Evidence Base — The Studies Behind Every Answer

What Five-Hour Nights Actually Do to Testosterone Researchers found one week of five-hour nights lowered daytime testosterone by 10–15% in ten young men. Other studies did not confirm it. The honest answer requires both findings. Leproult et al. · 2011 · JAMA · n=10

The rice was along for the ride Researchers pooled eleven controlled trials on carbs and muscle growth. The finding challenges what most lifters have been doing at every meal. 2:50 · Henselmans et al. · 2026 · Sports Medicine · n=0

The Phantom Dinner Keto Couldn't Feel Twenty adults on a metabolic ward ate two diets whose meals were rated with no significant difference in pleasantness. On one diet they quietly consumed about an extra dinner's worth of calories a day, on average, without rating themselves significantly hungrier on the other. 2:46 · Hall et al. · 2021 · Nature Medicine · n=20

Heavy Drinking Went With Excess Weight. Moderate Didn't. 127 studies, 1.6 million people: heavy drinkers had higher odds of overweight, moderate drinkers didn't, and every result was rated very low certainty. Lab studies add one number you can plan for: about 82 calories of extra food per drinking occasion. 2:13 · M et al. · 2021 · SSRN (preprint, not peer reviewed) · n=1,620,177

The squat stayed where it was Six controlled trials pooled 131 trained lifters on keto versus normal diets. The one-rep max data told a story most gym advice gets wrong. 2:33 · Vargas-Molina et al. · 2024 · Nutrients · n=131

You Spent 17 Hours Last Year on Advice From the 1960s Researchers pooled every controlled experiment on eating six meals a day for metabolism. The result across nine health markers was the same: nothing happened. 3:09 · Blazey et al. · 2023 · International Journal of Behavioral Nutrition and Physical Activity · n=279

The yogurt was never the enemy A meta-analysis pooled every sugar-and-weight trial ever run. The finding that dissolves a decade of label anxiety landed in a single number — and it's not where parents have been looking. 3:36 · Morenga et al. · 2013 · BMJ

Three biological systems responded when they ate the same food later Harvard researchers fed 16 adults identical meals on two different schedules. The late schedule triggered a triple hit — doubled hunger, reduced calorie burn, and shifted fat cell gene expression — that no other study has captured simultaneously. 3:26 · Vujović et al. · 2022 · Cell Metabolism · n=16

The Gym-Floor Myth 43 Studies Put to the Test The interference effect — the idea that cardio cancels out your strength gains — was tested across 43 studies and 1,090 people. In the 15 that measured muscle growth, the effect landed at one hundredth of a standard deviation. 2:52 · Schumann et al. · 2022 · Sports Medicine · n=1,090

The 0.69% That Rewrites the Weight Room Rules A 29-study meta-analysis found that the difference in muscle growth between men and women is so small, researchers couldn't determine which sex grows faster. The fitness industry built an entire product category on a distinction that barely exists. 2:48 · Refalo et al. · 2025 · PeerJ · n=2,815

The 1 kg Argument Thirty-six studies tried to settle the cardio-versus-weights debate for fat loss. The gap they found would fit on a kitchen scale — and when both sides did equal work, it vanished entirely. 3:43 · Lafontant et al. · 2025 · Journal of the International Society of Sports Nutrition · n=1,564

The Exercise Your Heart Can't Tell Apart Fifty-one trials compared running, lifting, and combined training for cardiovascular fitness after 60. The formal test for a winner came back empty — and resistance training was the most consistent performer. 2:35 · Azimkhani et al. · 2026 · Biological Research for Nursing · n=3,152

Your Set Counter Is Lying to You A 67-study meta-analysis found that the way every training app counts sets is wrong — and the real volume picture changes everything about diminishing returns. 2:57 · Pelland et al. · 2025 · Sports Medicine · n=67

Light vs Heavy Weights: Similar Growth When Every Set Hits Failure A meta-analysis of 21 studies found similar muscle growth with light and heavy weights, with a slight edge for heavy that was not statistically significant — provided one specific condition was met. 3:08 · Schoenfeld et al. · 2017 · Journal of Strength and Conditioning Research

No Significant Gain on Three Strength Tests. The Fourth Rests on Two Trials. Researchers pooled ten trials of vitamin D in 354 athletes. Blood levels rose by nearly 15 ng/mL. Bench, jump, and grip showed no significant gain. Quadriceps contraction did, in two small trials. 2:11 · Han et al. · 2024 · Frontiers in Nutrition · n=354

Resting Metabolism Did Not Explain It. Everyday Movement Did. Sixteen people ate the same extra 1,000 calories a day for eight weeks. The changes in their resting metabolism did not account for who gained fat. The rise in their everyday movement outside exercise predicted who resisted it. 2:42 · Levine et al. · 1999 · Science · n=16

The Number Your App Cannot See The equation behind MyFitnessPal was built to replace math calibrated on 1919 bodies. It explains 71% of your metabolism. The researchers tried everything to crack the other 29%. 3:00 · MD et al. · 1990 · The American Journal of Clinical Nutrition · n=498

You Burn What a Wolf Your Size Burns Isotope-labeled water tracked 647 people over two decades. What researchers found about activity, obesity, and wild mammals challenges every 'sitting is the new smoking' headline. 2:28 · KR et al. · 2013 · Frontiers in Physiology · n=647

Your Scale Celebrates the Exercise That Costs You the Most Muscle A 62-study ranking of eight exercise types during a calorie deficit found that the one dropping the most weight also ranked near-last for keeping muscle — and the fix contradicts universal gym advice. 2:34 · Xie et al. · 2025 · Frontiers in Nutrition · n=4,429

The Three Things Your Tracking App Doesn't Know Researchers built a model that predicted meal intake using three food characteristics no calorie app measures. One of them — protein — pulled in the direction nobody expected. 2:57 · TL et al. · 2023 · Nature Food · n=35

Equal Taste, 508 Hidden Calories They rated both diets the same. Then ate an entire extra meal's worth of calories on ultra-processed food — every day, for two weeks, without noticing. 3:18 · Hall et al. · 2019 · Cell Metabolism · n=20

The gym anxiety that is probably misplaced Seven of eleven studies measured whether exercise order affects muscle growth. The combined difference was close to zero. Then researchers separated strength from growth, and the data told a completely different story. 1:51 · JP et al. · 2021 · European Journal of Sport Science · n=268

The Mental Cue That Nearly Doubled Bicep Growth Has a Weight Limit Researchers tested whether thinking 'squeeze the muscle' during a rep could change eight weeks of growth. For biceps, it nearly doubled the results. For quads, the same cue made no measurable difference. 2:07 · BJ et al. · 2018 · European Journal of Sport Science · n=0

Everyone Follows This Rule. It Is Worth 2.8%. Twenty-four studies tested whether full range of motion actually matters for muscle growth. The overall advantage was 2.8%. Then researchers checked which kind of partial rep the evidence had been testing. 2:32 · M et al. · 2023 · International Journal of Strength and Conditioning · n=658

Leg press and leg extension each grow a different quad region more — one exercise may not be enough Burke's RCT found that the same person's quad grew most in different places depending on which exercise that leg performed. The front quad responded more to extensions. The outer quad responded more to press. Each exercise led in one region. 2:16 · Burke et al. · 2025 · Journal of Science in Sport and Exercise · n=28

Same total muscle. Completely different shape. Seven studies agree: isolation and compound exercises build identical total muscle. But one study found they build it in completely different places. 1:57 · Rosa et al. · 2023 · Strength & Conditioning Journal · n=193

No muscle-growth gap behind the gym equipment hierarchy Haugen's meta-analysis pooled 13 studies and 1,016 people comparing free weights and machines. In the five that measured muscle growth, the gap was too small to tell apart from chance. For upper-body strength on each group's own equipment, the hierarchy might run backwards. 3:01 · Haugen et al. · 2023 · BMC Sports Science, Medicine and Rehabilitation · n=1,016

18 Months of Free Meals, and a $25 Cheque That Didn't Move the Scale In a 1993 trial, people handed free breakfasts and dinners on top of a diet class lost far more by the one-year weigh-in. Cash for results didn't move the scale, and the weight still crept back. 3:40 · Jeffery et al. · 1993 · Journal of Consulting and Clinical Psychology · n=159

Shift Work Weight: The 13-Year Threshold A hidden timeline. A sex split two studies completely disagree on. And a causal question that 693,449 workers across 63 studies still can’t settle. 2:45 · Łagowska et al. · 2024 · American Journal of Human Biology · n=693,449

One Cookie and a Demolished Meal Schedule The calorie gap between shift workers and day workers is 63 per day. The eating pattern gap is 44 percent versus 6 percent for night snacking, three-meal collapse, and food quality inversion across every study that measured it. 2:20 · Clark et al. · 2023 · Advances in Nutrition · n=18,196

Fasting vs Dieting: Under 1 kg Apart The largest trial ever in night shift workers tested fasting against calorie cutting for six months. The difference was so small it was indistinguishable from noise — but the good news was hiding in the data every group shared. 2:02 · Bonham et al. · 2025 · eBioMedicine · n=250

The Test That Barely Tracks Its Own Treatment A meta-analysis of 24 studies found that the standard screening test for sarcopenia, grip strength, showed only a limited response to exercise. The leg strength test showed a large effect. 2:36 · Yan et al. · 2025 · Aging Clinical and Experimental Research · n=951

The Creatine Evidence Was Built on People Who Don't Buy It A meta-analysis of 1,093 adults revealed that 69% of the creatine evidence base is women over 55 — the demographic least likely to use the supplement. The strength benefit landed where aging hits hardest, with a safety record the kidney concern doesn't survive. 2:42 · Sharifian et al. · 2025 · European Review of Aging and Physical Activity · n=1,093

151 Trials Say Your 75-Minute Workout Is Triple the Volume for Half the Result The largest network meta-analysis ever conducted on training volume in older adults found a 2.22-to-1 advantage in muscle growth for shorter programs. The data doesn't just suggest doing less — it identifies exactly how much less. 2:46 · Radaelli et al. · 2025 · Sports Medicine · n=6,306

Same Weight Lost, Different Body After 50 20 trials show the protein threshold that determines whether weight loss after 50 costs you muscle — or preserves it. 1:59 · Kim et al. · 2016 · Nutrition Reviews · n=20

One Workout Plan, Three Ways In. One Lost Almost Half. Researchers gave 112 experienced lifters one 10-week plan through a coach, an app or a PDF. Far fewer made it to the end with the PDF, and the coach made a measurable difference on one of the two lifts tested. 2:58 · Gavanda et al. · 2025 · Journal of Strength and Conditioning Research · n=79

An Extra Litre, Zero Return Scheduled drinkers consumed an extra litre per session and performed slightly worse. Seven studies found that drinking to thirst worked at least as well as following any plan. 2:03 · Goulet et al. · 2019 · Sports Medicine · n=82

The Factory Was Running the Whole Time Researchers reviewed every study linking estrogen to muscle loss. The muscle-building machinery they expected to find failing was doing the opposite. 2:14 · Menzies et al. · 2026 · Journal of Cachexia, Sarcopenia and Muscle

Chosen for Attention, the One-to-One Option Came With Less Time Seven trials put weight-loss groups ahead of one-to-one appointments at a year. The catch is in the review's table: where hours were counted, the groups got more. 3:51 · Abbott et al. · 2021 · Journal of Human Nutrition and Dietetics · n=2,576

The Coaching Trial That Counted Most Was Fully Automated A 2023 review of 14 trials asked whether a coach adds weight loss to a tracking app. Against the same kind of app alone, the answer came out too uncertain to count, and two programmes it counted as coaching were marked automated. 4:02 · Chew et al. · 2023 · Journal of Medical Internet Research · n=2,478

Your protein shake just hit a ceiling Researchers pooled 49 studies and 1,863 people to find where more protein stops building more muscle. The number landed lower than almost everyone on the gym floor thinks — and the gap between what you eat and what the evidence says you need is probably one chicken breast. 2:42 · Morton et al. · 2018 · British Journal of Sports Medicine · n=1,863

The Afterburn: One Butter Pat Per Day Twenty-nine studies proved the afterburn effect is real. Then the scientists said it does not matter. The fat-loss advantage of HIIT over regular cardio is five grams per day — and the researchers who found it called it clinically meaningless. 3:35 · al. et al. · 2023 · International Journal of Environmental Research and Public Health · n=807

Same calories. Different body. The study nobody's quoting. Researchers pooled 24 trials and 1,063 dieters. The protein group and the standard group ate the same number of calories — but what happened inside their bodies diverged in a way the scale never revealed. 2:40 · Wycherley et al. · 2012 · The American Journal of Clinical Nutrition · n=1,063

The Metabolic Boogeyman Is One Spoonful Millions of dieters believe their metabolism crashed by hundreds of calories. A systematic review of 33 studies on 2,528 people found the real number — and it fits on a tablespoon. 2:49 · Nunes et al. · 2021 · British Journal of Nutrition · n=2,528

The Creatine Tub You Stopped Using Researchers pooled 143 trials to answer the creatine question: real muscle or just water? The answer involves an MRI scanner, a price comparison, and a myth about loading. 3:00 · Pashayee-Khamene et al. · 2024 · Journal of the International Society of Sports Nutrition · n=3,655

278 Extra Calories — or Zero? A controlled-feeding trial found low-carb burns 278 extra calories a day after weight loss. Then another team reanalyzed the raw data and the number kept getting smaller. The metabolic edge debate in four numbers. 3:40 · Ebbeling et al. · 2018 · The BMJ · n=164

Double the Vitamin E. Zero Vitamin C Lost. The Frozen Bag Has a Secret. UC Davis researchers split eight produce items at harvest and tested eleven nutrients. The frozen versions held up better than the fresh aisle would have you believe — though one orange pigment tells a very different story. 2:19 · Bouzari et al. · 2015 · Journal of Agricultural and Food Chemistry

The Number on Every Turmeric Label Comes From Dividing by Nearly Zero The 2000% absorption boost printed on millions of supplement bottles was calculated by dividing by a baseline so low the equipment could barely detect it. Four labs tried to replicate it. None could. 2:05 · Shoba et al. · 1998 · Planta Medica · n=8

Your Body Charges a Tax on Every Gram of Protein A meta-analysis of 52 studies measured exactly how many extra calories your body burns digesting protein. The number is bigger than most people think, but the boost works differently than the fitness industry promises. 2:55 · LL et al. · 2024 · Advances in Nutrition · n=1,232

Muscle Memory After 60: The Comeback Was Faster Than the Original Build Men aged 58 to 77 lost just 5 to 15 percent of their strength after three months off. When they returned, the comeback took a third less time and ended at a higher peak. The movement they'd trained most came back fastest. 2:13 · Blocquiaux et al. · 2020 · Experimental Gerontology · n=0

The Post-Diet Muscle Split Nobody Saw Coming Both groups lost the same fat. But researchers kept measuring for 10 more weeks — and found the only body composition difference happened after the diet ended, not during it. 2:33 · Conlin et al. · 2021 · Journal of the International Society of Sports Nutrition · n=23

Everything the Internet Knows About Fat and Testosterone Comes From 206 Men A meta-analysis found the testosterone drop was real — then a bigger study found it wasn't. The research behind the internet's most viral testosterone claim is thinner than anyone sharing it realizes. 2:05 · Whittaker et al. · 2021 · Journal of Steroid Biochemistry and Molecular Biology · n=206

They Mapped 124 Factors. One Outperformed All Others. Researchers tested every possible predictor of weight regain across 31,741 people — from metabolism to demographics to psychology. The most consistent one is something most people already own and stopped using. 2:19 · Varkevisser et al. · 2019 · Obesity Reviews · n=31,741

The Number on the Bottle Was a Guess Ninety researchers tracked water through 5,604 people with isotope tracers. The sex-based guideline collapsed when body composition entered the model, and the best prediction equation still could not explain more than half. 3:07 · Yamada et al. · 2022 · Science · n=5,604

The Rep Tax Nobody Named Twenty-eight studies found dehydration steals reps before it touches max strength — and how much water you lost had almost nothing to do with the damage. 2:24 · Savoie et al. · 2015 · Sports Medicine · n=28

Everything the Electrolyte Ad Does Not Tell You A systematic review of 38 studies found that pre-workout fluid loading can improve endurance in heat — but none tested a gym session, two-thirds reported gut symptoms, and almost nobody tested was female. 2:07 · Jardine et al. · 2023 · Sports Medicine 53(11):2111-2134 · n=403

The Water Target Earned Less Than a Glass Weighs Eight controlled trials tested whether water helps overweight adults lose weight. The pooled answer was nearly zero — and the one swap everyone recommends was linked to gaining 1.8 kilograms. 2:03 · Chen et al. · 2024 · Nutrients

Five ways to check and one answer nobody expected Researchers tested vegans and meat-eaters with five independent muscle measurements over 12 weeks. The results agreed — but the part nobody mentions changed the whole conversation. 2:18 · Hevia-Larraín et al. · 2021 · Sports Medicine · n=38

Your Whey Premium Bought You a Rounding Error A research team pooled every qualifying trial ever published on plant versus animal protein for muscle, all 43 of them. The soy-versus-whey data rewrites the gym-floor debate. 2:47 · Reid-McCann et al. · 2025 · Nutrition Reviews · n=1,538

Soy and Testosterone: What the Dose-Response Data Shows A meta-analysis of 11 trials tested soy isoflavones from 49 to 164 mg/day. Testosterone showed zero response at every dose. Estradiol had a threshold, above food intake. 2:22 · Rajaie et al. · 2025 · Food Frontiers · n=489

Three Plants That Matched Milk in a Lab A wheat + corn + pea blend delivered 91% fewer amino acids to the blood — and built the same amount of muscle. The scoreboard on every protein tub measures the wrong thing. 2:00 · Pinckaers et al. · 2022 · The Journal of Nutrition · n=24

The MPS Gap Was Measured. It Was 0.004. 12 studies compared the muscle-building spike from plant vs animal protein. The gap sat at the 0.35th percentile of variation, measured in a window covering one-fifth of the day. 2:22 · Mendes et al. · 2026 · Journal of the Academy of Nutrition and Dietetics · n=321

The Supplement That Works Harder When You Skip the Steak Same dose, more than double the response. A systematic review of nine studies reveals why vegetarian muscles handle creatine differently — and where the evidence stops being clear. 3:05 · Kaviani et al. · 2020 · International Journal of Environmental Research and Public Health · n=355

The Dusty Tub on the Bottom Shelf Won A meta-analysis of 62 trials found that the cheapest fiber supplement outperformed the trendiest one by two to one — and the expensive option's result wasn't even reliable. Plus: the first eight weeks are a dead zone. 2:43 · Jovanovski et al. · 2020 · The American Journal of Clinical Nutrition · n=3,877

Eight Hormones Voted for Regain. One Held the Line. Researchers tracked nine appetite hormones for a year after weight loss. The fullness signal didn't recover. It got worse. And neither study found those changes predicted who actually regained. 1:59 · Sumithran et al. · 2011 · New England Journal of Medicine · n=34

The Portion Plate Was Fighting the Wrong Battle A meta-analysis of 31 studies found that swapping meals to lower calorie density produced the largest sustained calorie savings researchers have measured — and your body barely compensates. 2:49 · Robinson et al. · 2022 · International Journal of Behavioral Nutrition and Physical Activity · n=31

The Protein Detector Your Body Never Told You About 38 trials found your body measures protein as a percentage, not grams, and below a specific threshold, it forces you to eat more of everything else. Neither carbs nor fat had any effect. 2:04 · Gosby et al. · 2014 · Obesity Reviews

107 Fiber Treatments Tested. Most Failed. Researchers ranked 38 fiber types by their effect on appetite. The most popular supplement — the one trending as a weight-loss hack — was among the worst performers. 2:06 · Clark et al. · 2013 · Journal of the American College of Nutrition

The Fat That Alcohol Doesn't Make Researchers tracked where every calorie of alcohol actually went inside the body. The answer upended the most common belief about drinking and belly fat — and revealed a mechanism nobody on social media got right. JJ et al. · 1988 · Journal of Clinical Investigation · n=14

The Exact Price Tag of Drinking After the Gym Researchers put the same eight lifters through the same workout three times — then measured what happened to muscle building when recovery included twelve drinks. Protein helped. But not how you'd expect. Parr et al. · 2014 · PLoS ONE · n=8

Your Nightcap's 2.4-to-1 Exchange Rate 27 sleep studies found the exact dose where alcohol starts disrupting brain recovery. The catch: the people in the studies couldn't feel it happening. C et al. · 2024 · Sleep Medicine Reviews · n=529

Your Body Has a Calorie Ceiling — And Your Apple Watch Doesn't Know About It A study of 332 adults found that above moderate activity, more exercise didn't mean more calories burned. The data behind your body's invisible energy budget — and why it hits hardest on the people trying hardest to lose weight. Pontzer et al. · 2016 · Current Biology · n=332

The Coin Flip That Ended the Zone 2 War Researchers pooled 23 studies on exercise intensity in older adults. The gap between going hard and going moderate came out smaller than the energy cost of standing up from a chair. Fosstveit et al. · 2024 · Scandinavian Journal of Medicine & Science in Sports · n=0

The Step Count Where Your Tracker Stops Being Right A 57-study review of 161,176 adults found exactly where step-count benefits flatten — and it's thousands of steps below the number your fitness app has been chasing. D et al. · 2025 · The Lancet Public Health · n=161

The Diet War That Ended in a Jar of Peanut Butter Stanford tracked 609 adults for a year on low-carb vs low-fat. The entire difference was less than a jar of peanut butter — and even that might not be real. Gardner et al. · 2018 · JAMA · n=609

The Iron Your Cereal Box Promises and the Iron Your Body Actually Gets A 42-study review measured how much mineral nutrition your body actually collects from grains and beans. The gap between the label and the blood turned out to be larger than anyone expected, and the fix turned out to be sitting in kitchens for centuries. Chondrou et al. · 2024 · Nutrients

Your Morning Coffee Is Stealing Your Iron The same iron on your plate gets absorbed at wildly different rates depending on what else you eat with it. A validated algorithm mapped the eight factors — and three of the biggest are in the most common breakfast on earth. Hallberg et al. · 2000 · American Journal of Clinical Nutrition · n=310

The Cold Rice Hack Used the Wrong Starch A Nature Metabolism trial showed real weight loss from resistant starch — then the researchers tested mice with no gut bacteria, and the starch did nothing. The viral version skipped that part. al. et al. · 2024 · Nature Metabolism · n=37

The Dressing That Cancelled Every Vitamin in the Bowl Researchers analyzed 39 studies on dietary fat and carotenoid absorption. What they found about fat-free dressing wasn't a percentage — it was a zero. Yao et al. · 2022 · Nutrition Reviews · n=39

Raw Lost 26 of 27 Antioxidant Tests Italian researchers measured what happens to antioxidant compounds when you boil, steam, and fry three common vegetables. The results upend the most common cooking advice. Miglio et al. · 2008 · Journal of Agricultural and Food Chemistry

McMaster Called It 'Hype.' Three Evidence Domains Explain Why Hormones Don't Predict Muscle Growth. A narrative review from one of the world's most respected exercise science labs examined whether hormones drive muscle growth across three independent evidence domains: post-workout spikes, sex-based testosterone differences, and menstrual cycle periodization. The answer was the same from all three. Every et al. · 2024 · Exercise and Sport Sciences Reviews

The Myth Inside the Meta-Analysis A meta-analysis pooled every study on the pill and muscle growth. The answer challenges a fear that started with a conference poster and became a TikTok consensus. Nolan et al. · 2023 · Sports Medicine · n=325

78 Studies, One Barbell Collar: The Real Size of Your Cycle's Effect The largest meta-analysis ever done on menstrual cycles and exercise found something the cycle-syncing industry built 294 million TikTok views around. The effect was smaller than the clip you put on the bar. McNulty et al. · 2020 · Sports Medicine · n=1,193

Menopause Changed Everything Except Your Weight A 17-year study tracked 1,246 women with body scans through menopause. Fat gain nearly doubled. Lean mass reversed. The scale showed nothing. Greendale et al. · 2019 · JCI Insight · n=1,246

The Fear That's Never Once Shown Up in a Lab Twenty-two studies tried to break lifters with overtraining. What researchers found, and didn't find, changes how you should read your next flat week. Grandou et al. · 2020 · Sports Medicine

The 0-for-6 Scorecard Your Supplement Aisle Hides Researchers tested your multivitamin across six body systems — cancer, heart, immunity, brain, eyes, and mortality. The score was not what the label promised. Wang et al. · 2025 · Ageing Research Reviews · n=5,535,426

14,600 days of stable metabolism — the age-30 crash never happened The largest metabolic database ever assembled shows your metabolism holds from 20 to 60. The real decline starts 33 years later than the myth claims, and it is cellular. P et al. · 2021 · Science · n=6,421

The protein gap nobody told you about after 40 Researchers found that the per-meal protein dose that maxes out muscle building in younger adults only gets older adults 60 percent of the way there. But the peak building rate? Identical. Moore et al. · 2015 · Journals of Gerontology - Series A Biological Sciences and Medical Sciences · n=108

Her Scale Said Nothing Changed. 101 Studies Showed Why. Researchers pooled 5,697 postmenopausal women and found that one exercise type rewrites body composition without moving the number on the bathroom floor. The data also revealed why age is irrelevant to the result. Khalafi et al. · 2023 · Frontiers in Endocrinology · n=101

The Best Diet Beats the Worst by One Grape Per Day Researchers pooled 121 trials and 21,942 people to compare every popular diet. The gap between the best and mid-ranked diet is absurdly small — and a separate trial reveals what actually predicts whether a diet works for you. Ge et al. · 2020 · BMJ · n=21

The Bag of Flour That Ended the Low-Fat Debate A Cochrane review pooled 37 trials and 57,079 adults to answer the question every gym bro has argued about. The answer weighed less than a bag of flour. L et al. · 2020 · Cochrane Database of Systematic Reviews · n=57,079

Same protein, different containers — one split built 25% more muscle Researchers tested the same people on two meal splits with identical total protein. The way they divided it changed their muscle building by a quarter — and the meal that mattered most was the one they gave least thought to. Mamerow et al. · 2014 · The Journal of Nutrition · n=8

The Post-Workout Shake Ritual That Was Optional All Along A meta-analysis of 23 studies found that the entire timing effect vanished once researchers controlled for a single variable. The lead author's response to his own findings surprised everyone. Schoenfeld et al. · 2013 · Journal of the International Society of Sports Nutrition · n=525

Their Testosterone Crashed 21%. Their Muscle Didn't Care. Researchers put 34 experienced lifters on 16:8 for eight weeks with matched calories and protein. The DXA scans told a story that rewrites what the fitness internet believes about fasting, hormones, and muscle. Moro et al. · 2016 · Journal of Translational Medicine · n=34

The man who owned the patent funded the only study that said ZMA works A 33.5% testosterone claim traces to one industry-funded study with 53% dropout. Two independent labs found zero. But athletes might actually need more zinc — just not from a branded tub. Te et al. · 2023 · Journal of Trace Elements in Medicine and Biology · n=38

Every Study Said Yes. Then Came the Asterisks. A systematic review filtered 1,254 papers on magnesium and soreness down to four. Every survivor found magnesium helped. But the protocol everyone recommends was never actually tested, and the evidence rests on 73 people. Tarsitano et al. · 2024 · Journal of Translational Medicine · n=73

The Iron Math Your Supplement Label Doesn't Show Twenty-three studies, one dose threshold, and a math problem that changes what "100% Daily Value" actually means for anyone who trains. M et al. · 2024 · Journal of Sport and Health Science · n=669

Fell Asleep Faster. Didn’t Sleep Longer. A meta-analysis pooled every trial on magnesium and sleep. One outcome improved. The other didn’t. The evidence base is thinner than the marketing suggests. Mah et al. · 2021 · BMC Complementary Medicine and Therapies · n=151

1,053 Calories Your Diary Cannot See Scientists used nuclear water to track every calorie for 14 days. The gap between what these dieters reported and what they actually ate explains why the scale never moved. SW et al. · 1992 · New England Journal of Medicine · n=10

What Biggest Loser Did to Their Metabolism Fourteen contestants lost massive weight on The Biggest Loser. Six years later, their resting metabolism was still burning 499 fewer calories per day — and the contestants who kept the most weight off had it worst. Fothergill et al. · 2016 · Obesity · n=14

Your Apple Watch calorie count is 28% wrong A 56-study meta-analysis found the Apple Watch overestimates calories by 28 percent while getting heart rate and steps right. No brand does better. Kang et al. · 2025 · Physiological Measurement · n=56

The Recovery Tool That Only Recovers One Thing A 21-study meta-analysis found foam rolling reliably reduces how sore you feel. The part that surprises: the recovery benefits most gym-goers expect did not reach statistical significance. Wiewelhove et al. · 2019 · Frontiers in Physiology · n=454

957-to-43 Against Your Ice Bath An 8-study meta-analysis calculates the probability your post-workout cold plunge is blunting your gains. When they tested both conditions inside the same body, the ice bath leg built a third of the muscle. 144s · Piñero et al. · 2024 · European Journal of Sport Science

Half the Muscle Growth, None of the Warning Signs Researchers gave young adults the ibuprofen dose printed on every bottle and tracked their muscles for eight weeks. What the MRI scans revealed cost them roughly half their gains — and their workouts never showed a single sign. Lilja et al. · 2018 · Acta Physiologica · n=31

The 16x Growth Hormone Claim Has Three Problems Nobody Mentions A systematic review checked forty sauna studies across twelve countries. The viral growth hormone claim traces to a single 1986 experiment that used two hours of heat exposure per day. What researchers found instead changes the conversation. Hussain et al. · 2018 · Evidence-Based Complementary and Alternative Medicine

Four Tests. Four Barely-Passing Grades. A 12-study meta-analysis found compression garments produce moderate effects on every recovery measure — but the scores barely cleared the threshold. And a newer analysis from the same lead researcher found nothing. Hill et al. · 2014 · British Journal of Sports Medicine · n=0

99 Studies Ranked Every Recovery Method. Your Routine Is Backwards. A meta-analysis compared ten recovery methods head-to-head for the first time. The method most lifters skip ranked first. The method most lifters do every day ranked last. Dupuy et al. · 2018 · Frontiers in Physiology · n=1,188

The Math Behind Your Month-Four Plateau Researchers built a mathematical model to test whether metabolism or compliance explains the six-month plateau. The equation settled the debate — and the answer will probably sound familiar. Thomas et al. · 2014 · American Journal of Clinical Nutrition · n=5,211

The App You Quit Was Already Working Researchers tracked what happened across 12 randomized trials and 1,190 people who kept logging their food. The per-meal math changes how you think about the app sitting unused on your phone. Berry et al. · 2021 · Obesity Reviews · n=1,190

Your Cooking Oil Is a Body Composition Variable Same surplus, same weight on the scale. But MRI scans revealed one cooking oil was building muscle while the other was storing fat — and nobody's scale caught the difference. Rosqvist et al. · 2014 · Diabetes · n=37

23 Studies Just Cleared Your Evening Workout The largest meta-analysis ever conducted on evening exercise and sleep found zero significant harm, with one specific, easy-to-avoid exception. Stutz et al. · 2019 · Sports Medicine · n=275

The Scale Lied: Same Diet, Opposite Results Researchers put the same people on the same calorie-restricted diet twice. Both times the scale dropped by 3 kg. But one group lost fat while the other lost muscle, and sleep was the only thing that changed. Nedeltcheva et al. · 2010 · Annals of Internal Medicine · n=10

Your Protein Shake Couldn't Fix What Sleep Broke Researchers fed the same protein meals to the same people after sleeping and after staying up all night. The sleep-deprived group built 18% less muscle — not because the food was wrong, but because the muscle couldn't use it. Lamon et al. · 2021 · Physiological Reports · n=13

They Slept In All Weekend. It Fixed 9%. A controlled lab study found weekend catch-up sleep barely dented the week's sleep debt — and the group that tried to recover ended up with worse insulin sensitivity than the group that never caught up. Depner et al. · 2019 · Current Biology · n=36

Your Strength Survives a Bad Night. Your Accuracy Doesn't. 77 studies measured exactly what short sleep costs every kind of workout — and the gap between what breaks and what holds is bigger than anyone admits. Craven et al. · 2022 · Sports Medicine · n=959

Your Body Fat Readings Looked Normal. The CT Scanner Disagreed. Researchers restricted sleep to four hours and measured body composition two ways. Standard tools saw nothing. CT imaging found 11% more fat wrapping the organs — and recovery sleep didn't reverse it. Covassin et al. · 2022 · Journal of the American College of Cardiology · n=12

The Night Your Brain Ordered an Extra Meal A meta-analysis of eleven studies found that sleep-deprived people eat a specific, measurable calorie surplus — and the explanation most health content gives for why is the wrong one. HK et al. · 2017 · European Journal of Clinical Nutrition · n=496

The Supplement 19 Million People Take for Muscle — Tested 15 Ways Researchers pooled every trial that measured fish oil's effect on muscle protein building. They tested dose, age, duration, and training status. The result was the same every time. Therdyothin et al. · 2024 · Nutrition Reviews · n=188

The 3.3 kg Receipt for Calorie Counting Oxford researchers put a number on the calorie counting debate across 37 studies. The answer satisfied neither side — and a finding about motivation shocked both. Hartmann-Boyce et al. · 2014 · Obesity Reviews · n=13,453

The Bag of Sugar That Ends the Diet Wars Sixty-one trials and 6,925 people measured the exact extra weight you lose by cutting carbs. The number fits in one hand — and it rewrites what you thought you knew about restriction. Naude et al. · 2022 · Cochrane Database of Systematic Reviews · n=6,925

The Protein That Wasn't Supposed to Work Collagen has almost none of the amino acid experts say you need for muscle. Researchers graded it on every training claim anyway — and the report card has grades nobody expected. Bischof et al. · 2024 · Sports Medicine · n=768

Three Out of Four Ingredients Had Nothing Researchers checked every ingredient in 50 testosterone boosters against the published research. The gap between what the labels claim and what the science shows is a canyon — and some ingredients may work against you. Clemesha et al. · 2020 · World Journal of Men's Health · n=50

The $50 Supplement That Lost to Free Researchers compared fat burner supplements to plain diet and exercise across 21 studies. The supplements didn't just fail to win — they lost at statistical significance. The confidence intervals tell the full story. Clark et al. · 2021 · Nutrition and Health · n=2,359

$480 a Year for a 4.5% Chance A systematic review pooled 22 BCAA trials. For muscle growth, only one came back positive. The research tells a very different story than the label. Julea et al. · 2025 · Cureus · n=511

They Ranked Every Protein Tub. Most Were Expensive Sugar Pills. The largest head-to-head comparison of protein supplements just tested all thirteen types against a placebo. The number that survived will change how you look at the supplement aisle. Drummond et al. · 2026 · Translational Sports Medicine · n=4,755

The $40 Tub Sits on a Line Scientists Drew Between Trivial and Real Ten crossover trials measured what pre-workout caffeine actually produces. The strength boost is real — and its size might change how you think about the price tag. J et al. · 2018 · Journal of the International Society of Sports Nutrition · n=149

The Willpower Tax Nobody Warned You About Twelve studies tested the one thing every dieter is afraid to try. The group that pushed straight through paid a metabolic price for zero extra fat loss. Poon et al. · 2024 · Nutrition Reviews · n=881

The Slow-Carb Tax: What Your Oat Upgrade Actually Bought Fourteen long-term trials pooled nearly 1,800 people on low-GI vs high-GI diets. The weight difference was smaller than a day's water fluctuation, but the metabolic story had a twist nobody expected. L et al. · 2013 · Nutrition, Metabolism and Cardiovascular Diseases · n=2,344

The post-workout clock was never ticking Nineteen sports nutrition researchers reviewed decades of timing evidence. What they found about your post-workout window, your gummy bears, and your glycogen tank rewrites the routine. Kerksick et al. · 2017 · Journal of the International Society of Sports Nutrition

The Breakfast Advice Your Parents Got Was Planted by Two Food Companies Seven controlled experiments tested the most inherited piece of nutrition advice in the world. The group told to eat breakfast gained more weight than the group told to skip it. Bonnet et al. · 2020 · Obesity (Silver Spring) · n=425

Your 5:45 Alarm Bought Nothing but Lost Sleep Researchers matched 20 women on every variable — calories, protein, treadmill sessions — and changed only one thing: whether they ate before or after exercise. The gap in fat loss was statistically indistinguishable from zero. Schoenfeld et al. · 2014 · Journal of the International Society of Sports Nutrition · n=20

They Gained Muscle on 40% Fewer Calories Forty men ran a severe calorie deficit for four weeks. The group that doubled its protein gained lean mass and lost more fat. The group that ate half as much protein still held every gram of muscle. And strength? Identical in both. Longland et al. · 2016 · The American Journal of Clinical Nutrition · n=40

800 extra calories a day from protein. Zero fat gained. Researchers pushed protein intake to the highest level ever recorded in a published study and ran body scans every two weeks. The calorie math said 12.8 pounds of fat should have appeared. It didn't. Antonio et al. · 2014 · Journal of the International Society of Sports Nutrition · n=30

The 30g Limit Was a Measurement Limit Researchers gave 36 men 100 grams of protein and tracked every gram for 12 hours. The body was still building at the end. The old studies just stopped watching too early. Trommelen et al. · 2023 · Cell Reports Medicine · n=36

The studies behind every answer.