What to track in a fight camp, and what to ignore

1. The claim under test

The claim a fighter meets everywhere is that more data produces a better camp: that a wearable, a jump mat, a smart scale and a recovery score together describe the athlete's state, and that reading them will catch problems before they cost a fight.

The literature supports a much narrower version. It supports the idea that monitoring something is better than monitoring nothing, that the athlete's own report carries information the instruments miss, and that a load number computed from perceived exertion tracks other load measures closely enough to be useful. It does not support the idea that any of it predicts a result, and it does not supply the one thing a weekly number needs to be interpretable, which is a threshold for what counts as a real change.

The distinction is not academic. A camp is eight to twelve weeks. A measurement whose week-to-week error exceeds the change a week can plausibly produce will generate a signal every week, and every one of those signals will be noise. That is the failure mode this article is organised around.

2. What the systematic review actually says

Saw, Main and Gastin published the anchor synthesis in the British Journal of Sports Medicine in 2016. Its abstract sentence is the one that gets quoted: "Subjective and objective measures of athlete well-being generally did not correlate. Subjective measures reflected acute and chronic training loads with superior sensitivity and consistency than objective measures."

The second sentence, less quoted, sets the size of the claim: within studies, subjective measures were more sensitive and consistent than objective measures in 22 of the 54 studies that permitted the comparison. That is a majority of nothing. It is a plurality finding — more often, not always.

The objective measures on the losing side of that comparison were named specifically, and the list is long: cortisol, testosterone, growth hormone, prolactin, ACTH, LH, dopamine, IGF-1, red blood cell count, haematocrit, haemoglobin, leukocytes, immunoglobulins, interleukins, TNF-α, CRP, oxidative stress markers, creatine kinase, urea, creatinine, resting and exercise heart rate, heart rate variability, blood pressure, lactate, VO₂, and performance tests of both sustained and short duration. That is the review's inclusion list rather than a ranking within it, but it is worth reading slowly, because almost every product sold to a fighter as an objective monitoring solution measures something on it.

Four limitations travel with the headline and are not optional. The review could not perform a meta-analysis because of heterogeneity in methods and reporting. It depended on studies that mostly reported p-values, carrying a Type II error risk. Subjective measures were often collected retrospectively while objective ones were concurrent, which is not a fair contest. Many studies had small samples, publication bias was flagged, and group-level reporting may have masked individual responses. Any sentence that reads "self-report beats the wearable" without those attached is overstating the paper.

3. Session-RPE, and the evidence borrowed sideways

Session-RPE is the athlete's rating of overall session intensity multiplied by session duration in minutes, producing a single arbitrary-unit load figure. It is the closest thing the field has to a validated combat-sport load measure, and the qualification in that sentence does a lot of work.

The validity coefficients, collected in a 2017 review in Frontiers in Neuroscience, are these. In taekwondo, session-RPE correlated with Banister's TRIMP at r = 0.56–0.90 across 308 individual sessions in young male athletes averaging 13.1 ± 2.4 years — aerobic sessions r = 0.57–0.60, technique and tactics r = 0.60–0.61. In karate, session-RPE against Edwards' TRIMP reached r = 0.84–0.97 over ten sessions in 11 male youth athletes, and a separate study in male and female karate athletes aged 24.2 ± 2.3 and 22.6 ± 1.2 years reported ICC = 0.81 with r = 0.65–0.95. In fencing, session-RPE against Banister's TRIMP ran r = 0.84–0.98 and against Edwards' r = 0.91–0.98, across 67 training sessions and 101 competition bouts. A mixed-gender taekwondo cohort produced r = 0.52–0.71.

Every session-RPE validity figure in combat sport comes from taekwondo, karate or fencing, and several cohorts are adolescents. The core recommendation of this article rests on evidence borrowed sideways. No fetched study validates session-RPE in MMA, professional boxing, Muay Thai, BJJ or wrestling — the five sports this article is tagged for.

The same review is useful about what actually drives the number. Time spent at high intensity, and only marginally the session duration, influenced the session-RPE. And in submaximal efforts under non-excessive loads, fatigue, stress, muscle soreness and sleep were not major contributors to the rating. That second finding carries a condition that matters enormously here: submaximal, non-excessive. It does not extend to a hard camp week, which is the only condition a fighter cares about. The deeper treatment of what load numbers can and cannot carry is in training load monitoring for fighters.

4. What a monitored ten-week block actually looks like

The most useful description of a real monitored block in a combat sport is a 2026 observational study of 20 elite junior female boxers in Türkiye — 18.9 ± 1.2 years, 165.4 ± 5.8 cm, 59.8 ± 6.1 kg, all with at least three years' experience — followed across ten weeks of normal training with no experimental manipulation.

Weekly session-RPE load peaked at 3,322.36 ± 487.43 AU in week 1 and bottomed at 1,239.48 ± 279.37 AU in week 7, F(9,171) = 101.32, p < 0.001, ηp² = 0.842. Training monotony was highest in week 1 at 1.14 ± 0.08 and lowest in week 5 at 0.38 ± 0.00. Strain peaked in week 2 at 3,734.96 ± 771.65 and bottomed in week 5 at 616.31 ± 23.99. Repetitive-jump performance improved in week 2, declined across weeks 3 to 5, and partially recovered in week 6.

That last sentence is the reason to read the study. The performance dip is not a warning sign in isolation; it is what a block of accumulated load looks like from the inside, visible in a metric that costs nothing but a number written down after each session. Without the load series beside it, a coach reading three consecutive weeks of falling jump numbers has no way to tell an intended dip from an unravelling athlete.

The authors' own limitations are extensive and belong here. The sample is elite junior female boxers only, which they say limits generalisability to other age groups, competitive levels, or male athletes. No boxing-specific external load was recorded — no punch frequency, no contact intensity. No hormonal status, energy availability, sleep, heart rate variability or muscle-damage markers were collected. And there was no menstrual cycle monitoring, which the authors call a particularly relevant limitation in research involving female athletes. The design is observational, so no mechanism can be inferred from it.

5. Measurement error is the spine of the whole problem

Here is the argument the rest of the article hangs on. A fighter's true weekly change is frequently smaller than the error a meal introduces into the measurement of it.

Kerr, Slater and Byrne measured this directly in 32 resistance-trained males with at least two years' training and a BMI of 25 or above. Under standardised conditions — overnight fasted and rested — measurement errors were trivial to small across DXA, BOD POD and three- and four-compartment models, with bioimpedance spectroscopy fat mass the one substantial exception. Under non-standardised conditions, meaning ad libitum food and fluid plus activity, the errors were mostly moderately or substantially large across all techniques. Bioimpedance fat-mass typical error was 879 g; the three- and four-compartment models each carried a typical error of 631 g for fat mass.

The acute experiments are the vivid part. A single 500 g meal, measured fifteen minutes later, moved bioimpedance fat mass by 774 g. A 500 g meal plus 1 L of water moved bioimpedance fat mass by 1,400 g, BOD POD fat mass by 820 g, and DXA fat-free mass by 1,211 g. The authors' rule follows from that: DXA and BOD POD remain acceptable if acute food and fluid intake stays below 500 g, with assessment in an overnight-fasted and rested state.

Read that against a camp. A week in which an athlete genuinely lost 400 g of tissue is a week whose signal is comfortably inside the shift a breakfast produces in a body-composition estimate. The conclusion the arithmetic supports is not that a better device is needed. It is that the same time, the same state and the same conditions, every time, is worth more than any upgrade — which is also the argument made at greater length in reading a bodyweight trend in camp.

Two conditions on those numbers. The cohort is 32 resistance-trained males with a BMI of 25 or higher, and the food and fluid experiments are acute and same-day. We could not confirm one further figure from this paper — a 1,960 g shift in total body water under fully non-standardised conditions — against the abstract we could reach, so it does not appear as a claim here. The 1,400 g and 1,211 g figures carry the same argument and are confirmed.

6. The jump mat, and the change nobody has published

Countermovement jump height is the most common neuromuscular monitoring test in sport, and its device-level reliability has been measured. Across 60 jumps in 15 healthy males aged 27.0 ± 5.0 years and weighing 72.2 ± 8.28 kg: Myotest ICC 0.97, CV 4.2%; Ergojump 0.93 / 3.2%; Optojump 0.87 / 4.2%; MyJump 0.97 / 3.9%.

The smallest worthwhile change for jump height in that study ranged 0.008 to 0.015 m — 8 to 15 mm depending on the device — computed as 0.2 times the between-subject standard deviation. That range, rather than a single number, is what the paper reports, and the difference between its ends decides whether a given day's reading means anything.

Do the arithmetic on a 40 cm jump. A coefficient of variation of 4% is roughly 1.6 cm of typical measurement noise. Against the low end of the smallest worthwhile change that is about twice the change worth caring about; against the high end it is roughly equal to it. Either way, a single Tuesday reading 1 cm below Monday's is inside the noise of the instrument, and treating it as information about the athlete is a category error.

The cohort condition is severe. Those 15 participants were sports-science students of mixed training background, not combat athletes, and the authors state explicitly that generalisation to professional athletes needs further study.

And this is where the audit of the combat-sport testing literature lands hardest. Across 39 studies — judo 39%, taekwondo 20%, karate 18%, fencing 10%, wrestling 8%, amateur boxing 5% — only 51% reported reliability data at all. Where intraclass correlations were reported they ranged from 0.43 to 1.00. Criterion validity appeared in 54% of studies, construct validity in 31%, and predictive validity in exactly one. Seventy-four per cent had samples under 30. Sixty per cent gave no full account of familiarisation. Fifty-nine per cent gave no detail on the stability of testing conditions. Only 13% examined test sensitivity at all, and minimal detectable change was investigated in three studies.

For most combat-sport-specific tests, nobody has published a minimal detectable change. The central question of camp monitoring — is this week's number bigger than the noise? — mostly cannot be answered from the literature. The MDC-adjacent figures that do exist come from non-combat cohorts.

7. What the blood says, and what the athlete says

A three-week structured strength and conditioning block in ten male professional MMA athletes — 26.2 ± 0.9 years, 7.0 ± 2.0 years of experience — was sampled at four timepoints: a baseline and the end of each week.

Cortisol rose significantly after week 1 and then declined in the final week. High-sensitivity CRP declined between week 2 and week 3 (p < 0.05). Testosterone and catecholamines stayed stable. Creatine kinase stayed elevated throughout. Meanwhile the Profile of Mood States fatigue score accumulated progressively to its highest value at week 3 (p < 0.0001), and vigour fell across the same period (p < 0.0001).

Biochemistry partly recovered while subjective fatigue did not. That is the specific version of the general finding from the systematic review, observed in the population this article is about.

It is a small study and its authors say so: small sample, short duration, no control group, no sleep monitoring, unstandardised nutrition, no performance outcomes, and in their own phrase, limited generalisability. It is also male only, n = 10.

A second, hedged data point points the same way. In 12 elite boxers — 8 male, 4 female — across a 12-week pre-competition block for the 2018 Commonwealth Games, significant week effects were reported for training load with a large effect size and for change in creatine kinase with a moderate one, greatest during competition-specific taper phases. Of the wellness variables collected, only muscle condition changed significantly over time (p < 0.001), with small-to-moderate correlations between load, change in creatine kinase and wellness. The publisher blocked direct retrieval of that paper, so this is summary-level reporting and should be weighted accordingly.

The three-subscale morning questionnaire that shape of study uses — mood disturbance, muscle condition disturbance, sleep quality disturbance, delivered by phone — is widely described as British boxing practice. We could not fetch a federation document confirming that, so it is attributed here to the boxing study and not to any governing body.

8. The scale, the descent, and the conditions on every percentage

The 2025 International Society of Sports Nutrition position stand on combat sports is explicit that monitoring the athlete's rate of weight descent is part of the job, and recommends taking measurements of body weight preceding and following training sessions to gauge sweat rate and fluid replenishment. On intake, its line is plain: "Daily food logs or dietary tracking apps are accessible to most individuals and are ideal for assessing and monitoring dietary intake."

For off-camp mass, the stand recommends a walk-around weight of 12–15% above the athlete's desired weight class, and gives its own worked example: a UFC male middleweight making 185 lb sustaining 207–212 lb. That framing is professional MMA, off-camp general preparation phase, and male.

Now the acute figures, with every condition physically attached, because they do not survive being separated from them.

The figures below were measured under professional MMA weigh-in conditions, with a 24 to 36 hour window between the weigh-in and competition; at a same-day weigh-in — the norm in much amateur boxing, BJJ and Muay Thai — they describe nothing. The stand states: "Suitable losses in body mass range from 6.7% at 72 h, 5.7% at 48 h, and 4.4% at 24 h, prior to weigh-in." Those are three answers to one question asked at three separate moments. They are not stages, they do not sequence, and they do not sum — an added figure of roughly 17% is a fabrication. The stand attaches them to the sentence: "During fight week, acute water loss strategies, including sauna, hot water immersion, and mummy wraps, can be used effectively with appropriate supervision (optimally ~2–4% of body mass within 24 h of weigh-in)." An athlete with under four hours between weigh-in and first bout cannot use these strategies at all, and the stand does not specify refeed protocols or a defined cohort behind the three percentages.

The same document states plainly that "The long-term effects of frequent weight cuts on health and performance are unknown, necessitating further research," and that "Few published studies address the complexities and unique training demands of combat sports." It does not recommend urine specific gravity.

One promoter-published claim circulates alongside these and should be named rather than repeated. A UFC Performance Institute volume is reported, in secondary coverage, to state that a fighter can lose 10 percent of body mass without putting health at risk. The primary document could not be retrieved — the volume one PDF fails on an expired TLS certificate and volume two is not openly hosted — it is a promoter publication rather than a peer-reviewed paper, and per the same secondary reporting the document elsewhere cites the World Health Organization position that 10% dehydration is life threatening and concedes that dehydration beyond 4–5% becomes difficult and dangerous. A document that contradicts itself is not a source, and no figure in this article is described as safe.

9. Sleep trackers, and what the meta-analysis did not measure

A 2025 meta-analysis pooled 24 studies and 798 participants comparing consumer wrist-worn devices against polysomnography, the laboratory standard.

The pooled biases: total sleep time −16.85 minutes (95% CI −26.33 to −7.38, 22 studies); sleep efficiency −4.69% (−7.08 to −2.30, 18 studies); sleep onset latency +2.57 minutes (0.61 to 4.54, 18 studies); wake after sleep onset +13.26 minutes (4.52 to 21.99, 20 studies). Fitbit devices specifically showed no significant difference in wake after sleep onset but overestimated latency by about 5.6 minutes; other devices measured latency accurately but overestimated wake after sleep onset by roughly 24 minutes.

Two conditions decide how much of that transfers. The cohorts are general population, not athletes. And sleep staging was not assessed in this meta-analysis at all — which means nothing here licenses any claim about the accuracy of a device's deep sleep or REM numbers, in either direction. Any percentage a fighter has been shown for staging error is not coming from this evidence base.

What the biases do support is a modest reading: a wrist device is plausibly useful for tracking whether the athlete is in bed for a consistent duration across a camp, and it is not a measurement of sleep architecture. On the order of a quarter-hour of systematic bias in total sleep time is small against the size of the changes a bad camp produces and large against the day-to-day differences people try to interpret.

10. Heart rate variability, and the formula this article will not print

Heart rate variability is the most heavily marketed monitoring variable in sport, and the honest position on it is uncomfortable.

The most-cited review of vagally-derived HRV in elite athletes concluded that most supporting research sits in recreational and well-trained rather than elite populations; that elite findings are equivocal, with both increases and decreases in HRV associated with negative adaptation; and that signs of positive adaptation, in the form of increased cardiorespiratory fitness, have been observed alongside atypical decreases in HRV. Its stated conclusion is that practical ways to use HRV to monitor training status in elite athletes are yet to be established. That review is endurance athletes, and we could only reach it at abstract level, so it is hedged here.

There is also a rule circulating in coaching and vendor material that a meaningful HRV change is 0.5 times the standard deviation of the seven-day rolling average. Every reachable statement of that rule sat on a vendor page, a coaching blog, or a figure thumbnail; the primary papers were behind authentication walls. We could not source it, so this article does not state it as established.

Note also where HRV sits in the systematic review of section 2: it appears in the list of objective measures that generally did not correlate with subjective well-being and were less sensitive to load in the studies that compared them directly. And no fetched source examines HRV monitoring in a combat sport at all.

11. Hydration, and the marker that flags almost everyone

Urine specific gravity is the most commonly used biochemical hydration marker in combat sports, and its problem is well described even in the paper that describes its use.

Recent investigations report a prevalence of "hypohydration" by urine specific gravity of around 90% in combat-sport athletes, frequently despite stable body mass — which the authors of that review treat as evidence of potential false-positive misclassification rather than as a description of a dehydrated population. The confounders they raise are specific: laboratory-derived thresholds applied in field conditions, inconsistent sampling, heavier solutes such as glucose and creatinine biasing density, and poorly characterised renal adaptation to dehydration in athletes. Their recommendation is to combine body-mass change, urine specific gravity and thirst rather than to use the marker alone.

That whole paragraph is hedged: the publisher blocked full-text retrieval and this is abstract-level content. It is included because a marker that flags nine in ten athletes as hypohydrated is either describing a crisis or describing itself, and a fighter deciding between those two readings should know the question exists.

Separately, a passing test at a commission's threshold is a licence to weigh in. It is not a statement about anything else, and the position stand above does not recommend the marker for camp monitoring at all.

12. A worked scenario

This is arithmetic on published figures. It is not a client, not a case study, and not advice.

Take Mara Delgado, an invented flyweight four weeks from a fight, competing at a 56.7 kg division. She started camp at 64.0 kg — 12.9% above her division, which sits inside the 12–15% off-camp band the position stand describes for professional MMA, though that band was framed around a male middleweight. She is now at 59.6 kg. Her log carries one open knee injury at severity 2, limiting, opened at 3, and one closed rib contusion.

What the scale can tell her. Over the last week her morning mass moved 0.4 kg. Against the measurement literature, that is a difficult week to interpret: in resistance-trained males, a 500 g meal plus 1 L of water moved bioimpedance fat-mass estimates by 1,400 g and DXA fat-free mass estimates by 1,211 g on the same day. A week's genuine change of that size is not distinguishable from the state of the athlete at the moment of measurement unless the measurement is standardised — overnight fasted, rested, under 500 g of acute food and fluid, same time, same conditions. Standardising it is worth more than a better scale. A single-week reading in either direction is best treated as one point in a series rather than as an event.

What the jump mat can tell her. If her countermovement jump reads 1 cm below last Tuesday's, that is inside the instrument. On a 40 cm jump, a 4% coefficient of variation is roughly 1.6 cm of typical noise, against a smallest worthwhile change of 8 to 15 mm depending on the device — so the noise is between one and two times the smallest change worth caring about, and the direction of a single reading carries no information. That comparison is drawn from 15 sports-science students, not fighters, and no combat-sport minimal detectable change has been published for the test.

What the questionnaire can tell her. If her load has been high for three weeks and her self-reported fatigue is climbing while nothing measurable has changed, the MMA study in section 7 is the closest published analogue: cortisol falling and hs-CRP declining in the final week while POMS fatigue reached its highest value and vigour its lowest. Ten men, three weeks, no control group.

What the load series can tell her. The junior-female boxing block ran from 3,322 ± 487 AU in week 1 to 1,239 ± 279 AU in week 7, with jump performance dipping across weeks 3 to 5 and partially recovering in week 6. A dip inside a planned block looks like that. A dip with no planned reduction beside it looks the same on the performance chart and means something else entirely, which is the reason the load number is worth the ten seconds it costs.

What the injury log can tell her. A severity that opened at 3 and now reads 2 is a record of what she could and could not do, dated. It is not a diagnosis and it does not describe the tissue. The distinction between the two axes is set out in injury severity and tissue grade.

The scenario stops here. Nothing above describes how to make weight, and none of the descent figures in section 8 were measured in an unsupervised athlete.

13. What breaks first

Monitoring systems do not usually fail because the measure was wrong. They fail because the measurement stopped happening.

The qualitative work on this is the most practically useful paper in the whole set: 30 participants — 8 athletes, 7 coaches and 15 sports science or medicine staff — across 20 programmes, 10 individual, 4 team and 6 youth. What determined whether self-report monitoring survived contact with a real programme was measure design (its accessibility, its timing, its interface) and social environment (coach buy-in, feedback to the athlete, and reinforcement). Not the measure's validity. Its friction, and whether anyone visibly did anything with it.

That study is qualitative, and its authors state that it did not attempt to reveal an exhaustive list of factors or to represent all end-users and sport settings. It reports no compliance percentages, and a widely circulated figure attaching an 84% adoption rate to this group of researchers does not appear in it; we could not reach a primary for that number and it is not printed here.

The practical implication is unglamorous. A one-minute daily self-report that the coach reads and responds to outperforms a comprehensive battery that runs for three weeks and then quietly stops. That is why the honest minimum for most camps is small: a load number after each session, a short daily wellness entry, a standardised morning mass, and a dated injury log — which is roughly the set Fighter Cut puts on one screen, because the failure mode being designed against is the athlete who stops entering.

14. The populations this evidence does not cover

The literature behind every figure above is adult, overwhelmingly male, and institutional.

The Olympic combat-sport testing literature is 61% male-only, 5% female-only and 28% both. The one all-female camp study here is junior, Turkish, n = 20, and did not monitor menstrual cycle status — a limitation its own authors call particularly relevant. The MMA biochemistry cohort is ten men. The body-composition error work is 32 resistance-trained males. The jump-reliability work is 15 male sports-science students. The sleep meta-analysis is 798 general-population participants of unstated athletic status.

None of that is adjustable with a coefficient. Body composition, total body water as a fraction of mass and cycle effects on fluid balance differ in ways no multiplier accommodates, and where a governing body has legislated for the difference it did so with a separate rule and a different number rather than a scaling factor — the NCAA's minimum wrestling weight thresholds are 5% body fat for men and 17% for women, assessed by a designated assessor behind a hydration gate.

Adolescents appear here only awkwardly. Several of the session-RPE cohorts are 13 to 19 year old taekwondo and karate athletes, which is a young sample rather than a deliberate youth evidence base. And no fetched source addresses monitoring in amateur or hobbyist competitors — who are the majority of BJJ and Muay Thai entrants and the readers with the least support around them.

Finally, the honest ceiling on the whole enterprise: nothing in this literature links a monitoring variable to a fight outcome. Predictive validity was examined in exactly one study across the entire Olympic combat-sport test literature. No source here shows that tracking anything improves competitive results. What monitoring is supported for is describing what happened and how the athlete responded to it, which is a smaller claim and a real one.

What we could not verify

• Any session-RPE validation in this article's five tagged sports. Every combat-sport validity coefficient we could reach is taekwondo, karate or fencing, several in adolescents. No fetched study validates session-RPE in MMA, professional boxing, Muay Thai, BJJ or wrestling.
• A minimal detectable change for almost any combat-sport-specific test. Three studies out of 39 investigated it. The MDC-adjacent numbers used above come from non-combat cohorts, and we did not find combat-sport values.
• The 1,960 g total-body-water shift attributed to the body-composition error study under non-standardised conditions. It did not appear in the abstract we could retrieve, which expresses that error as a coefficient of variation instead. Dropped from the article; the 1,400 g and 1,211 g figures carry the same argument and are confirmed.
• Any HRV monitoring evidence in a combat sport. None located. The review we relied on is endurance athletes, reached at abstract level only, and its own conclusion is that elite application is not established.
• The HRV smallest-worthwhile-change formula. Every reachable statement of the 0.5 × SD rule was vendor, coaching-blog or thumbnail level. Not printed.
• Any sleep-tracker validation in athletes, and any sleep-staging accuracy figure. The meta-analysis is general population and explicitly did not assess staging, so no REM or deep-sleep error percentage appears here.
• Grip strength as a camp monitoring variable. It is frequently named, but we located no primary source giving reliability, minimal detectable change or camp-monitoring data for it, so no number is attached to it.
• The 84% self-report adoption figure commonly attached to this group of researchers. It is not in the paper we fetched, which is qualitative and reports no compliance percentages at all.
• A federation document for the morning three-subscale wellness questionnaire. The design matches a published boxing study; we could not confirm a governing-body primary, so it is attributed to the study.
• The UFC Performance Institute primary document. The volume one PDF fails on an expired certificate and volume two is not openly hosted. Only secondary reporting was reachable.
• The long-term consequences of repeated weight cuts. The position stand states outright that these are unknown.
• That tracking anything improves fight outcomes. Predictive validity was examined in one study across the whole literature. Nothing here establishes it.

Sources

Sourced to

1. 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
2. Monitoring the athlete training response: subjective self-reported measures trump commonly used objective measures: a systematic review — Saw AE, Main LC, Gastin PB, British Journal of Sports Medicine, 2016;50(5):281–291. DOI 10.1136/bjsports-2015-094758, PMID 26423706
3. Monitoring Athletes Through Self-Report: Factors Influencing Implementation — Saw AE, Main LC, Gastin PB, Journal of Sports Science and Medicine, 2015;14(1):137–146. PMID 25729301
4. Session-RPE Method for Training Load Monitoring: Validity, Ecological Usefulness, and Influencing Factors — Haddad M, Stylianides G, Djaoui L, Dellal A, Chamari K, Frontiers in Neuroscience, 2017;11:612. DOI 10.3389/fnins.2017.00612
5. Tests for the Assessment of Sport-Specific Performance in Olympic Combat Sports: A Systematic Review With Practical Recommendations — Chaabene H, Negra Y, Bouguezzi R, Capranica L, Franchini E, et al., Frontiers in Physiology, 2018;9:386. DOI 10.3389/fphys.2018.00386
6. Impact of food and fluid intake on technical and biological measurement error in body composition assessment methods in athletes — Kerr A, Slater GJ, Byrne N, British Journal of Nutrition, 2017;117(4):591–601. DOI 10.1017/S0007114517000551
7. Countermovement Jump Analysis Using Different Portable Devices: Implications for Field Testing — Rago V, Brito J, Figueiredo P, et al., Sports (Basel), 2018;6(3):91. DOI 10.3390/sports6030091
8. Weekly Fluctuations in Internal Load and Neuromuscular Performance Across a 10-Week Training Period in Elite Female Boxers — Aydın AS, Altuğ T, Yılmaz C, Badau A, Söyler M, Life (Basel), 2026;16(3):386. DOI 10.3390/life16030386, PMID 41900905
9. Biochemical and psychological markers of fatigue and recovery in mixed martial arts athletes during strength and conditioning training — Ostapiuk-Karolczuk J, Dziewiecka H, Bojsa P, et al., Scientific Reports, 2025;15:24234. DOI 10.1038/s41598-025-09719-z
10. 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
11. Training Load Is Correlated with Changes in Creatine Kinase and Wellness over a 12-Week Multi-Stage Preparatory Training Block for a Major Competition in International Boxers — McCabe D, Martin D, McMahon G, Physiologia, 2023;3(4):43. DOI 10.3390/physiologia3040043 (publisher blocked direct retrieval; reported at summary level)
12. Urine specific gravity as an indicator of dehydration in Olympic combat sport athletes; considerations for research and practice — Zubac D, Reale R, Karnincic H, Sivric A, Jelaska I, European Journal of Sport Science, 2018;18(7):920–929. DOI 10.1080/17461391.2018.1468483 (abstract-level only)
13. Training adaptation and heart rate variability in elite endurance athletes: opening the door to effective monitoring — Plews DJ, Laursen PB, Stanley J, Kilding AE, Buchheit M, Sports Medicine, 2013;43(9):773–781. DOI 10.1007/s40279-013-0071-8, PMID 23852425 (abstract-level only)
14. UFC Performance Institute claims fighters can lose 10% body mass during fight week — Combat Sports Law, 17 May 2021; secondary reporting of a promoter-published, non-peer-reviewed document whose primary could not be retrieved

Read next