108 corrections on 13 pages. Most arrive together: one paper read again against every page that cites it, and every page fixed the same day.
Rosqvist et al. 201423 corrections · 4 pages
November 22, 2026 · We got it wrong
A Short called lean tissue the trial's primary outcome, and liver fat a secondary one
What it said
Its notes for researchers put the lean tissue result under "Primary outcome:" and the visceral and liver fat results under "Secondary outcomes:".
What it says now
They say "Lean tissue (a secondary outcome):" and "Liver fat (the primary outcome) and visceral fat:".
Why
Rosqvist and colleagues (2014) name liver fat as the trial's primary outcome; lean tissue was a secondary outcome, and the authors say the lean tissue finding needs to be replicated.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short set sunflower oil against the oils its reader kept, when the trial compared it with palm oil only
What it said
It said "A randomized trial put that exact hierarchy to the test", that the seed oil did better "than the oil you kept", and its search title said the best oil "Wasn't the One You Kept".
What it says now
It says "A trial in Sweden put two oils to the test: sunflower oil and palm oil. Coconut oil and avocado oil were not part of it."
Why
The trial by Rosqvist and colleagues (2014) baked refined sunflower oil or refined palm oil into muffins; coconut, avocado and olive oil were not tested, and the authors see no clear reason to think sunflower oil would differ from other oils rich in polyunsaturated fat. The Short told the trial as beating the oils its reader kept.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a recipe as the trial's exact substitution, and itself as explaining a mechanism no paper found
What it said
Its recipe notes said a turkey burger with olive oil and avocado makes "the exact fat-source substitution the Rosqvist study tested" and that "This Short explains the mechanism behind why fat type matters for body composition".
What it says now
The Short no longer links the turkey burger, and its salad note now says "The 2014 trial's authors say their results back the advice to swap some saturated fat for polyunsaturated fat".
Why
The trial by Rosqvist and colleagues (2014) compared refined sunflower oil with refined palm oil, baked into muffins; it did not test olive oil or avocado, so nothing in it ties the burger's fats to the trial. The authors write that the mechanisms behind the differences are unknown and remain to be determined; they also write that their results support the advice to partly replace saturated fat with polyunsaturated fat, found in nuts, seeds and non-tropical vegetable oils.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a trial's gain in lean tissue as muscle built, and a nearly threefold rise as three times
What it said
Its title said "The Oil You Threw Away Built Three Times More Muscle", and it said the sunflower oil "built more muscle per calorie of surplus" and that the surplus "split roughly half and half between muscle and fat".
What it says now
Its title asks "Does Your Cooking Oil Change How Much Muscle You Build?", and it says "Lean tissue is not just muscle: it also counts organs and other tissue that is not fat."
Why
Rosqvist and colleagues (2014) measured lean tissue, by MRI and by air-displacement plethysmography, not muscle on its own; lean tissue counts muscle but also organs and other tissue that is not fat. With sunflower oil it rose nearly three times as much as with palm oil. The Short told the result as muscle built, and as three times more.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said fat type in a surplus determines more than weight gain, when the trial found similar weight gain with both fats
What it said
Its answer for AI systems and its spoken summary opened "Fat type during a caloric surplus determines tissue partitioning, not just weight gain".
What it says now
They open with the trial and its people, "37 young, lean adults ate extra muffins made with palm oil or sunflower oil", and say "Both groups gained similar weight."
Why
In the trial by Rosqvist and colleagues (2014) both groups gained similar weight, 1.6 kg on average; the type of fat changed what the gain was made of, not how much was gained. Read as written, the old opening set weight gain among what fat type determines.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said visceral fat doubled, when its rise was twice the other group's
What it said
It said "Visceral fat around the organs doubled", its takeaway that one surplus "built equal parts lean tissue and fat", and its notes said "~2× more visceral adipose tissue" and "~3× more lean tissue".
What it says now
It says "Its visceral fat, the fat around the organs, rose twice as much" and "nearly three times more lean tissue".
Why
In the trial by Rosqvist and colleagues (2014), visceral fat rose by 0.22 litres with palm oil and 0.11 litres with sunflower oil, from about 0.8 litres at the start: a twofold larger rise, not a doubling of visceral fat. Lean tissue rose nearly three times as much with sunflower oil, and the ratio of lean tissue to fat was about one to one, not equal parts.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short's search title gave one group's lean-to-fat ratio as what fat calories in a surplus build
What it said
Its search title said "Fat Calories During a Surplus — MRI Found a 1:4 Tissue Split".
What it says now
It reads "Fat Calories in a Surplus: Sunflower vs Palm Oil Under MRI".
Why
In the trial by Rosqvist and colleagues (2014) the gain was about one part lean tissue to four parts fat with palm oil, and about one to one with sunflower oil. Alone in a search result, the title gave the palm oil group's ratio as what fat calories in a surplus build.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told the type of fat as the only difference between two trial groups
What it said
It said the fat type directed the calories "while every other variable held steady", and its recording said "The only variable that separated them was fat type".
What it says now
It says "The muffins were the same except for the oil", and its notes for researchers that the oils "also differed in monounsaturated fat".
Why
In the trial by Rosqvist and colleagues (2014) the muffins were identical except for the oil, but the two oils also differed in monounsaturated fat, 37% of palm oil's fatty acids against 24% of sunflower oil's, and in vitamin E, which the authors name as a limitation though they judge its levels most likely too low to have had an effect; the saturated fat group's intake of monounsaturated fat also rose more.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said liver fat depends on the type of fat more than the amount, which neither trial tested
What it said
It said "Liver fat accumulation is primarily determined by the type of dietary fat, not the total amount" and "The variable that determined liver fat was not the total amount of dietary fat."
What it says now
It says "Both trials compared two types of fat at a similar surplus. Neither tested eating more fat against less."
Why
Both trials by Rosqvist and colleagues (2014, 2019) gave muffins that differed only in the type of fat; neither compared amounts of fat. The 2014 paper notes that other studies found high-fat diets raise liver fat compared with low-fat diets.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said saturated fat produced fourteen times more liver fat, dividing two changes, one of them near zero
What it said
It said "saturated fat produced fourteen times more liver fat than polyunsaturated fat" and "The saturated fat group accumulated fourteen times more liver fat", and its chart showed a 14× callout.
What it says now
It says "In the saturated fat group, liver fat rose by about half a percentage point, from a low start of about 1%" and "The polyunsaturated group's liver barely moved."
Why
In the trial by Rosqvist and colleagues (2014), liver fat rose by 0.56 percentage points with palm oil and 0.04 with sunflower oil, from about 1% at the start. Dividing one change by the other gives fourteen, but the paper gives no such ratio: it says saturated fat markedly increased liver fat compared with polyunsaturated fat. Neither group ended with fourteen times the other's liver fat.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told the liver as reading and sorting fat by type, a mechanism neither trial showed
What it said
It said the liver "was reading what arrived and responding accordingly", that "It sorts by type", and that "the liver receives a signal to store fat locally".
What it says now
It says "That is a link, not proof of a cause, and the authors say the reasons for the difference are not known."
Why
Rosqvist and colleagues (2014) found that blood levels of the two fats tracked the liver fat changes, an association, and wrote that the mechanisms behind the difference are unknown. In their 2019 trial a small exploratory test found no difference in how much palmitate the liver took up, and the authors say the mechanisms remain elusive.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Three Shorts cited the Rosqvist papers with a composed title, the wrong journal or DOI, or the journal as publisher
What it said
The cooking oil Short cited the 2014 paper as "Overfeeding polyunsaturated and saturated fat in humans: effects on body composition and gene expression" and its takeaway as "Rosqvist et al. 2014 · American Journal of Clinical Nutrition"; the liver fat Short printed the 2019 paper as "Rosqvist et al. (2019) · American Journal of Clinical Nutrition" with the 2014 paper's DOI, and a note, "Satellite data from same research group", where its DOI goes. Two of the Shorts' data blocks gave the journal's name as the 2014 paper's publisher, and the liver fat Short's named only the 2014 paper beside the 2019 trial's figures.
What it says now
They cite the 2014 paper as "Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans", in Diabetes, and the 2019 paper as "Rosqvist et al. (2019) · Journal of Clinical Endocrinology and Metabolism" with its own DOI, 10.1210/jc.2019-00160; the data blocks name the American Diabetes Association as the publisher, and each Short now cites the later papers it reports.
Why
The 2014 paper by Rosqvist and colleagues is titled "Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans" and was published in Diabetes (doi:10.2337/db13-1622) by the American Diabetes Association. The 2019 paper appeared in the Journal of Clinical Endocrinology and Metabolism (doi:10.1210/jc.2019-00160), and the 2024 paper, which one Short's takeaway rested on without naming it, in the American Journal of Clinical Nutrition.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Three Shorts told the similar weight gain of two trial groups as identical
What it said
The cooking oil Short said "despite identical weight gain on the scale" and "Identical weight."; the fat calories Short said "1.6 kilograms on the scale, identical"; the liver fat Short said "Both groups gained exactly 1.6 kilograms" and "The scale saw identical outcomes".
What it says now
The cooking oil Short says "Weight gain was similar"; the fat calories Short says "1.6 kilograms on the scale, on average"; the liver fat Short says "Both groups gained 1.6 kilograms on average."
Why
In the 2014 trial by Rosqvist and colleagues both groups gained 1.6 kg on average, 0.02 kg apart; the paper calls the weight gain similar, and the gains varied from person to person. The Shorts called it identical.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told a surplus of muffins as extra fat alone, and weeks of it as every time you eat fat
What it said
The liver fat Short said "This was overfeeding — extra fat added on top of a normal diet" and "Every time you eat fat, the organ doesn't just store it."; the fat calories Short said "one group got their surplus from saturated fat, the other from polyunsaturated fat".
What it says now
The liver fat Short says "The muffins were about half fat and nearly half carbohydrate." and "about three extra muffins a day on top of a normal diet, for seven or eight weeks"; the fat calories Short says "Two groups ate extra muffins for seven weeks, on top of their usual food."
Why
In both trials by Rosqvist and colleagues the extra calories came from muffins that drew 51% of their energy from fat and 44% from carbohydrate, eaten for seven or eight weeks; in the 2014 trial more than half of that carbohydrate was sugar, and its authors say fructose and saturated fat may have acted together on liver fat. The Shorts told the surplus as fat alone, and one of them the result as what happens each time you eat fat.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told two trial groups as building opposite bodies, when both gained fat and lean tissue
What it said
The cooking oil Short said "The oil on the pan decided where the surplus landed" and "Completely different body composition."; the fat calories Short said "The Surplus Built Two Different Bodies on the Same Scale", "Opposite architecture." and that saturated fat "concentrated that fat in the worst places".
What it says now
The cooking oil Short says "The oil in the muffins changed the mix of what each group gained"; the fat calories Short says "Both groups gained some of each."
Why
In the trial by Rosqvist and colleagues (2014) both groups gained fat and lean tissue: about one part lean to one part fat with sunflower oil, and one to four with palm oil. Palm oil raised visceral fat twice as much, yet most of each group's new fat was not visceral fat. The Shorts told the groups as building opposite bodies.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told an overfeeding trial in people who were not training as an answer for muscle building or as a bulk
What it said
The cooking oil Short closed "Your cooking oil changes what a surplus builds inside your body, if your body is lean enough for the partition to matter"; the fat calories Short said the study "pulled back the curtain on what that budget actually built", and its recording opened "You're seven weeks into a bulk", told the listener "Now imagine someone slides you into an MRI machine" and said "Two groups did exactly this."
What it says now
The cooking oil Short says "People who did regular hard exercise for more than three hours a week were left out of both trials, and everyone kept their usual activity"; the fat calories Short says "They kept their usual activity, and anyone who did regular hard exercise for more than three hours a week was left out."
Why
In the 2014 trial by Rosqvist and colleagues, and in the same team's repeat reported in 2019 and 2024, people who did regular heavy exercise for more than three hours a week were excluded, and everyone kept their usual diet and activity while muffins were added to it. Neither trial tested training, and the 2024 authors call the question of resistance exercise unclear. The Shorts told the trial as an answer for building muscle and as a bulk.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told one trial's lean tissue result as settled, and fat type as deciding what a surplus builds
What it said
The cooking oil Short said "The partitioning advantage held in lean adults" and graded it Strong; the fat calories Short said "During a caloric surplus, the type of fat you eat changes what your body builds with those extra calories" and "The fat source column in your tracker was never cosmetic."
What it says now
The cooking oil Short says "When the same team repeated the trial in older, heavier adults, lean tissue rose by a similar amount with both oils" and that the first result "may have been chance"; the fat calories Short gives the authors' own words, that fat type "seems to be a new and important factor" and that "the lean tissue result needs to be repeated in other studies."
Why
In the trial by Rosqvist and colleagues (2014) lean tissue was a secondary outcome, and the authors wrote that the finding needs to be replicated; they said fat type seems to be an important determinant during moderate weight gain. The same team's repeat in older adults with overweight did not find the lean tissue difference (2024). The Shorts told the result as settled, and as a rule for the reader's own tracker.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told the trial's enrolment, its surplus and its people beyond what the paper says
What it said
Their notes for researchers said "39 enrolled" and a set surplus of 750 kcal a day ("Participants received 750 kcal/day surplus" on the fat calories Short), and the fat calories Short called the volunteers Swedish.
What it says now
They say "41 randomized, 39 started" and that the muffins were "adjusted to reach a 3% weight gain (on average 3.1 a day, about 750 kcal)"; the fat calories Short says the trial was "studied in Uppsala, Sweden".
Why
In the trial by Rosqvist and colleagues (2014), 41 people were randomized and 2 dropped out before it began, leaving 39 with baseline data and 37 analysed; the muffins were added to the usual diet and adjusted person by person to reach a 3% weight gain, on average 3.1 a day or about 750 kcal. The trial ran in Uppsala, Sweden; the paper gives no nationality for its people.
Rosqvist F, Iggman D, Kullberg J, Cedernaes J, Johansson HE, Larsson A, Johansson L, Ahlström H, Arner P, Dahlman I, Risérus U. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said the type of fat determines where it lands and called the biology real and established; the studies' authors said their results seem to, or suggest
What it said
The page said "the type of fat determines where it lands" and "The underlying biology is real — glycogen buffers exist, fat type matters". Its notes for AI systems called the carbohydrate finding an "established mechanism".
What it says now
It says "The authors say the type of fat seems to shape where a modest weight gain goes."
Why
Rosqvist and colleagues (2014) say the type of fat "seems to be" an important factor in where weight gain goes, and that their lean tissue result needed to be repeated in more studies. Acheson and colleagues (1988) and Levine and colleagues (1999) say their data "suggest" their findings. Three men, 39 young adults and 16 volunteers do not settle the biology.
Rosqvist F, Iggman D, Kullberg J, et al. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368. doi:10.2337/db13-1622.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said saturated fat caused twice the visceral fat and polyunsaturated fat three times the lean tissue; the trial reports a twofold larger rise and a nearly threefold larger rise
What it said
Its notes for AI systems said "SFA increased liver fat and caused 2x more visceral fat, while PUFA caused 3x more lean tissue growth".
What it says now
Its notes for AI systems now say "caused a twofold larger increase in visceral fat, while PUFA caused a nearly threefold larger increase in lean tissue".
Why
Rosqvist and colleagues (2014) report "a twofold larger increase in VAT" and "a nearly threefold larger increase in lean tissue": the rises, compared, and nearly three times, not three.
Rosqvist F, Iggman D, Kullberg J, et al. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368. doi:10.2337/db13-1622.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said fat from food reaches your fat cells within hours; none of its studies timed that
What it said
The page said "Dietary fat reaches your fat cells within hours." and "The fat clock is shorter — and stranger." Its answer for AI systems said "Dietary fat reaches storage faster".
What it says now
It says "For fat, the trial here asked a different question: not how fast, but where the gain goes."
Why
None of the page's studies timed dietary fat. Acheson and colleagues (1988) fed carbohydrate. Rosqvist and colleagues (2014) measured body fat and lean tissue after seven weeks, and Levine and colleagues (1999) after eight weeks.
Rosqvist F, Iggman D, Kullberg J, et al. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368. doi:10.2337/db13-1622.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said both fat types gave identical weight gain; the trial reports similar weight gain
What it said
Its section for researchers said "Weight gain identical; body composition diverged by fat type." and its notes for AI systems said "despite identical weight gain".
What it says now
Its section for researchers says "Weight gain similar" and gives the difference.
Why
Rosqvist and colleagues (2014) report that both groups gained similar weight, 1.6 kg on average, a difference of 0.02 kg between them.
Rosqvist F, Iggman D, Kullberg J, et al. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368. doi:10.2337/db13-1622.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said extra saturated fat went to the liver and organs while polyunsaturated fat built lean tissue instead; both groups gained fat and lean tissue, in different proportions
What it said
The page said "two completely different bodies. Saturated fat drove storage into the liver and around the organs. Polyunsaturated fat built lean tissue instead." Its FAQ said "the body underneath was completely different".
What it says now
It says "Both groups gained about the same, 1.6 kg on average, and both gained some fat and some lean tissue." and "But the palm oil group gained more body fat, more fat in the liver and around the organs, and less lean tissue."
Why
In the trial by Rosqvist and colleagues (2014), young adults ate extra muffins made with palm oil or sunflower oil for seven weeks. Both groups gained 1.6 kg, and both gained body fat and lean tissue. The palm oil group gained more total, visceral and liver fat and less lean tissue: lean to fat about one to four, against about one to one with sunflower oil.
Rosqvist F, Iggman D, Kullberg J, et al. Overfeeding Polyunsaturated and Saturated Fat Causes Distinct Effects on Liver and Visceral Fat Accumulation in Humans. Diabetes. 2014;63(7):2356-2368. doi:10.2337/db13-1622.Read the paper
Found by
Our full re-read of the page against all its papers.
A Short said a sports nutrition position stand cites the trial's finding, with no copy of the stand to check it against
November 22, 2026 · We got it wrong
What it said
Its notes for researchers said "ISSN position stand on diets and body composition (2017) cites the ~1:1 vs ~1:4 lean:fat ratio finding", and its grade of the evidence said "ISSN position stand cites this finding".
What it says now
Neither line is on the page: FitChef holds no copy of the position stand to check the claim against.
Why
FitChef holds no copy of the International Society of Sports Nutrition position stand on diets and body composition (2017), so the claim that it cites this trial's lean-to-fat ratios cannot be checked. No finding on the page rests on it, and the lines are removed.
Our full re-read of the page against all its papers.
Rosqvist et al. 20193 corrections · 2 pages
November 22, 2026 · We got it wrong
A Short said no symptom shows until liver damage is advanced and that you cannot check your liver, with no source
What it said
It said "No symptom fires until the damage is advanced", "you have no way to check what's happening" and "Silently, without symptoms, inside an organ you cannot monitor."
What it says now
The lines are gone; the page opens "The bathroom scale tells you what you weigh. It can't tell you how much fat sits inside your liver, the organ behind your ribs."
Why
Neither trial on the page says anything about symptoms or liver damage, and liver fat can be measured: both trials measured it by MRI, and the 2019 trial also measured a liver enzyme in the blood. The lines were FitChef's own and are removed.
Rosqvist F, Kullberg J, Ståhlman M, Cedernaes J, Heurling K, Johansson HE, Iggman D, Wilking H, Larsson A, Eriksson O, Johansson L, Straniero S, Rudling M, Antoni G, Lubberink M, Orho-Melander M, Borén J, Ahlström H, Risérus U. Overeating Saturated Fat Promotes Fatty Liver and Ceramides Compared With Polyunsaturated Fat: A Randomized Trial. J Clin Endocrinol Metab. 2019;104(12):6207-6219.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts gave one Rosqvist trial's figures to the other
What it said
The cooking oil Short gave the 2014 trial in lean adults "liver fat up 50%" and "increased liver fat by 50%"; the liver fat Short gave the 2019 trial in adults with overweight "8 weeks, +750 kcal/day" and named its data block "Liver Fat Divergence by Dietary Fat Type at Identical Calorie Surplus".
What it says now
The cooking oil Short says liver fat "rose by about half a percentage point" in the 2014 trial; the liver fat Short gives the 2019 trial "about 2.9 muffins (about 40 g of oil) a day".
Why
In the 2014 trial in young, lean adults (Rosqvist and colleagues, Diabetes), liver fat rose by 0.56 percentage points with palm oil, and the paper gives no relative figure; the 50% and 53% rises are the 2019 trial's, in adults with overweight. That 2019 trial gives no calorie figure for its muffins, about 2.9 a day and about 40 g of oil; 750 kcal a day was the 2014 trial's, and in 2019 energy intake rose by different amounts in the two groups.
Rosqvist F, Kullberg J, Ståhlman M, Cedernaes J, Heurling K, Johansson HE, Iggman D, Wilking H, Larsson A, Eriksson O, Johansson L, Straniero S, Rudling M, Antoni G, Lubberink M, Orho-Melander M, Borén J, Ahlström H, Risérus U. Overeating Saturated Fat Promotes Fatty Liver and Ceramides Compared With Polyunsaturated Fat: A Randomized Trial. J Clin Endocrinol Metab. 2019;104(12):6207-6219.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts told sunflower oil as lowering liver fat, when the trial found it did not change it
What it said
The cooking oil Short said sunflower oil "slightly decreased it"; the liver fat Short said "the liver was actively clearing fat when the right type arrived", and its chart said "Liver actively cleared fat".
What it says now
The cooking oil Short says "while sunflower oil did not raise it"; the liver fat Short says "It dipped by 2%, which the authors describe as no change, or at most a slight decrease."
Why
In the 2019 trial in adults with overweight (Rosqvist and colleagues), liver fat changed by −0.09 percentage points, a 2% decrease, with sunflower oil; the authors write that overfeeding polyunsaturated fat did not alter liver fat and that, if anything, there was a slight decrease. The Shorts told it as a fall, and the liver as clearing fat.
Rosqvist F, Kullberg J, Ståhlman M, Cedernaes J, Heurling K, Johansson HE, Iggman D, Wilking H, Larsson A, Eriksson O, Johansson L, Straniero S, Rudling M, Antoni G, Lubberink M, Orho-Melander M, Borén J, Ahlström H, Risérus U. Overeating Saturated Fat Promotes Fatty Liver and Ceramides Compared With Polyunsaturated Fat: A Randomized Trial. J Clin Endocrinol Metab. 2019;104(12):6207-6219.Read the paper
Found by
Our full re-read of the page against all its papers.
Rosqvist et al. 20242 corrections · 4 pages
November 22, 2026 · We got it wrong
Three Shorts misreported what the repeat trial in adults with overweight found
What it said
The cooking oil Short said "Carry more body fat, and the oil type stops separating what the surplus builds" and "the internal fat routing difference persists"; the fat calories Short said "Starting body composition likely influences how fat type routes surplus calories"; the liver fat Short said, after the trial in adults with overweight, "all sorted differently depending on which fat carried the surplus", and asked "Does being overweight make dietary fat worse for your liver?".
What it says now
The cooking oil Short says "Lean tissue rose by a similar amount with both oils, measured two different ways, and so did visceral fat and total body fat" and "The authors say their first lean tissue result may have been chance"; the fat calories Short says "In the repeat with heavier adults, only liver fat differed"; the liver fat Short says "In the second trial, none of these differed. Only liver fat did." and that the two trials "do not show that body weight itself made the difference".
Why
When the same team repeated the trial in older adults with overweight (reported in 2019 and 2024), lean tissue, visceral fat and total body fat rose by similar amounts with both fats; only liver fat still rose more with saturated fat. The 2024 authors say the first lean tissue result may have been chance, or partly due to the different people, who were older, heavier and more often women. The Shorts told the lost lean tissue difference as a matter of body fat, the other differences as holding in both trials, or the two separate trials as showing what body weight does to the liver.
Rosqvist F, Cedernaes J, Martínez Mora A, Fridén M, Johansson HE, Iggman D, Larsson A, Ahlström H, Kullberg J, Risérus U. Overfeeding polyunsaturated fat compared with saturated fat does not differentially influence lean tissue accumulation in individuals with overweight: a randomized controlled trial. Am J Clin Nutr. 2024;120(1):121-128.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories told a fat-type trial in young adults of normal weight as the answer and left out the same team's repeat, in which older, heavier adults gained similar body fat and lean tissue on both fats
What it said
To the question "Does the type of fat you eat affect where surplus calories go?" its FAQ answered "Yes — dramatically." from the trial in young adults of normal weight alone; the page named no later trial.
What it says now
It says "The same team then repeated the trial for eight weeks in 61 older, heavier adults." and "But the gains in body fat, fat around the organs and lean tissue were similar with both fats." Its FAQ now answers "In one trial it did; in a repeat, only for liver fat."
Why
The trial by Rosqvist and colleagues (2014) was in young adults of normal weight. The same team repeated it for eight weeks in 61 adults with overweight or obesity, older and heavier (Rosqvist and colleagues, 2019 and 2024). Palm oil again raised liver fat, by 53%, while sunflower oil did not; but visceral fat, total body fat and lean tissue rose by similar amounts with both oils. The authors say the first lean tissue result may have been chance, or partly due to the different people.
Rosqvist F, Cedernaes J, Martínez Mora A, Fridén M, Johansson HE, Iggman D, Larsson A, Ahlström H, Kullberg J, Risérus U. Overfeeding polyunsaturated fat compared with saturated fat does not differentially influence lean tissue accumulation in individuals with overweight: a randomized controlled trial. Am J Clin Nutr. 2024;120(1):121-128. doi:10.1016/j.ajcnut.2024.04.010.Read the paper
Found by
Our full re-read of the page against all its papers.
Two Shorts made claims about what fitness culture and everyone does, with no source
November 22, 2026 · We got it wrong
What it said
The cooking oil Short said "The oil that fitness culture ranks last" and "The oil everyone in the fitness world tells you to avoid"; the fat calories Short said "The one line in the macro tracker that nobody optimizes is fat source".
What it says now
They tell the reader's own scene instead: "In one trial, sunflower oil, the seed oil you took off the shelf, came out ahead of palm oil on lean tissue" and "The fat line holds whatever was in the kitchen."
Why
No source on either page says what fitness culture ranks, what everyone in the fitness world advises or what nobody optimizes; the lines were FitChef's own claims.
Our full re-read of the page against all its papers.
Levine et al. 199912 corrections · 3 pages
November 22, 2026 · We got it wrong
A 10,000-steps Short answered how much everyday burn varies between people with how much NEAT rose when people overate
What it said
Its FAQ asked "What controls the everyday calorie burn that varies between people?" and answered with how much NEAT rose while people overate, up to 692 calories a day.
What it says now
It asks "What is NEAT, and what happened to it when people overate?"
Why
Levine and colleagues (1999) report how NEAT changed while 16 adults ate 1,000 extra calories a day for 8 weeks: from a fall of 98 to a rise of 692 calories a day. That is a response to overeating, not how much people's everyday burn differs, and the answer gave it as one.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A 10,000-steps Short printed its own walking estimate under two studies' names, left a review off its takeaway's credit, and said every calculator agrees with it
What it said
Its key takeaway began "Walking 10,000 steps burns roughly 350 to 500 calories depending on body weight — a sound estimate of energy spent during the walk" above the credit to Levine 1999 and Pontzer 2021, and the page said "Every calculator on the internet agrees on a number in that range, and they are all approximately right." Its takeaway's line "The body can compensate, for example through appetite" stood under the same credit.
What it says now
It says "By FitChef's estimate, walking 10,000 steps burns roughly 350 to 500 calories, depending on body weight" and "That range is FitChef's own estimate.", and its takeaway's credit now also names "Westerterp 2013 · Frontiers in Physiology (review)".
Why
The 350-to-500 figure is FitChef's arithmetic on the oxygen volunteers used during a 10-minute treadmill walk in Levine and colleagues' 1999 study, which did not count steps. The page said so in its notes for researchers, but its takeaway printed the figure under the two studies' names. No source on the page compares online calculators. The takeaway's line that the body can compensate, for example through appetite, rests on Westerterp's 2013 review, which its credit did not name.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short's notes for AI systems graded one small study's results as high certainty, and named the wrong method for one of them
What it said
Its notes for AI systems graded its three results from the 1999 study as high certainty, one of them on the basis of "same controlled metabolic study with DLW methodology".
What it says now
They grade them moderate, as results of one study of 16 adults, and name indirect calorimetry as the way resting metabolism was measured.
Why
The three results come from one supervised study of 16 adults without obesity. In it, resting metabolism was measured by indirect calorimetry; doubly labeled water measured total energy use.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short's opening told the reader they have NEAT's role in their daily burn wrong; its study measured how NEAT changed when people overate
What it said
Its opening said NEAT's "role in your daily calorie burn is the part most people have radically wrong".
What it says now
It asks "When people eat more than they need, which bucket’s burn goes up the most?"
Why
Levine and colleagues (1999) report how the parts of daily energy use changed when 16 adults overate for 8 weeks: on average, the rise in movement outside exercise was two-thirds of the rise in energy use. They do not report its share of anyone's daily burn. Of the daily burn itself, resting metabolism is the largest part: 1,693 of 2,807 calories a day in the same volunteers before overfeeding. Westerterp's 2013 review also calls resting metabolism the largest part of daily energy use, and activity the part that varies most.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short gave the average rise in NEAT as 336 calories a day beside its table's spread; the table's own average is 328
What it said
Its section for researchers said "mean 336 ± 256 SD".
What it says now
It says "mean 328 ± 256 SD in Table 2; the text gives the average rise as 336".
Why
Levine and colleagues (1999) give the average rise in NEAT as 336 calories a day in their text, and as 328, with a standard deviation of 256, in their Table 2. The page put the text's average beside the table's spread.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short's recording told how resting metabolism changed during overeating as what a fast or slow metabolism decides; the study did not test that
What it said
Its recording said a metabolism "decides less than you think" and closed "So the next time you hear someone say their metabolism is slow — they might be partly right, just about the wrong part of it."
What it says now
Its recording says "Here’s what one study measured" and closes "In the overfeeding study, what tracked who gained less fat wasn’t resting metabolism."
Why
In Levine and colleagues' 1999 study, resting metabolism rose by about 5% when 16 adults overate for 8 weeks. That rise took up 8% of the extra energy, and how much it changed did not account for who gained fat. The study did not test whether a faster or slower metabolism to begin with decides weight gain, as a correction of 12 October 2026 on the same pages also says (2026-10-12-neat-shorts-8-percent-told-as-share-of-fat-gain). The recording now keeps to what was measured.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short's title and opening told one study's average as what happens when anyone overeats, as FitChef's own finding
What it said
The page and its recording were titled "Overeat, and Movement Does Most of the Extra Burning", and its opening said "Outside of workouts, every calorie you burn today sorts into three buckets."
What it says now
They are titled "In a Study, 16 People Overate. Movement Did Most of the Extra Burning.", and the opening says "Scientists sort every calorie you burn outside of workouts into three buckets."
Why
In Levine and colleagues' 1999 study, 16 adults without obesity were overfed by 1,000 calories a day for 8 weeks. On average, two-thirds of the rise in the energy they burned came from movement outside exercise, but the change ran from a fall of 98 calories a day in one volunteer to a rise of 692 in another. That is one study's average, not a rule about overeating: the titles and the page's notes for AI systems now report it as the study's, and the opening says the three buckets are how scientists sort the calories you burn. On 12 October 2026 we retitled the page (correction 2026-10-12-neat-shorts-overfeeding-response-told-as-daily-movement) so that it spoke of the rise in movement during overeating, not of everyday movement, but the new title still stated it as FitChef's own rule.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short told the authors' strolling equivalent for the largest NEAT rise as movement a volunteer was seen to make
What it said
It asked "What does a 692-calorie rise actually look like?" and said the volunteer "added the equivalent of about 15 minutes of light strolling per waking hour"; its recording said "The person whose movement rose most" and "It looks like nothing."
What it says now
It says "The authors worked out that it would match about 15 extra minutes of light strolling in every waking hour."
Why
Levine and colleagues (1999) worked out NEAT from total energy use, less resting and digestion costs, not from a record of how the volunteers moved. They wrote that the largest rise, 692 calories a day, "could be accounted for" by about 15 more minutes of strolling in each waking hour: an energy equivalent, not movement anyone was seen to make.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts on daily calorie burn told the change in NEAT during overeating as a rise in everyone; in one volunteer it fell
What it said
The NEAT Short said "How much it rose differed between people by hundreds of calories a day, the largest rise being 692"; the 10,000-steps Short said "How much their movement outside exercise rose differed widely".
What it says now
The NEAT Short says "How much it changed differed between people by hundreds of calories a day, from a fall of 98 to a rise of 692"; the 10,000-steps Short says "How much their movement outside exercise changed differed widely".
Why
In Levine and colleagues' 1999 study, the change in NEAT ran from a fall of 98 calories a day to a rise of 692. It fell in at least one volunteer, so the pages now speak of how much it changed.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said your body pushes back and fights extra calories; the study measured 16 volunteers
What it said
Its audio said "Your body is not sitting there passively storing everything. It pushes back." Its FAQ said "Your body actively resists caloric surplus through multiple defense mechanisms". The page closed "The clock your body is running answers a different question".
What it says now
It says "The eight-week study answers a different question" and its audio says "Their bodies did not store all of it."
Why
Levine and colleagues (1999) overfed 16 adults who were not obese by 1,000 calories a day for eight weeks. On average they burned off 531 of those calories a day, and the change in everyday movement outside exercise differed widely between them. The study measured those volunteers, not the reader.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said fat gain from the same surplus varies tenfold between people; the range was measured in 16 volunteers
What it said
The page said "Fat gain from an identical surplus varies tenfold between people." and "Individual fat gain from the same surplus varies tenfold."
What it says now
It says "In an eight-week study, 16 volunteers each ate 1,000 extra calories a day, and their fat gain ranged tenfold."
Why
Levine and colleagues (1999) found that fat gain "varied 10-fold among our volunteers": 16 adults aged 25 to 36 who were not obese, overfed by 1,000 calories a day for eight weeks.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said three processes had been running since your meal two hours ago; its studies overfed people for a week, seven weeks and eight weeks
What it said
The page said "Your body is not running that clock. It is running three." and "Three processes have been running in parallel".
What it says now
It says "Three overfeeding studies tested what extra food does over time." and "They fed people extra for a week, for seven weeks and for eight."
Why
Acheson and colleagues (1988) overfed three men for seven days, Rosqvist and colleagues (2014) overfed young adults for seven weeks, and Levine and colleagues (1999) overfed sixteen volunteers for eight weeks. None of them measured what one meal does to body fat. This corrects the rest of the lines the record 2026-10-12-overfeeding-study-eight-weeks-told-as-a-weekend-or-one-meal left on this page: it fixed the movement clause only.
Levine JA, Eberhardt NL, Jensen MD. Role of Nonexercise Activity Thermogenesis in Resistance to Fat Gain in Humans. Science. 1999;283:212-214.
Found by
Our full re-read of the page against all its papers.
Pontzer et al. 20212 corrections · 1 page
November 22, 2026 · We got it wrong
A 10,000-steps Short did not say it read Pontzer 2021 in the authors' accepted manuscript, and counted a review as a study
What it said
Its section for researchers said "This Short synthesizes findings from three peer-reviewed studies", and named Pontzer and colleagues' 2021 paper without saying which version was read.
What it says now
It says "This Short draws on two studies (Levine et al. 1999; Pontzer et al. 2021) and a review (Westerterp 2013)" and that Pontzer 2021 was "read here in the authors' accepted manuscript, PMC8370708".
Why
FitChef holds Pontzer and colleagues' 2021 paper as the authors' accepted manuscript on PubMed Central (PMC8370708), not as published in Science, and the page's notes for researchers now say so. Westerterp's 2013 paper is a review of other studies, not a study.
Pontzer H, Yamada Y, Sagayama H, et al. Daily Energy Expenditure through the Human Life Course. Science. 2021;373(6556):808-812. Read in the authors' accepted manuscript on PubMed Central (PMC8370708). doi:10.1126/science.abe5017.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A 10,000-steps Short said lean mass decides what you burn and steps are a rounding error; the study behind it spanned babies to 95-year-olds and never measured steps
What it said
It said "Not step count. Not exercise. Lean mass." and "steps are a rounding error on top of that structure", and its recording said "what determines whether you burn high or low was decided by what you're built from, not what you walked today".
What it says now
It says "Body size explained 83% of the differences between them, lean mass most of all" and "Even among people with the same lean mass, fat, sex and age, the daily burn still varied by more than 20% either way."
Why
Pontzer and colleagues (2021) measured daily energy use in 6,421 people aged 8 days to 95 years. Body size, particularly fat-free mass, accounted for 83% of the differences, across a group running from newborns to the very old. Among people with the same body composition, sex and age, energy use still varied by more than 20% either way. The study did not measure steps or compare them with lean mass, and its authors write that differences in physical activity and in how the body's tissues burn energy both shape daily energy use across a lifetime.
Pontzer H, Yamada Y, Sagayama H, et al. Daily Energy Expenditure through the Human Life Course. Science. 2021;373(6556):808-812. Read in the authors' accepted manuscript on PubMed Central (PMC8370708). doi:10.1126/science.abe5017.Read the paper
Found by
Our full re-read of the page against all its papers.
Westerterp 201313 corrections · 3 pages
November 22, 2026 · We got it wrong
A 10,000-steps Short said activity drops above a BMI of 35; its source reports a lower ratio of total to resting energy use there, a finding it cites from another paper
What it said
Its answer on body weight said total daily activity expenditure ends up similar "(except above BMI 35, where activity drops)".
What it says now
It says "Only above a BMI of 35 is their activity level, daily burn as a multiple of resting burn, lower."
Why
Westerterp's 2013 review reports that activity energy is similar or even higher in heavier people, and that only above a BMI of 35 is the physical activity level lower: total energy use divided by resting energy use, a finding it cites from Prentice and colleagues (1996).
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90. doi:10.3389/fphys.2013.00090.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A 10,000-steps Short said the body takes back the calories of exercise through hunger; its source only infers that hunger rose in one training study, whose finishers still lost fat
What it said
It said "The calories they burned during training were reclaimed through hunger before anyone noticed", "Walking triggers a quieter version of the same adjustment" and, in its recording, "The deficit exists on paper and dissolves in your body".
What it says now
It says "Those who finished did not lose weight", "The 2013 review that reports it says the training apparently made them hungrier over time" and "Younger and middle-aged adults who ate as much as they liked kept up their other movement, so their total activity went up."
Why
Westerterp's 2013 review reports a study in which sedentary adults aged 30 to 40 trained for about 40 weeks to run a half marathon. Those who finished did not lose weight, but the women lost about 2 kg of fat and gained 2 kg of fat-free mass, and the men lost about 4 kg of fat and gained 3 kg. The review infers that hunger rose ("Apparently ..."); it reports no measure of hunger. Nothing on the page shows that the calories of a walk are eaten back. In the training studies the review gathers, younger adults eating as they liked kept up their other movement, while older adults, and obese women on a diet, moved less outside their training.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90. doi:10.3389/fphys.2013.00090.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short gave the genetic share of how much people move as a reason NEAT may not change; the review that reports it reads it the other way
What it said
Its answer on raising NEAT gave the genetic share as a reason NEAT may not change, without what the review that reports it says of it.
What it says now
It adds "The review that reports it says this does not mean people inclined to sit must move less: it just takes them more effort to reach the same level."
Why
Westerterp's 2013 review writes that a high genetic share does not automatically mean that people predisposed to sit are less active than others: the activity level they reach comes from genes and environment together, and it only takes them more effort to reach the same level.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90. doi:10.3389/fphys.2013.00090.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short said twin studies put 72-78% of differences in movement down to genes; the figure comes from one twin study of 20 pairs
What it said
Its answer on raising NEAT said "twin studies estimate 72–78% of movement variation is genetic".
What it says now
It says "One twin study of 20 pairs worked out that genes account for 72–78% of the differences in how much people move".
Why
Westerterp's 2013 review reports one twin study, by Joosen and colleagues (2005): 20 same-sex twin pairs, aged about 25, whose activity was measured over two weeks. In it, genetic factors were calculated to account for 72-78% of the variance in physical activity. The page made one study into several.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90. doi:10.3389/fphys.2013.00090.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts on daily calorie burn cited Westerterp's 2013 review under a wrong or shortened title
What it said
The NEAT Short cited it as "Physical activity and obesity: contribution of energy expenditure (Westerterp 2013)", a title it does not have; the 10,000-steps Short gave only the first half of its title.
What it says now
Both cite it as "Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects".
Why
The review Klaas Westerterp published in Frontiers in Physiology in 2013 is titled "Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects". One page cited it under a title it never had, and the other cut the title short.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90. doi:10.3389/fphys.2013.00090.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on former athletes said their weight gain comes from an appetite trained up by years of training that does not reset; its source reports no such finding
What it said
The Short said former athletes gain weight because of "an appetite that was biologically driven upward by years of heavy training, and that doesn't reset when the training stops". It said "The appetite doesn't follow." and "A former athlete isn't overeating by choice." Its chart labelled about four times resting energy use "appetite still set here".
What it says now
It asks "Why do former athletes gain weight when they stop training?" and says "The research this page rests on never followed athletes after they quit." It reports a study of young adults in which "the ones who were more active at the start gained more fat, not less", and the review's account, in which "a switch to a less active routine does not cut how much people eat by the same amount".
Why
The page rests on one paper, a 2013 review by Klaas Westerterp of research that measured how much energy people use. No line of it is about athletes after they stop training, or about appetite staying high. Its one line on hunger is the author's reading of a training study in previously inactive adults aged 30 to 40: while they trained for a half marathon, the extra energy the training needed "eventually increased hunger". Its line on becoming less active is about young adults followed for more than ten years. It says an active lifestyle leads to a bigger fall in daily energy use later in life, and that a switch to a less active routine "does not induce an equivalent reduction of energy intake". The page said the fall in energy use was not what drives the gain; the review names both.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on former athletes said an athlete's daily energy use falls to 2.5 times resting after retirement; 2.5 is the most non-athletes keep up, and its source reports no measurement of retired athletes
What it said
It said of professional endurance athletes "After retirement, that falls to around 2.5 times. The same ceiling as everyone else." and called the fall from about four to 2.5 times "a potential 40-50% reduction in daily energy expenditure".
What it says now
It says "The most that people who are not athletes can keep up is about 2 to 2.5 times" and "No study in Westerterp's 2013 review measured athletes after they retired."
Why
The review gives 2.5 times resting energy use as the most non-athletes reach, and 2.0 to 2.5 as the top of the range people can keep up; adults average about 1.7 to 1.8. It reports no measurement of athletes after they retire. A fall from 4.0 to 2.5 would also be 37.5%, not 40 to 50%.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on former athletes said its findings came from professional endurance athletes; only one figure did, and its ten-year finding was in young adults
What it said
It said "these findings are from professional endurance athletes, the extreme end of the spectrum", and its section for researchers said "These findings are from professional endurance athletes." To the question "Do the most active athletes gain the most weight after stopping?" it answered yes, from the ten-year data.
What it says now
It answers that question "No study in Westerterp's 2013 review followed athletes after they stopped, so it cannot say." Its section for researchers says "The 4.0 figure is from professional endurance athletes; the 10-year fat-gain finding is from young adults".
Why
Only one of those findings comes from professional endurance athletes: that they can reach about four times their resting energy use for several weeks. The fat gain over more than ten years was measured in young adults, and the line on hunger comes from previously inactive adults aged 30 to 40 while they trained.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on former athletes credited a review's 647-person data table with findings that came from other studies
What it said
It said "Westerterp's doubly labeled water review (n=647) found that the most physically active people gained the most fat over a decade of follow-up", described the method as "Doubly labeled water (DLW) review across 647 subjects", and its evidence notes for AI systems named the 647 beside each finding.
What it says now
Its section for researchers says "The review's own data table covers 647 healthy adults living in Northern Europe, each measured over 14 days" and "The findings below come from separate studies it reports."
Why
In the review, 647 is the number of adults in its own table of two-week measurements at one site (301 women and 346 men). The fat gain over more than ten years comes from a separate study of young adults the review reports (Westerterp and Plasqui, 2009), the four-times figure from a study of athletes (Westerterp and others, 1986), and the hunger line from a training study of 32 people (Westerterp and others, 1992).
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said professional endurance athletes burn four times their resting energy every day; the review says they can reach it for several weeks
What it said
It said "professional endurance athletes burn roughly four times their resting metabolism", its section for researchers said they sustain it, and its recording said "your body was burning four times what it burns at rest. Every single day."
What it says now
It says "Professional endurance athletes can reach about four times their resting energy use, and keep that up for several weeks."
Why
The review says professional endurance athletes "can reach a value around 4.0 and can maintain this values for several weeks". It does not say they keep it up every day.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said the method that measures daily energy use captures every calorie; it is accurate to within a few percent
What it said
It said daily energy use was "tracked with a method precise enough to capture every calorie from training sessions to sleep".
What it says now
Its section for researchers says validation studies found the method "accurate to 1–3% (precision 2–8%) against respirometry".
Why
The review says validation studies of doubly labeled water found an accuracy of 1 to 3% and a precision of 2 to 8%, compared with respirometry. Close, but not every calorie.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short's recording told its listeners the page's claims about former athletes as what their own body did
What it said
Its recording said "Your daily burn fell from roughly four times your resting level to around 2.5." and "Your instinct is correct." It said "You're eating exactly what your body is telling you to eat." and "You're not battling a broken metabolism."
What it says now
The recording now asks the page's question and says "The research behind this episode never followed athletes after they quit." It tells no listener what their own body did.
Why
No study in the review measured the listener, or any former athlete. The four-times figure is from professional endurance athletes in training, and 2.5 is the most the review gives for non-athletes. Neither was measured in anyone after they stopped training.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on former athletes graded a ten-year link between activity and fat gain as high-certainty evidence; the review reports it from one observational study
What it said
Its evidence notes for AI systems said "People who started at the highest activity levels gained the most fat over more than a decade of follow-up" and graded it "High — longitudinal observational data within DLW-measured cohort". Its section for researchers said "Subjects with the highest initial activity levels gained the most fat".
What it says now
Its section for researchers says "In young adults observed over an average of more than 10 years (Westerterp and Plasqui, 2009), subjects with a higher activity level at the start had a higher fat gain at follow-up." Its evidence notes for AI systems now grade that link moderate.
Why
The review reports the finding from one study of young adults followed over an average of more than ten years (Westerterp and Plasqui, 2009): "subjects with higher activity level at the start were those with a higher fat gain at follow up after more than 10 years". That is a link seen in one observational study, and the review does not say how many people it followed. Its words compare higher and lower activity at the start; they do not single out the most active.
Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol. 2013;4:90.Read the paper
Found by
Our full re-read of the page against all its papers.
A 10,000-steps Short said walking supports the lean tissue that drives your burn; no source on the page tested it
November 22, 2026 · We got it wrong
What it said
It said "the movement supports the lean tissue that actually drives your burn, and nobody should stop".
What it says now
It says "Walking is not a waste. The calories it burns are real."
Why
None of the page's sources tested what walking does to lean tissue. The review it cites reports lean mass gained over a running programme, not from walking.
Our full re-read of the page against all its papers.
Levine et al. 20052 corrections · 1 page
November 22, 2026 · We got it wrong
A NEAT Short's recording told a small pilot study's result as a rule about lean and heavier people
What it said
Its recording said "Lean people naturally stand and walk about two hours more per day than people carrying extra weight."
What it says now
Its recording says "In a small pilot study, the lean volunteers stood and walked over two hours more a day than the mildly obese ones."
Why
Levine and colleagues (2005) measured 10 lean and 10 mildly obese volunteers who were all sedentary, for 10 days. The lean volunteers were upright about 152 minutes a day longer. The authors call it a pilot study whose results need to be confirmed in larger studies.
Levine JA, Lanningham-Foster LM, McCrady SK, Krizan AC, Olson LR, Kane PH, Jensen MD, Clark MM. Interindividual Variation in Posture Allocation: Possible Role in Human Obesity. Science. 2005;307:584-586. doi:10.1126/science.1106561.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A NEAT Short used a 2005 posture study without citing it, and gave that study and a twin study's figure one DOI, that of a 2013 review
What it said
Its lines on posture rested on a 2005 study that was in neither its citations nor its sources, and its evidence notes for AI systems put that study and the twin figure in one claim under the DOI of Westerterp's 2013 review.
What it says now
It cites the study as "Interindividual variation in posture allocation: possible role in human obesity (Levine et al. 2005)", and its notes for AI systems give each paper its own claim and its own DOI.
Why
Levine and colleagues' posture study was published in Science in 2005, under its own DOI. The twin figure comes from a 2005 twin study by Joosen and colleagues, as Westerterp's 2013 review reports it. One DOI cannot point to both, and the page now names the posture study among its sources.
Levine JA, Lanningham-Foster LM, McCrady SK, Krizan AC, Olson LR, Kane PH, Jensen MD, Clark MM. Interindividual Variation in Posture Allocation: Possible Role in Human Obesity. Science. 2005;307:584-586. doi:10.1126/science.1106561.Read the paper
Found by
Our full re-read of the page against all its papers.
A NEAT Short said what most people believe about their metabolism and what they track, with no source for it
November 22, 2026 · We got it wrong
What it said
It said NEAT's role "is the part most people have radically wrong" and that "most people put resting metabolism on top", and called resting metabolism "the component everyone blames" and movement "the component almost nobody tracks".
What it says now
It puts these as the reader's own likely view: "Stacked in order of importance, you might put resting metabolism on top" and "the component you might blame first".
Why
No study on the page measured what people believe about their metabolism or what they keep track of. The page now says what a reader might think, without a claim about most people.
Our full re-read of the page against all its papers.
A Short told the muscle question as settled, leaving out that the review it leans on cites more muscle protein lost in a fasted session
November 22, 2026 · We got it wrong
What it said
It said "What is not real is the conclusion most people draw from it" and gave the fear's biology as fat fuel only.
What it says now
It says "And one study cited in a 2017 review found a greater loss of muscle protein in a single fasted session than in a fed one."
Why
Hackett and Hagstrom (2017) note that an acute bout of fasted exercise has been shown to cause a significantly greater loss of muscle protein than fed exercise, which could matter over weeks or months, and that their short trials of women still found no difference in lean mass.
Hackett D, Hagstrom AD. Effect of Overnight Fasted Exercise on Weight Loss and Body Composition: A Systematic Review and Meta-Analysis. J Funct Morphol Kinesiol. 2017;2(4):43.Read the paper
Found by
Our full re-read of the page against all its papers.
Schoenfeld et al. 20146 corrections · 2 pages
November 22, 2026 · We got it wrong
A Short told a four-week result in young women as true for anyone who trains
What it said
It said "anyone eating reasonably and training within human ranges — the fasted session is not costing them muscle".
What it says now
It says "these trials studied young adults for four to six weeks, and only women had their lean mass measured".
Why
Schoenfeld and colleagues (2014) studied young, non-obese women for four weeks and write that the results cannot necessarily be generalized. In the 2017 review only the trials of women measured lean mass.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said a trial measured everything the myth predicts and nothing changed, when weight and fat fell in both groups
What it said
It called it "The most direct test of this question", said the gap was "well inside measurement noise", and that the trial "measured everything the myth predicts should change — and nothing did".
What it says now
It says "Fat-free mass did not change in either group" and "Both groups lost weight and some fat, and the fat they lost did not clearly differ".
Why
In Schoenfeld and colleagues' trial (2014), weight and fat mass fell significantly in both groups; fat-free mass did not change. The paper reports no measurement error and no source ranks the trial.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short gave a trial's starting body mass index as about 22; it was about 23
What it said
Its research notes said "20 young women (BMI ~22)".
What it says now
They say "20 young women (BMI ~23)".
Why
Schoenfeld and colleagues (2014) report a starting body mass index of 23.4 in the fasted group and 23.3 in the fed group.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short gave its finding as FitChef's own verdict in its title and the lines that travel
What it said
Its title said "Fasted Cardio Burns Something Real. It Isn't Muscle." and its search title "Cardio on an Empty Stomach Won't Burn Muscle".
What it says now
Its title asks "Does Fasted Cardio Burn Muscle? What the Trials Found" and its search title asks "Does Cardio on an Empty Stomach Burn Muscle?"
Why
Lines that travel without the page ask the question or name the research. The finding is a four-week trial's: fat-free mass did not change in 20 young women whether they trained fasted or fed.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said fasted and fed cardio gave identical fat loss and that fasting bought nothing, from one small trial that found no clear difference
What it said
It said "Fasted and fed exercise produced identical fat loss" and "The empty stomach bought nothing", its recording said "Identical fat loss whether people ate before or not", and its notes for AI systems said fasted and fed exercise "produce identical fat loss and body composition changes".
What it says now
It says "Both groups lost weight and fat, and the trial found no clear difference between them", and that "a small edge for either side cannot be ruled out".
Why
Schoenfeld and colleagues (2014) had 20 young women on a calorie-reduced diet do the same cardio for four weeks, either fasted or after a meal. Both groups lost weight and fat, with no significant difference between them, and the authors write that with a small sample and a short study they cannot rule out a small benefit for either. The Short told that as identical fat loss, and as fasting buying nothing.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a trial's explanation from earlier research, that the body evens out its fat burning over a day, as something the trial measured
What it said
It said "The body adjusts its fuel mix hour by hour", "A fasted session shifts toward fat during the workout, then compensates afterward" and "Over a full day, the totals even out", its recording said "The body adjusts across the day", and its sources line said the trial showed "substrate compensation over 24 hours".
What it says now
It says "Its authors point to earlier research: burning more fat during one part of the day is balanced by burning more carbohydrate later."
Why
Schoenfeld and colleagues (2014) measured weight and body composition, not how much fat was burned through the day. That fat used during exercise is balanced by more carbohydrate used later in the day is the authors' reading of earlier research they cite. The Short told it as the trial's result, and the day's totals evening out as a fact.
Schoenfeld BJ, Aragon AA, Wilborn CD, Krieger JW, Sonmez GT. Body composition changes associated with fasted versus non-fasted aerobic exercise. J Int Soc Sports Nutr. 2014;11:54.Read the paper
Found by
Our full re-read of the page against all its papers.
Acheson et al. 19885 corrections · 1 page
November 22, 2026 · We got it wrong
A Short on extra calories answered in its search, share and answer lines as FitChef's own verdict, with no study named
What it said
Its answer for AI systems began "Extra calories do not convert to body fat on a single predictable timeline." and its share description said "Three separate pathways".
What it says now
Its answer begins "Three overfeeding studies give parts of the answer, and none of them measured what one meal does to body fat."
Why
The page rests on three overfeeding studies (Acheson and colleagues 1988, Rosqvist and colleagues 2014, Levine and colleagues 1999). Lines read away from the page now name that research, or ask, instead of stating a verdict about the body as FitChef's own.
Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt JP, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr. 1988;48(2):240-7. doi:10.1093/ajcn/48.2.240.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said almost nobody knows the body stores carbohydrate as glycogen; no source says so
What it said
The page said "a buffer system stands between you and fat storage — and almost nobody knows it exists".
What it says now
It says "If most of the surplus came from carbohydrates, a buffer system stands between you and fat storage."
Why
No study on the page says how many people know about glycogen.
Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt JP, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr. 1988;48(2):240-7. doi:10.1093/ajcn/48.2.240.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said no carbohydrate is turned into fat until the glycogen store has grown by 500 grams; the study says fat-making only overtook fat burning then
What it said
The page said "before any conversion to body fat begins" and its answer for AI systems said "before any conversion to fat begins".
What it says now
It says "they began making fat faster than they burned it" once the store had grown by about 500 grams.
Why
Acheson and colleagues (1988) say the stores can take a gain of about 500 grams "before net lipid synthesis contributes to increasing body fat mass", and that the stores must grow by 500 grams "before appreciable de novo lipogenesis begins". Net means more fat made than burned, and appreciable means a noticeable amount: neither means none at all.
Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt JP, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr. 1988;48(2):240-7. doi:10.1093/ajcn/48.2.240.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said a quarter of the energy is lost as heat when carbohydrate becomes fat, as if measured; the researchers estimated that cost from the energy each step needs
What it said
The page said "And the conversion itself is expensive" and that the energy "gets burned off as heat". Its audio said "it burned a quarter of the energy just doing the conversion" and "A quarter lost to heat."
What it says now
It says "The same team worked out that turning carbohydrate into fat should cost about" a quarter of its energy, and "Overall, the men kept about three-quarters of the extra energy they ate."
Why
Acheson and colleagues (1988) calculated that about 25% of the energy of glucose turned into fat "can be expected to be needed for this process", from the cost of each step. What they found from their measurements is that the men kept 75% of the energy they ate above their estimated needs and burned off 25%.
Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt JP, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr. 1988;48(2):240-7. doi:10.1093/ajcn/48.2.240.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on extra calories said the body's carbohydrate store holds 500 grams and must fill over two days before any fat is made; in the study, fat-making overtook fat burning once the store had grown by about 500 grams, and it kept filling
What it said
The page said "Until it fills, your body is storing the excess as quick-access fuel, not as fat." and that the tank took two full days "to reach capacity before the body began converting any carbohydrate into fat". Its audio said "Five hundred grams of stored glucose in your muscles and liver have to reach capacity before any conversion begins."
What it says now
It says "One study put the tank's full size at about 15 grams per kilogram of body weight." and that once the store had grown by about 500 grams, "they began making fat from carbohydrate faster than they burned fat." It adds "The tank kept filling until day four or five, growing by 810 grams on average."
Why
Acheson and colleagues (1988) overfed three young men with huge amounts of carbohydrate for seven days. When their glycogen stores had grown by about 500 grams, at the end of the second day, burning and storing carbohydrate could no longer keep up, and the excess was turned into fat. The stores kept filling until day four or five, by 810 grams on average. The paper puts the store's full size at about 15 grams per kilogram of body weight. Our page told the 500 grams as the store's whole size, filled in two days before any fat was made.
Acheson KJ, Schutz Y, Bessard T, Anantharaman K, Flatt JP, Jéquier E. Glycogen storage capacity and de novo lipogenesis during massive carbohydrate overfeeding in man. Am J Clin Nutr. 1988;48(2):240-7. doi:10.1093/ajcn/48.2.240.Read the paper
Found by
Our full re-read of the page against all its papers.
King et al. 20228 corrections · 1 page
November 22, 2026 · We got it wrong
A Short cited a review under a cut title and a FitChef claim under a title the claim does not carry
What it said
It cited King and colleagues' review as "The Ergogenic Effects of Acute Carbohydrate Feeding on Resistance Exercise Performance (King et al. 2022)" and FitChef's claim as "Does carb timing matter for fat loss? (FitChef)".
What it says now
It cites them as "The Ergogenic Effects of Acute Carbohydrate Feeding on Resistance Exercise Performance: A Systematic Review and Meta-analysis (King et al. 2022)" and "Does Carb Timing Actually Matter? What 4 Analyses Found (FitChef)".
Why
The review's title, as printed in Sports Medicine (2022), runs on ": A Systematic Review and Meta-analysis", and the claim page the citation links to is titled "Does Carb Timing Actually Matter? What 4 Analyses Found"; it is not about fat loss.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said two things decide whether pre-workout carbs help, leaving out the one its review tested and confirmed: the number of all-out sets
What it said
It said "What decides whether pre-workout carbs help is how long the session runs and how long you've gone without eating" and "Two other things were", its speakable said "What decides whether pre-workout carbs help is training longer than 45 minutes or training after a long fast", its recording said "What did matter was two things almost nobody mentions", its sources line said "session >45 min and fast >=8 h are the moderators", and its research notes said "session duration and fasting status do".
What it says now
It says "And the more all-out sets a session held, the bigger the benefit", and its research notes say the number of maximal-effort sets moderated the effect, while "sub-groups showed an effect in sessions over 45 min and after fasts of 8 h or longer".
Why
King and colleagues (2022) found the benefit of carbohydrate more likely in sessions longer than 45 minutes and after a fast of 8 hours or more, by sub-group; their analysis across studies found that the number of all-out sets, not the dose, changed the size of the benefit. The Short named the two sub-groups, left out the sets, and called them what decides.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said the carbs that power today's workout are mostly the ones eaten yesterday, and gave an untested reason as fact
What it said
Its shareable line and recording said "The carbs that power today's workout are mostly the ones you ate yesterday", its share description said "the carbs powering today's workout are mostly the ones you ate yesterday", it said "There is a simple reason a short, fed workout doesn't lean on a pre-gym carb hit" and "The fuel you banked this morning, and across yesterday, is still mostly there", an answer said a short session "leaves most of your stored carbohydrate in place" and "so you walk in with plenty already in the tank", and its recording said "You're nowhere near empty."
What it says now
It says "Shorter sessions showed no clear gain from carbs, and why is not known", and "The review's authors call the glycogen explanations speculative."
Why
No source on the page measures how much of a session's fuel comes from the day before, or tested why short sessions showed no clear gain. King and colleagues (2022) call their glycogen explanations speculative, and report that a carbohydrate meal after an overnight fast can raise muscle glycogen within hours; the position stand cites one study's 39% drop after six sets of leg extensions. This takes off the page words the correction of 14 October 2026 (2026-10-14-pre-workout-carbs-short-one-leg-extension-study-told-as-any-session) had written ("leaves most of your stored carbohydrate in place"), because they gave the 39% as the reason.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a review's finding that the carb dose did not seem to change the benefit as a settled result
What it said
It said "the size of the carb dose made no measurable difference to how the session went" and that the research "already says plainly" that the gram-counting before the gym "is aimed at the dial that barely turns", its takeaway said "For an ordinary hour of lifting, the amount barely changes your performance", its recording said "The big dose and the small dose came out the same" and "The grams right before barely move the needle", and its chart said the amount has "no real effect".
What it says now
It says "the size of the carb dose did not seem to change how much" carbs helped, its takeaway says "For lifting, a 2022 review found the amount did not seem to change how much carbs helped", and its chart says "no clear effect".
Why
King and colleagues (2022) write that carbohydrate dose "does not seem to influence" the benefit; the dose was tested across studies that used different amounts, its estimate was wide, and the authors rate the evidence low to moderate. The Short told that as the doses coming out the same, as an amount that barely changes performance, and as something the research says plainly.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a review of carbs taken before or during lifting as 21 trials of carbs before it, all pooled together
What it said
It said "When 21 trials testing carbs before resistance training were pooled together", its recording said "gather up every study that tested carbs before lifting, twenty-one of them", and its research notes gave the volume result as from "21 crossover trials".
What it says now
It says "A 2022 review gathered 21 studies that gave lifters carbs before or during a session", and its research notes say "12 studies pooled for this outcome".
Why
King and colleagues (2022) reviewed 21 crossover studies of carbohydrate taken before and/or during resistance training, compared with a placebo or water; the pooled result for training volume came from 12 of them, and the authors tie the long-session result to carbohydrate taken before and during the session. The Short told them as 21 trials of carbs before training, all pooled.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told carbs before a workout as barely changing a normal, hour-long lift, where the review it rests on found they helped, more likely past 45 minutes
What it said
Its search title said "Carbs You Eat Before a Workout Barely Change a Normal Lift", and its recording said "a normal, fed workout doesn't depend on it".
What it says now
Its search title asks "How Many Carbs Before a Workout? What a Review of 21 Lifting Studies Found", and its recording says that past about forty-five minutes, "carbs taken before and during the session were more likely to help".
Why
King and colleagues (2022) found that carbohydrate improved training volume compared with a placebo or water, more likely in sessions longer than 45 minutes; the page itself calls a normal session an hour in the weight room. What did not seem to matter was the dose, not the carbs.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told carbs before fasted morning lifting as a real boost, more than at almost any other time, where the review found a small benefit that depended on when they were eaten
What it said
It answered "More than at almost any other time", said eating carbs first "gives a real boost to how much work you can finish" and that for a session a few hours after a meal "the effect mostly fades", and its recording said "carbs beforehand made a real difference".
What it says now
It says "carbs helped a little when the session came after a fast of 8 hours or more", and "The two studies with the biggest gains gave the carbs half an hour to two hours before."
Why
In the review by King and colleagues (2022), carbohydrate after a fast of 8 hours or more gave a small benefit; the two studies with the largest effects gave it half an hour to two hours before training, and two that gave it just before and during training found no gain. After a shorter fast the benefit was not clear, which the Short told as an effect that mostly fades.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told what carbs before a workout do as its own verdict in its title and in lines shown away from the page, naming no research
What it said
Its title said "Carbs Before a Workout: The Grams Matter Less Than the Clock", its search description said "For how many carbs to eat before a workout when you lift, the amount barely matters", and its summary for search and AI systems opened "For a typical gym workout, the amount of carbohydrate eaten beforehand barely changes performance".
What it says now
Its title asks "How Many Carbs Before a Workout? What 21 Lifting Studies Found", and its search description says "In a review of 21 lifting studies, the amount did not seem to change the benefit."
Why
Lines that travel without the page (its title, search lines, shareable line and summaries) state no fitness verdict as FitChef's own: they ask the page's question or report the research by name. These lines gave the finding of King and colleagues (2022) as FitChef's own verdict, and "the clock" as settled.
King A, Helms E, Zinn C, Jukic I. The ergogenic effects of acute carbohydrate feeding on resistance exercise performance: a systematic review and meta-analysis. Sports Med. 2022;52(11):2691-2712.Read the paper
Found by
Our full re-read of the page against all its papers.
Kerksick et al. 20178 corrections · 3 pages
November 22, 2026 · We got it wrong
A Short's research notes gave the position stand's pre-exercise carbohydrate figure in other units, and one leg exercise's glycogen drop as all resistance exercise
What it said
Its research notes said the stand "recommends 1–4 g/kg of carbohydrate in the hours before prolonged" exercise, and "Resistance exercise depletes vastus lateralis glycogen by about 39%".
What it says now
They say the stand gives it "printed as 1–4 g/kg/day", that "this page reads it as 1–4 g/kg in the hours before the session", and "In one study, six sets of 12RM leg extension reduced vastus lateralis glycogen by 39%."
Why
The ISSN position stand (Kerksick et al. 2017) prints the recommendation as meals or snacks high in carbohydrate (1–4 g/kg/day) for several hours before longer, harder exercise; the page reads it as grams per kilogram before the session, and now says so. The 39% drop is from one study of six sets of leg extensions, which the stand cites as such. The correction of 14 October 2026 (2026-10-14-pre-workout-carbs-short-one-leg-extension-study-told-as-any-session) fixed the reader lines but left the research notes.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said endurance athletes eat toward the top of the pre-exercise carbohydrate range before a long race, which no source on it says
What it said
It said "Endurance athletes eat toward the top before a long race because they are about to burn straight through it."
What it says now
That line is gone, and it marks its food examples as its own: "By our rough count, that is about a banana with toast and a glass of juice."
Why
No source on the page says what endurance athletes eat before a race, or why. The foods matched to 70 and 280 grams of carbohydrate are FitChef's own examples, and the page now says so.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said the day's carbohydrate outweighs the snack before the gym, which no study on it compared, and handed the reader an athletes' daily range built on endurance athletes
What it said
It said "That is why the daily habit outweighs the pre-gym snack" and that a body eating 5 to 12 grams per kilo a day "walks in already fueled", its summary said "Total daily carbohydrate intake (about 5 to 12 g/kg/day) matters more than the pre-workout amount", its recording said "Where it actually moves is your daily total", its link line spoke of "when the daily total quietly does all the work", and its research notes said "total daily intake remains the priority".
What it says now
It says the position stand ranks timing below the day's amount, "when the day's carbs meet the body's needs and fast recovery isn't needed, the timing of carbs matters much less", and that its daily range for athletes "was built mostly on endurance athletes".
Why
No study on the page compared the day's carbohydrate with carbohydrate eaten before training. The ISSN position stand (Kerksick et al. 2017) says timed carbohydrate matters much less when daily intake meets energy needs and fast recovery is not needed; it reports 5 to 12 g/kg/day as commonly recommended daily intakes for athletes, notes that most carbohydrate recommendations are based on endurance athletes, and says strength-power athletes may need less, though more research is needed.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a possible blood sugar dip at the start of exercise as a crash in the middle of the session
What it said
It spoke of the fear that eating carbs right before training "triggers a mid-session energy crash", and its recording said it "gives you a sugar crash mid-session".
What it says now
It says "Carbs eaten shortly before training can make blood sugar dip at the start of exercise", and its recording says the fear is that it "makes your blood sugar dip as the session starts".
Why
The ISSN position stand (Kerksick et al. 2017) places the possible drop in blood sugar after carbohydrate eaten before exercise at the onset of activity, which is why, in the one study it cites, the equivalent of a proper warm-up negated it.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short told a blood sugar dip after carbs just before training as a worry to retire and a question of comfort, where the position stand says a number of athletes may be affected
What it said
It said "One old worry deserves retiring" and "Eating close to training is a comfort question, not a performance threat", answered that eating close to a workout "is really a comfort choice, not a performance risk", and its recording said "So eating close to your workout is about comfort, not performance."
What it says now
It says "One old worry needs a closer look", and that the dip "can affect some athletes, but in one study a normal warm-up cancelled it, and eating closer to the start of training made it less likely".
Why
The ISSN position stand (Kerksick et al. 2017) says a number of athletes may be affected by the dip, and that too much carbohydrate, especially fructose, in the hours before exercise may hurt performance; in one study the equivalent of a proper warm-up negated the dip, and eating closer to the start made the symptoms less likely. It does not say the worry can be retired, or that eating close to training is only a matter of comfort. The correction of 14 October 2026 (2026-10-14-pre-workout-carbs-short-rebound-low-told-as-rare) fixed how often the dip happens but left these lines.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short described how insulin refills muscle on a rest day and how changing carbs between days amplifies the swing, which no source on it reports
What it said
It said "Your body registers that depletion immediately. Insulin drives glucose back into the depleted muscle cells", "Rest-day eating accelerates the process", and "the swing gets amplified. Lower carbs on rest days mean partial depletion."
What it says now
It says "Eating carbohydrate after training refills it", and that if carb intake changes between training and rest days, "your glycogen may swing more too. None of this page's studies tested that."
Why
The ISSN position stand (Kerksick et al. 2017) reports that carbohydrate eaten after exercise restores muscle glycogen. None of the page's sources describes insulin driving glucose into muscle at once, says that eating on a rest day speeds the process, or tested what changing carbohydrate between training and rest days does to glycogen or water.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short on when to stop eating before a workout closed by telling readers that when they eat never mattered, only how much, where the trial it leads with found the time mattered
What it said
It closed with "You were counting minutes. The evidence was counting grams." and "The backward calculation that brought you here was tracking the wrong number", and its recording closed with "Not when to eat. Whether you ate enough."
What it says now
It closes with "How long you train, how big the meal was and how your stomach feels shape the timing, not a fixed clock", and its recording with "So the clock was never the whole answer."
Why
The trial the Short leads with, which the ISSN position stand (Kerksick et al. 2017) cites, found that a carbohydrate drink 30 minutes before hard cycling gave greater gains in exercise capacity than the same drink 2 hours before: the time mattered. The 2018 review the Short cites compared eating with not eating before exercise, and no study on the page tested how much is eaten before a session. Aragon and Schoenfeld (2013) leave the exact timing to preference, tolerance and how long the session lasts.
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
Two Shorts cited the ISSN position stand on nutrient timing under a shortened title
What it said
They cited it as "ISSN Position Stand: Nutrient Timing".
What it says now
They cite it as "International society of sports nutrition position stand: nutrient timing".
Why
The paper's title, as printed, is "International society of sports nutrition position stand: nutrient timing" (J Int Soc Sports Nutr 14:33, 2017).
Kerksick CM, Arent S, Schoenfeld BJ, et al. International society of sports nutrition position stand: nutrient timing. J Int Soc Sports Nutr. 2017;14:33.Read the paper
Found by
Our full re-read of the page against all its papers.
A Short said caffeine follows the same logic as its carbohydrate finding, where caffeine's dose is what is set, by body weight
November 22, 2026 · We got it wrong
What it said
It said "The same logic sits underneath the other thing people load up on before training", and its stored link note said a reader is "primed for the same 'the amount isn't the lever' lesson on caffeine".
What it says now
It says "Caffeine, the other thing people load up on before training, runs on its own dose rule".
Why
The ISSN position stand on caffeine (Guest and colleagues, 2021), which the linked caffeine page rests on, gives caffeine's dose by body weight, 3 to 6 mg per kilogram, and says very high doses bring many side effects and do not seem to be needed. For caffeine the dose is set to the person, not a fixed scoop. The carb page's finding, that the amount did not seem to change the benefit, does not carry over.
Our full re-read of the page against all its papers.
A Short's answer gave a 4.8% one-week jump in fat-free mass without the ten weeks on a ketogenic diet that came before it
November 22, 2026 · We got it wrong
What it said
Its answer said "participants who reintroduced carbohydrates saw their fat-free mass jump 4.8% in one week", with no word of the diet before it.
What it says now
It says "people who went back to eating carbohydrates after ten weeks on a ketogenic diet saw their fat-free mass jump 4.8% in one week".
Why
Henselmans and colleagues (2026) report a trial by Wilson and colleagues: after ten weeks on a ketogenic diet, participants who went back to eating carbohydrates in the final week gained a further 4.8% in fat-free mass, likely from glycogen refilling. The answer gave the gain without the ketogenic diet before it, as if it followed any return to carbohydrates.
Henselmans M, Vårvik FT, Izquierdo M. The Effect of Carbohydrate Intake on Muscle Hypertrophy: A Systematic Review and Meta-analysis. Sports Med. 2026;56:691-702.Read the paper
Found by
Our full re-read of the page against all its papers.
Fernández-Elías et al. 20152 corrections · 1 page
November 22, 2026 · We got it wrong
A Short called a study that tested nine people on two occasions a controlled trial
What it said
It said "Muscle biopsies from a controlled trial measured the ratio directly".
What it says now
It says "A study took muscle samples from nine trained people and measured the ratio directly".
Why
Fernández-Elías and colleagues (2015) tested the same nine aerobically trained people on two occasions, once drinking little water with their carbohydrate and once enough to replace what they had lost; their abstract names no control group and no random assignment.
Fernández-Elías VE, Ortega JF, Nelson RK, Mora-Rodriguez R. Relationship between muscle water and glycogen recovery after prolonged exercise in the heat in humans. Eur J Appl Physiol. 2015;115:1919-1926.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short still told rest-day puffiness as glycogen and water for certain, in lines an earlier correction left
What it said
It said "Nothing went wrong. What you are seeing is what recovery actually looks like from the inside out", "Not because you lost fat overnight, but because fuel left and the water went with it", "Your visual cycle tracks your fuel, not your progress" and "Much of what changes day to day is glycogen and water", its share line said "A puffier rest-day mirror is not a sign that recovery failed", its search title said "Often Fuel, Not Fat", and its recording said "Same muscle. Same fat. Different fuel level.", "The puffiness follows the fuel" and "The glycogen-water mechanism is well-established at the tissue level."
What it says now
It says "What you are seeing can be what recovery looks like from the inside out", and its share line asks "Does a puffier rest-day mirror mean recovery failed?"
Why
No source on the page measured rest-day puffiness or what causes it. Fernández-Elías and colleagues (2015) measured the glycogen and water in muscle samples from nine people recovering from cycling in the heat, and call the ratio a long-held notion their findings agree with; the ISSN position stand (2017) cites one study's 39% glycogen drop after six sets of leg extensions. The correction of 15 October 2026 (2026-10-15-rest-day-bloating-short-puffiness-told-as-glycogen-for-certain) made the summary say puffiness can come from glycogen refilling; these lines still told it as certain.
Fernández-Elías VE, Ortega JF, Nelson RK, Mora-Rodriguez R. Relationship between muscle water and glycogen recovery after prolonged exercise in the heat in humans. Eur J Appl Physiol. 2015;115:1919-1926.Read the paper
Found by
Our full re-read of the page against all its papers.
A Short still listed a 2021 diet trial among its sources, though no sentence on the page rests on it
November 22, 2026 · We got it wrong
What it said
It listed the 2021 diet trial by Hall and colleagues among its sources, with that trial's finding on first-week weight loss.
What it says now
Its sources no longer list that trial.
Why
No sentence of the page uses the trial by Hall and colleagues (2021); only its sources list named it. This takes off the page the citation and the sources line that the correction of 15 October 2026 (2026-10-15-rest-day-bloating-short-hall-2021-title-and-first-week-loss) had put right, because no line rests on them.
Hall KD, Guo J, Courville AB, et al. Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nat Med. 2021;27:344-353.Read the paper
Found by
Our full re-read of the page against all its papers.
Aragon et al. 20132 corrections · 1 page
November 22, 2026 · We got it wrong
A Short gave a review's 20 to 40 gram protein dose as what its per-kilogram guideline works out to for most people
What it said
It said that for 0.4 to 0.5 grams of protein per kilogram of lean body mass, "For most people, that works out to about 20 to 40 grams of protein per meal".
What it says now
It gives the review's own example, "someone with 70 kg of lean body mass would eat roughly 28 to 35 grams at each", and says the guideline reflects evidence that "20 to 40 grams at a time gives the largest short-term muscle-building effect".
Why
Aragon and Schoenfeld (2013) suggest 0.4 to 0.5 g of protein per kg of lean body mass at the meals before and after training, and give one example: 70 kg of lean body mass, roughly 28 to 35 g. The 20 to 40 g is the dose range the guideline reflects, from evidence of the largest short-term muscle-building effect, not what the guideline comes to for most people; no source says that.
Aragon AA, Schoenfeld BJ. Nutrient timing revisited: is there a post-exercise anabolic window? J Int Soc Sports Nutr. 2013;10:5.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A Short said four things about pre-workout meals and the research behind them that none of its sources says
What it said
It said "Most pre-exercise timing research used controlled meal compositions in lab settings", "A banana thirty minutes before is already fuel" and "A smaller meal or snack needs far less buffer", and its recording called a full meal's extra time "comfort, not physiology".
What it says now
Those lines are gone. Its recording says one review notes that "a meal's nutrients can take three to six hours or more to return to normal levels in your blood, depending on its size".
Why
None of the page's sources says how most timing research was run, that a banana eaten 30 minutes before training is already fuel, or that a smaller meal needs far less time before exercise. Aragon and Schoenfeld (2013) note that a meal's circulating nutrients take 3 to 6 hours or more to return to baseline depending on its size, so meal size is not only a matter of comfort.
Aragon AA, Schoenfeld BJ. Nutrient timing revisited: is there a post-exercise anabolic window? J Int Soc Sports Nutr. 2013;10:5.Read the paper
Found by
Our full re-read of the page against all its papers.
Gautam et al. 201011 corrections · 1 page
November 22, 2026 · We got it wrong
A garlic and onion Short said sulfur compounds wrap iron and shield it from phytates, a mechanism its study's summary does not describe
What it said
The page said "Garlic and onion contain sulfur compounds that wrap around iron in grains and legumes, shielding it from phytates that would otherwise lock it away." and "The sulfur wraps around iron ions and holds them in a form your gut can actually reach." Its audio said "They wrap around iron before the phytates can, forming a shield that keeps the iron available."
What it says now
It gives sulfur as the researchers' reason for trying garlic and onion: "Both are rich in sulfur compounds, and earlier reports said building blocks of protein that contain sulfur improved the mineral levels of lab animals." It adds "The summary gives that as the reason for the test, and does not say how it would work."
Why
The study's summary gives sulfur only as the reason the researchers chose garlic and onion: earlier reports said building blocks of protein that contain sulfur improved the mineral levels of lab animals. Its result is that garlic and onion raised how much of the iron and zinc in food grains came free to be absorbed. It describes no way the effect works: no wrapping or shielding of iron, no phytates, no pathway.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short's title promised an iron rescue in every clove and its search title said they help your body absorb 73% more iron; its study measured iron freed from food, not iron absorbed by people
What it said
The page was titled "The Iron Rescue Hidden in Every Clove of Garlic", its search title was "Onion and Garlic Help Your Body Absorb 73% More Iron", and its heading said "How Onion and Garlic Help Your Body Absorb More Iron". The title also printed on the Shorts, recipes and collections that link to this page.
What it says now
It is titled "Garlic and Onion Go In for Flavor. Do They Free Up Iron Too?", and its search title and heading ask "Do Onion and Garlic Help Your Body Absorb More Iron?"
Why
The study measured how much of the iron in two cereals and two pulses came free in a form that can be absorbed (its bioaccessibility) when garlic or onion was added. In the pulses that rose by 9.9% to 73.3%. Its summary reports nothing measured in people, and calls a food-based way to help the body take up more of these minerals a "potential application". 73% is the top of a range, not what a body absorbs.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short said no human trial has tested or confirmed the effect; nothing on the page shows what other trials exist
What it said
The page said "No human absorption trial has confirmed it yet.", "The proof from inside a human body does not exist yet." and "the study nobody has run yet". Its search description said "No human absorption trial exists."
What it says now
It says "This is one lab study, and its summary reports what happened to the iron in the food", with nothing measured in people.
Why
The page rests on one study's summary. It can say what that summary reports, and that it reports nothing measured in people. It cannot show what no other study has done.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short said the effect survives cooking and runs separately from allicin; its study's summary does not say whether the garlic and onion were cooked, and never names allicin
What it said
The page said "Allicin is garlic's famous health compound" and "The iron mechanism is not allicin." Its audio said "Cooking does not destroy it. The garlic in your finished soup is still doing this." Its answer to "Does cooking garlic destroy its iron-absorption benefit?" began "No."
What it says now
It says "The team tested raw grains and cooked ones." and "What the summary does not say is whether the garlic and onion were cooked along with the grain or added afterward." Its answer to "Does cooking garlic take away its effect on iron?" begins "The 2010 study's summary cannot settle it."
Why
The summary says the cereals and pulses were studied raw and cooked, with garlic and onion added, and that the effect was generally seen in both. It does not say whether the garlic and onion were cooked with the grain. It never mentions allicin.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short named lentils, chickpeas, wheat and rice as the foods its study tested; the study's summary names none
What it said
The page said "In pulses (lentils and chickpeas, the foods where plant iron matters most)", and its audio said "When this was tested in a lab across lentils, chickpeas, wheat, and rice".
What it says now
It says "They added garlic or onion to two cereal grains and two pulses, raw and cooked" and "Its summary also does not name the four foods the team tested."
Why
The summary says "Two representative cereals and pulses each" and names no cereal and no pulse.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short called one study's result real, solid and reproducible in every food type; the study's summary says the effect was generally seen
What it said
The page said "The chemistry is real and reproducible across food types." Its audio said "The chemistry is solid and reproducible." and "It works in every food type that was tested." Its chart said "Confirmed in raw and cooked food".
What it says now
It says "The rise generally showed up in both." Its audio says "It is one study. And its own summary says the rise showed up generally, which is not the same as every time." Its chart says "From 9.9% to 73.3%, with the pulses raw and cooked".
Why
The page rests on one study, known from its summary, which says the effect "was generally evidenced". Nothing on the page shows the result repeated, or seen in every food at every level.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short told a lab result as what happens in your own pot, and its kitchen note gave it as an instruction
What it said
Its kitchen note said "Add garlic or onion to any pot of lentils, rice, or beans before cooking". The page said "every time garlic or onion goes into a pot of lentils, rice, or chickpeas, there is a grounded chemical case that more of the iron in those grains becomes available."
What it says now
It says "If what happened in the lab also happens in a pot, the garlic and onion you add for flavor could be doing a second job for the iron." and "That is a maybe, not a finding." Its kitchen note says "What that means for a home pot, the study's summary does not say."
Why
The study added fixed amounts of garlic or onion to two cereals and two pulses and measured the iron freed from them. Its summary says nothing about a home pot or a meal.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short said phytates lock iron away and explained how vitamin C works; nothing on the page supports either line
What it said
The page said "Grains and legumes are loaded with phytates, molecules that grab iron and lock it into clusters your body cannot use." and "Vitamin C works by changing iron's chemical form so your gut can absorb it."
What it says now
It says "Iron from plants is notoriously hard to absorb. The researchers start from that same point." The vitamin C lines are gone, and its link to the spinach page asks "Do the oxalates in spinach really block it?"
Why
The page's one study says only that iron and zinc are poorly taken up from plant foods. Its summary names no phytates and no vitamin C, and the page gave no other source for those lines.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short's recipe notes said two recipes match what the study tested; the recipes cook differently, and the study's summary names no food
What it said
Its recipe notes said "Garlic and onion are sautéed directly with the chickpeas — the exact allium-pulse pairing the study tested." and "Garlic and onion sautéed alongside kidney beans deploy the sulfur compound iron rescue".
What it says now
They say "This pilaf starts with sautéed onion and garlic, then simmers bulgur and chickpeas in the same pan." and "This casserole starts with sautéed onion and garlic and adds kidney beans for the last minutes of the simmer."
Why
Both recipes sauté the onion and garlic first and add the chickpeas or kidney beans near the end. The study's summary names no pulse, so no recipe can be called the pairing it tested, and it describes no rescue.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short's chart showed the top of a range as the result, and its search description gave the pulses' figure for grains
What it said
Its chart showed "73.3%" as "more iron freed from pulses" under the label "In the lab". Its search description said "keeps iron available in grains — up to 73.3% more in lab tests".
What it says now
The chart's label says "In the lab, up to" and its line below gives the range, "From 9.9% to 73.3%, with the pulses raw and cooked". The search description says "garlic or onion added to pulses raised the iron freed for absorption by 9.9% to 73.3%".
Why
In the study, the rise in pulses ran from 9.9% to 73.3%, and in cereals from 9.4% to 65.9%. 73.3% is the highest figure for pulses, not a typical one, and it is not the cereals' figure.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
November 22, 2026 · We got it wrong
A garlic and onion Short graded each of its claims as moderate evidence; it rests on one study of a food measure, known from its summary
What it said
Its evidence notes for AI systems graded each claim "moderate", for example "moderate — mechanism characterized in vitro".
What it says now
Its evidence notes for AI systems now grade each claim low, for example "low — one study of a food measure, known here from its abstract; it reports nothing measured in people".
Why
The page rests on one study, known from its summary. That summary reports a measure made on food, gives ranges and the word "generally", and describes no mechanism.
Gautam S, Platel K, Srinivasan K. Higher bioaccessibility of iron and zinc from food grains in the presence of garlic and onion. J Agric Food Chem. 2010;58(14):8426-9.Read the paper
Found by
Our full re-read of the page against all its papers.
A per-meal protein claim told AI systems its evidence covers muscle growth, which none of its studies measured
November 22, 2026 · We got it wrong
What it said
The tags FitChef gives AI systems and apps for the claim listed "muscle-growth" among the topics its evidence addresses.
What it says now
They now list protein intake, muscle protein synthesis, meal timing and protein dose.
Why
The studies behind the claim (Trommelen et al., 2023; Macnaughton et al., 2016; Mamerow et al., 2014) measured how fast muscles built protein over 5 to 24 hours, after a protein drink or through a day of meals. None of them measured muscle growth. This tag, made from the page's early notes, was not shown when the audit first read the page. When the page's tags were later read against the papers, it was kept by mistake.