Medicine, 2021;77:128–135. DOI 10.1016/j.sleep.2020.11.028, PMID 33352457
21. Benefits of Sleep Extension on Sustained Attention and Sleep Pressure Before and During Total Sleep Deprivation and Recovery — Arnal PJ, Sauvet F, Leger D, et al., Sleep, 2015;38(12):1935–1943. DOI 10.5665/sleep.5244, PMID 26194565
22. The influence of sleep and training load on illness in nationally competitive male Australian Football athletes: A cohort study over one season — Fitzgerald D, Beckmans C, Joyce D, Mills K, Journal of Science and Medicine in Sport, 2019;22(2):130–134. DOI 10.1016/j.jsams.2018.06.011, PMID 29945830
23. Short sleep duration is associated with inadequate hydration: cross-cultural evidence from US and Chinese adults — Rosinger AY, Chang AM, Buxton OM, Li J, Wu S, Gao X, Sleep, 2019;42(2):zsy210. DOI 10.1093/sleep/zsy210, PMID 30395316
24. Sleep drives metabolite clearance from the adult brain — Xie L, Kang H, Xu Q, et al., Science, 2013;342(6156):373–377. DOI 10.1126/science.1241224, PMID 24136970
25. Sparring and the Brain: The Associations between Sparring and Regional Brain Volumes in Professional Mixed Martial Arts Fighters — Esagoff AI, Heckenlaible NJ, Bray MJC, et al., Sports Medicine, 2023;53(8):1641–1649. DOI 10.1007/s40279-023-01838-9, PMID 36972014
26. Consensus statement on concussion in sport: the 6th International Conference on Concussion in Sport — Amsterdam, October 2022 — Patricios JS, Schneider KJ, Dvorak J, et al., British Journal of Sports Medicine, 2023;57(11):695–711. DOI 10.1136/bjsports-2023-106898, PMID 37316210
27. 2023 International Olympic Committee's (IOC) consensus statement on Relative Energy Deficiency in Sport (REDs) — Mountjoy M, Ackerman KE, Bailey DM, et al., British Journal of Sports Medicine, 2023;57(17):1073–1097. DOI 10.1136/bjsports-2023-106994, PMID 37752011
28. The influence of aerobic capacity on the loads and intensities of mixed martial arts sparring bouts — Kirk C, Clark D, Langan-Evans C, Journal of Sports Sciences, 2024;42(22):2093–2102. DOI 10.1080/02640414.2024.2419239, PMID 39551930

Read next