Science
Every study behind the method, including the ones that cut against it
In an industry where every landing page says “backed by science” and links to nothing, here’s the actual list. Every claim on this site carries a tag. verified means I read the paper or its abstract and the numbers are quoted from it. source pending means it’s widely reported but I haven’t pulled the primary paper yet, so treat it as provisional. If a sentence anywhere on this site says more than this page can back, that sentence is wrong.
Fat leaves as carbon dioxide, not heat
verified
When 10 kg of fat is metabolised, 8.4 kg leaves the body as exhaled CO₂ and 1.6 kg as water. The lungs are the primary excretory organ for fat. Most people, including many clinicians surveyed in the paper, assume fat is “burned” into heat or converted to muscle. It isn’t. It’s a mass balance.
Meerman R, Brown AJ. When somebody loses weight, where does the fat go? BMJ 2014;349:g7257. doi:10.1136/bmj.g7257
This is the one citation the whole site hangs on. A calorie measures the food. Mass measures you. Once you’ve actually read that paper you can’t look at a scale the same way.
The overnight drop is mostly water, and here’s exactly how much I can prove
source pending
You’ll read “80% of overnight weight loss is water” all over the internet, and I used to repeat it. Here’s what’s actually in the literature: an adult loses roughly 400 to 800 ml of water a day without noticing, about half through the skin and half through breathing. The one paper on the overnight share is a teaching exercise that reasons from those daily figures. Nobody has weighed people to the gram across a night and published it.
Weissenberg S. Insensible water loss during sleep: a theoretical exercise. Adv Physiol Educ 2005;29(4):213–215. doi:10.1152/advan.00028.2005. Daily insensible-loss range from standard physiology references; a measured overnight percentage does not exist, so treat any percentage, including mine, as an estimate.
What is measured is the two-week version, in the next section: 84% of short-term weight change is water and lean tissue. So the honest sentence is that your delta is mostly water and breath, and that the fat shows up in the trend of your maximum over weeks. That one constraint shapes every word on this site.
Your weight swings several pounds a day, and you are not imagining it
verified
Forty-six adults were tracked for two weeks with no diet and no instructions. Their weight moved by 0.26 ± 1.2 kg over the fortnight, and 84% of every unit of change was fat-free mass, which over two weeks is mostly water. Sixteen percent was fat. A 1 to 3 kg short-term swing carried an energy density of about 2,380 kcal/kg, a third of what pure fat would.
Bhutani S, Kahn E, Tasali E, Schoeller DA. Composition of two-week change in body weight under unrestricted free-living conditions. Physiol Rep 2017;5(13):e13336. doi:10.14814/phy2.13336
verified
In 80 adults logging 4,657 weigh-ins, weight peaked on Sunday and Monday and bottomed out on Friday and Saturday. Among people losing weight, the weekly maximum landed on Sunday or Monday 59% of the time. That is why the baseline is 14 days and not 7: it has to contain two weekends.
Orsama AL, Mattila E, Ermes M, van Gils M, Wansink B, Korhonen I. Weight Rhythms: Weight Increases during Weekends and Decreases during Weekdays. Obes Facts 2014;7(1):36–47. doi:10.1159/000356147
Part of the mechanism is glycogen: each gram is stored with at least 3 grams of water, so a low-carb day drops pounds of water and a refeed puts them back, with no fat involved either way (Murray B, Rosenbloom C. Nutr Rev 2018;76(4):243–259, doi:10.1093/nutrit/nuy001). Salt does the same thing: in the DASH-Sodium trial, 412 adults fed the same calories at three sodium levels weighed slightly more on the saltiest one (Juraschek SP et al. Hypertension 2020;75(3):723–729, doi:10.1161/HYPERTENSIONAHA.119.13932). And the menstrual cycle? You’ll see “2 to 5 lb of water retention” quoted everywhere. The one recent study that measured body composition at four points in the cycle found no significant effect in 54 people (Stein-Brüggemann D et al. Eur J Sport Sci 2025;25(5):e12283, doi:10.1002/ejsc.12283). Some people see a swing around their cycle. How big, and how consistently, is actually not settled, which is why the app gives you a marker instead of an assumption.
So when your Monday number feels random, it isn’t you. You’re being handed the last frame of a movie and asked what the plot was. That’s why the app draws your 7-day moving average in bold and the raw points faint: signal and noise, separated. It’s also why the app plans from your average delta and never from one high or low reading. An extreme reading is followed by a less extreme one for statistical reasons alone (regression to the mean), and a plan built on one peak is a plan built on a fluke.
The 2pm cutoff has been tested at exactly that clock time
verified
Eleven adults with overweight ate all their food between 8:00 am and 2:00 pm for four days, then, after a washout, the same food between 8:00 am and 8:00 pm. Same calories both times. On the early window, 24-hour mean glucose fell by 4 ± 1 mg/dl (p = 0.0003) and morning fasting insulin fell by 2.9 ± 0.4 mU/l (p < 0.0001).
Jamshed H, Beyl RA, Della Manna DL, Yang ES, Ravussin E, Peterson CM. Early Time-Restricted Feeding Improves 24-Hour Glucose Levels and Affects Markers of the Circadian Clock, Aging, and Autophagy in Humans. Nutrients 2019;11(6):1234. doi:10.3390/nu11061234
Eleven people for four days is a mechanism study, not a weight-loss trial, so don’t let anyone (including me) sell it as one. For weight, the trial you actually want is the next one.
verified
Ninety adults with obesity were randomised to an 8-hour early window (7:00 am to 3:00 pm) or a window of 12 hours or more, both with the same calorie reduction, for 14 weeks. The early-window group lost an additional 2.3 kg (95% CI −3.7 to −0.9; p = 0.002).
Jamshed H, Steger FL, Bryan DR, et al. Effectiveness of Early Time-Restricted Eating for Weight Loss, Fat Loss, and Cardiometabolic Health in Adults With Obesity: A Randomized Clinical Trial. JAMA Intern Med 2022;182(9):953–962. doi:10.1001/jamainternmed.2022.3050
That trial’s window ends at 3pm. Ours ends at 2pm. This is about as close to a direct evidence base as a consumer protocol ever gets, and I’ll be straight with you: the hour between 2 and 3 is mine, not the trial’s.
A longer fast does not speed up your metabolism
verified
Early time-restricted feeding (8:00 am to 2:00 pm) reduced appetite and increased fat oxidation but did not change energy expenditure compared with a 12-hour window. Your body burns the same total. It burns a higher share of it as fat.
Ravussin E, Beyl RA, Poggiogalle E, Hsia DS, Peterson CM. Early Time-Restricted Feeding Reduces Appetite and Increases Fat Oxidation but Does Not Affect Energy Expenditure in Humans. Obesity (Silver Spring) 2019;27(8):1244–1254. doi:10.1002/oby.22518
The title of that paper is a warning label and I treat it as one. You won’t read “boosts your metabolism” anywhere on this site. Well, the truth is your metabolism doesn’t speed up. What changes is the fuel: a longer fasting window shifts your body toward burning fat rather than the food you just ate.
Earlier beats later
verified
Delaying meal times reduces fat oxidation in a randomised crossover controlled-feeding design. Across the literature, eating windows of 8 to 10 hours placed early in the day look like the most sustainable balance of effect and adherence.
Randomised crossover controlled feeding study, PMC9691571. Full citation to be added.
verified
Here’s the same 8-hour window placed late. 116 adults with overweight or obesity were randomised to eat noon to 8pm or to three structured meals, for 12 weeks. The late-window group lost 0.94 kg, the control group 0.68 kg, and the difference between them was −0.26 kg (95% CI −1.30 to 0.78; p = .63). Nothing.
Lowe DA, Wu N, Rohdin-Bibby L, et al. Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity: The TREAT Randomized Clinical Trial. JAMA Intern Med 2020;180(11):1491–1499. doi:10.1001/jamainternmed.2020.4153
So here’s why 2pm and not the 16:8 you’ve seen everywhere. Eight hours ending at 3pm lost an extra 2.3 kg in 14 weeks. Eight hours ending at 8pm lost nothing extra in 12. Same window length, different intervention. Earlier wins, and the noon-to-8 version that most people actually try is the one with the null result.
The honest caveat: a lot of the benefit is eating less
verified citation pending
A systematic review and meta-analysis of 19 studies, 22 intervention arms, 1,201 participants found that the advantages of early over late time-restricted eating in waist circumference, fat mass, fasting insulin and HbA1c appeared only when energy intake was not prescribed. Once total calories were controlled, the differences disappeared. A separate crossover trial in 31 women with overweight or obesity found that intended-isocaloric TRE shifted circadian clocks but did not improve cardiometabolic health.
Numbers verified during research; the review’s full citation is being attached to this page.
What that actually means: a good chunk of the benefit is that you eat less when the window is shorter. That’s not a hole in the method, that’s the product. Eating less without counting is the entire pitch. But it does mean I can’t claim some metabolic magic that works independent of what you eat, so I don’t.
Two meals beat six at the same calories
verified
Fifty-four adults with type 2 diabetes ate either two larger meals (breakfast and lunch) or six smaller meals, at identical calories (500 below requirement), for 12 weeks each in a crossover. Two meals produced more weight loss (−3.7 kg vs −2.3 kg), more reduction in liver fat and fasting glucose, and a bigger rise in insulin sensitivity. In the follow-up hormone analysis, weight loss was −3.8 kg (95% CI −4.1, −3.4) vs −2.4 kg (95% CI −2.8, −2.0), p = 0.0005. Both meals were finished by 4pm.
Kahleová H, Belinova L, Malinska H, et al. Eating two larger meals a day (breakfast and lunch) is more effective than six smaller meals in a reduced-energy regimen for patients with type 2 diabetes: a randomised crossover study. Diabetologia 2014;57(8):1552–1560. doi:10.1007/s00125-014-3253-5
Belinova L, Kahleova H, Malinska H, et al. The effect of meal frequency in a reduced-energy regimen on the gastrointestinal and appetite hormones in patients with type 2 diabetes: a randomised crossover study. PLoS One 2017;12(4):e0174820. doi:10.1371/journal.pone.0174820
verified
Zoom out to every trial. A 2024 meta-analysis of 29 randomised trials, 2,485 adults, running 12 weeks or longer, found that eating fewer meals moved weight by −1.85 kg (95% CI −3.55 to −0.13), eating earlier in the day by −1.75 kg (−2.37 to −1.13), and time-restricted eating by −1.37 kg (−1.99 to −0.75), each against standard advice. The authors call the effects “small and of uncertain clinical importance” and flag high risk of bias. A 2023 meta-analysis that split trials at three meals a day found no difference in weight (−0.62 kg, 95% CI −2.76 to 1.52; p = 0.57), on very low certainty evidence.
Liu HY, Eso AA, Cook N, O’Neill HM, Albarqouni L. Meal Timing and Anthropometric and Metabolic Outcomes: A Systematic Review and Meta-Analysis. JAMA Netw Open 2024;7(11):e2442163. doi:10.1001/jamanetworkopen.2024.42163
Blazey P, Habibi A, Hassen N, Friedman D, Khan KM, Ardern CL. The effects of eating frequency on changes in body composition and cardiometabolic health in adults: a systematic review with meta-analysis of randomized trials. Int J Behav Nutr Phys Act 2023;20(1):133. doi:10.1186/s12966-023-01532-z
Read those two together and you get the truthful version: fewer meals and earlier meals each nudge the scale by a kilo or two over three months, the trials are messy, and nobody should promise you more than that. I don’t. The method’s engine is the ceiling, not the meal count.
Three limits on the Kahleová trial, plainly. Those subjects had type 2 diabetes and were on a prescribed deficit. The comparison was two meals versus six, not two versus three. And fasting ghrelin, the hunger hormone, went up on two meals. Hunger is a real cost of eating this way, and part of my case for a 2pm cutoff over a late window is that 9pm hunger is easier to sleep through than to snack through.
verified
In a cohort of 8,874 Chinese adults over 45 followed for four years, eating two meals a day showed no significant association with developing type 2 diabetes compared with three. Four meals a day was associated with lower risk (RR 0.76, 95% CI 0.60–0.97), a signal that held only in people with BMI under 25.
Wang X, Hu Y, Qin LQ, Dong JY. Meal frequency and incidence of type 2 diabetes: a prospective study. Br J Nutr 2022;128(2):273–278. doi:10.1017/S0007114521003226
That study kills “two meals is dangerous.” It doesn’t say two meals is better, and I won’t pretend it does. Two meals is how an early window works in practice, not a mechanism of its own.
Two 20-minute sessions produce more afterburn than one 40-minute session
verified
The combined excess post-exercise oxygen consumption from split sessions (7,410 ± 1,851 mL O₂) was significantly greater than from one continuous session (5,278 ± 1,305 mL O₂). The authors concluded that splitting exercise can significantly increase post-exercise caloric expenditure.
Almuzaini KS, Potteiger JA, Green SB. Effects of split exercise sessions on excess postexercise oxygen consumption and resting metabolic rate. Can J Appl Physiol 1998;23(5):433–443. PMID 9738129
verified
“The absence of a sustained EPOC after exercise seems to be a consistent finding in studies with low exercise intensity and/or duration.” The size of the afterburn depends on both how long and how hard.
Børsheim E, Bahr R. Effect of exercise intensity, duration and mode on post-exercise oxygen consumption. Sports Med 2003;33(14):1037–1060. PMID 14599232
Two things follow. Your 20 minutes have to actually raise your heart rate, or the citation doesn’t apply to what you did. And the afterburn is real but small: tens of calories, not hundreds. It helps. It isn’t the engine. The engine is the ceiling.
One to two pounds a week is the range, and the app caps at the top of it
verified
“People who lose weight at a gradual, steady pace—about 1 to 2 pounds a week—are more likely to keep the weight off than people who lose weight quicker.”
CDC, Steps for Losing Weight, last reviewed 17 January 2025.
The app defaults to 1 lb a week. You can adjust it. You can’t set it above 2. My own 40 lbs in 52 weeks came to 0.77 lb a week, inside the range and close to the default, which is not a coincidence.
Weighing every day works, and two readings extend that rather than contradict it
verified
In a six-month trial, adults who weighed themselves every day lost 9.2 kg; those who weighed less often lost 3.1 kg (difference −6.1 kg, 95% CI −10.2 to −2.1, p = .004). The less-often group still weighed 5.4 days a week. Daily weighers also adopted more weight-control behaviours.
Steinberg DM, Bennett GG, Askew S, Tate DF. Weighing everyday matters: Daily weighing improves weight loss and adoption of weight control behaviors. J Acad Nutr Diet 2015;115(4):511–518. doi:10.1016/j.jand.2014.12.011
My position is that one morning number is incomplete, not wrong. Two numbers give you a pattern. Nobody has tested twice-a-day against once-a-day and I’m not going to claim it wins. What the app does is teach you to tell fluctuation from change, which is the exact reason the trial’s authors gave for why daily weighing helps.
The “91% higher risk” headline, and what it measured
In March 2024 the American Heart Association issued a press release on a conference abstract linking eating windows under 8 hours to a 91% higher risk of cardiovascular death. It was widely reported. Here is what it was: an abstract, not a peer-reviewed paper; NHANES survey data from around 20,000 adults; the eating window classified from two days of self-reported recall; and, in the release’s own words, “this does not mean that time-restricted eating caused cardiovascular death.” Factors outside eating duration were not in the model.
It’s on this page because leaving it off would be the one thing that makes it look true. It’s an observation worth following as the full paper lands. It isn’t evidence that an early window harms you, and it isn’t evidence that it doesn’t.
What you won’t read on this site
| Not on this site | What we say instead |
|---|---|
| “Burn fat overnight” | “Your overnight mass change” |
| “Boosts your metabolism” | “Shifts your body toward burning fat” |
| “Lose X lbs in Y weeks” | “1 lb a week is the default target” |
| “Calorie counting doesn’t work” | “Calorie counting asks you to estimate a number that was never about you” |
| “Three meals a day is unhealthy” | “Three meals a day is a habit, not a law” |
| “Typical users lose 40 lbs” | “I lost 40 lbs. Your body is not my body.” |
| Any disease treatment or prevention claim | Nothing. Not one. |
Still being sourced
A measured (not theoretical) overnight water figure, if one ever gets published. The full citation for the 19-study early-versus-late review. A clean two-versus-three-meal trial in adults without diabetes. When they land, the tags above change. Know one I’ve missed? Tell me.