FitChef

The log: one day.

June 3, 2026

5 corrections on 8 pages. Most arrive together: one paper read again against every page that cites it, and every page fixed the same day.

The carbs guide's headline participant total was borrowed from one claim instead of computed from the guide's own studies

June 3, 2026 · We got it wrong

What it said
The guide's headline participant total was taken from a single claim's evidence set, most of it one 68,128-participant study, rather than from the studies the guide itself analyses.
What it says now
The headline total is computed from the guide's own study set, and a verification step blocks any headline stat that cannot be reconstructed from the studies shown on the page.
Why
Internal audit found the borrowed figure made the headline unrepresentative of the guide as a whole.
Built on it

3 shorts · 2 recipe hubs

Found by
Our own checks of the site.
A recipe carried a 'High Protein' badge below the EU threshold for that claim

June 3, 2026 · We got it wrong

What it said
The stuffed portobello mushrooms recipe displayed a 'High Protein' badge. Protein provides 16.7% of its energy, below the 20% that the EU nutrition-claim regulation (1924/2006) requires for 'high protein'.
What it says now
The badge reads '35g Protein', a factual statement rather than a regulated claim. The recipe pipeline now checks the threshold before assigning any nutrition badge.
Why
Found by the deep pre-scale audit's EU health-claim compliance check.
Found by
Our own checks of the site.
A meta-analysis page stated a participant total (684) that its paper does not report

June 3, 2026 · We got it wrong

What it said
The Schoenfeld 2017 light-versus-heavy-load meta-analysis page stated '684 participants' across its 21 studies, in the hero, the introduction, the search description and the citation fields. The training guide's headline participant total (9,350) and a cross-reference in the Zhang 2025 exercise-and-deficit study used the same figure.
What it says now
The participant count is removed and the page keeps '21 studies', which is what the paper reports. The training guide's headline total is 8,666, and the Zhang 2025 cross-reference no longer carries the figure.
Why
Checked against the full paper (Journal of Strength and Conditioning Research 31(12):3508-3523): it reports no pooled participant total. Its Table 1 sums to 630, but that includes control groups excluded from the analysis, so the analysed number is lower and unreported; 684 appears nowhere in the paper. An unverifiable number is treated the same as a fabricated one and removed rather than replaced with an estimate. The figure had been written as a clean number in the correction of 2026-05-25.
The paper
Schoenfeld BJ et al. (2017). Journal of Strength and Conditioning Research 31(12):3508-3523
Found by
Our own checks of the site.
A sugar meta-analysis page, and the claim citing it, showed the number of pooled studies as a participant count

June 3, 2026 · We got it wrong

What it said
The sugar-and-body-weight meta-analysis page, and the claim that cites it, showed 'Participants: 68'. Sixty-eight is the number of studies the meta-analysis pooled (30 randomised trials and 38 prospective cohort studies), not a participant count.
What it says now
The figure is removed from both pages. The paper reports per-analysis effect sizes rather than one participant total, so no number replaces it.
Why
Internal audit traced the value to the study count. Rather than invent a total, we removed it.
Built on it

1 guide · 9 shorts · 11 recipes · 12 recipe hubs

The paper
Te Morenga et al. (2013). BMJ
Found by
Our own checks of the site.
A study headline called a 20-minute intake figure '340 extra calories', implying a between-diet difference that is actually 508 a day

June 3, 2026 · We got it wrong

What it said
The headline on the Hall 2019 ultra-processed-food study read '340 extra calories'. The 340 figure is how many calories the measured eating rate delivers in a 20-minute meal; 'extra' implied it was the difference between the two diets, which is 508 calories a day.
What it says now
The headline reads '340 calories', without 'extra'. The 508-calorie daily difference stays in the key finding, where it was already stated correctly.
Why
Internal audit found the word turned a grounded number into a misleading one.
Built on it

1 claim · 1 guide · 9 shorts · 2 recipe hubs

Found by
Our own checks of the site.

Newer day: June 5, 2026Older day: June 2, 2026The whole log

The full picture

How every claim on FitChef is checked, page by page.

We did the thinking. You just eat.

Get my week

It takes a minute.

ListenListen0:00Help