~/labs local-first · official food tables · sources cited
Fat loss tracker: calories, body fat and sleep
Type “hamburger” and the tracker gives the calories of a fast-food-style hamburger at the usual French portion: about 465 kcal for 185 g (a small fast-food burger is about 90 g), from the French national food table (Anses Ciqual 2025) and the portions adults in France reported in the national INCA3 survey (2014-2015). Log meals, weight, body fat and sleep in a few seconds, morning, noon and evening. It runs in your browser, without an account.
Type a dish, get its calories
The search runs on your phone, with no AI and no server: it reads a list of about 300 everyday foods taken from official tables (Anses Ciqual 2025 for France, the Max Rubner-Institut's BLS 4.0 for German dishes, USDA FoodData Central for some unit weights). It understands French, English, German and Luxembourgish names, typos and quantities: “2 croissants”, “½ baguette”, “e Patt Wäin”. You can also tap an icon, or repeat a recent breakfast in one tap. Luxembourg dishes that no table lists by name show their nearest official equivalent or a plate estimate, labelled as such; the few with no official value yet are logged with calories only.
Body fat, three ways, never mixed
A tape measure once a week (US Navy formula; typical error about 4 points, sometimes 8 or more, a slow trend only; Foulis et al. 2023), the body-fat % your scale shows, typed at the morning weigh-in (often off by several kilos of fat; Frija-Masson et al. 2021), and an estimate from your BMI (Deurenberg et al. 1991) when you have neither. Each method keeps its own line: averaging them would hide their different errors.
A target you can check
Resting energy from the Mifflin-St Jeor equation (1990), times your daily-life activity level (FAO/WHO/UNU 2004), minus a deficit sized for 0.5 % of your weight a week by default (Helms et al. 2014; Garthe et al. 2011). The energy in a kilo lost depends on your body fat (Hall 2008), so the tool does not use the flat 7,700 kcal rule. Runs you log are added back on the day, and the deficit stops when your weight trend reaches your target. The tool never goes below 1,200 kcal a day (1,500 with the male formula), plans at least 30 kcal per kg of estimated lean mass after your logged runs (IOC 2023), and never cuts more than 750 kcal a day (AHA/ACC/TOS 2013). Both lean mass and running cost are estimates, so the warning signs listed in the tracker matter more than the figure.
Training for a race?
If you use Race Prep, the tracker shows your training phase and adds your logged runs to the day's target. In peak, taper and race week it switches to maintenance by default, so that hard training and a calorie deficit do not stack up into low energy availability (IOC REDs consensus, Mountjoy et al. 2023). Sleep sits next to food because, in a 14-day diet study with 10 people, short nights meant less fat and more lean mass lost (Nedeltcheva et al. 2010).
Your data
Everything you enter stays in this browser (IndexedDB and local storage). No account, no upload. Page views of this presentation page are counted anonymously; the tracker screen sends no audience measurement. Export a backup file whenever you like.
On your phone
Open the tracker, then add it to your home screen: on iPhone, Share then Add to Home Screen (install it before you start, since the home-screen app keeps its own data); on Android, the browser menu then Install app. It opens like an app. After the first visit it opens and works without a connection. Calendar reminders are made on your phone, from Me > Reminders.
Questions
- Is the calorie count exact?
- No count is: a hamburger varies by recipe (251 kcal per 100 g in Ciqual 2025, 190 in Germany's BLS 4.0, 288 for USDA's 'Hamburger, NFS'). The tool shows the source of every food, rounds to 5 kcal, and lets you change the grams. Studies with doubly labelled water find that people under-report what they eat, by 11 % on average with a structured recall (Moshfegh et al. 2008); the evening review asks about commonly forgotten foods for that reason.
- Which body-fat method should I trust?
- None is precise. The tape, measured the same way each week, is the cheapest way to follow a slow trend: its typical error is about 3.5 points (IOM 1998), but in Army recruits it read about 6 points lower than a DXA scan and missed short-term change (Foulis et al. 2023). Scale readings swing with meals and water (Slinde and Rossander-Hulthén 2001): weigh in at the same moment every morning and read the 7-reading line. The BMI estimate cannot tell muscle from fat (NICE NG246).
- Why 0.5 % of my weight a week by default?
- A review for natural bodybuilders recommends 0.5 to 1 % of body weight a week to keep muscle (Helms et al. 2014); in athletes, 0.7 % a week kept more lean mass than 1.4 % (Garthe et al. 2011). You can choose from 0.25 % to 1 %, never more: at most 0.75 % while training for a race, and 0.5 % while your body fat comes from the BMI estimate.
- Who should not use fat-loss mode?
- Anyone pregnant or breastfeeding, under 18, or with a current or past eating disorder: the tool then tracks without a deficit. If you take regular medicines or live with a long-term condition, talk to your doctor before eating less. This is a lifestyle and training tool, not a medical device.
- Does my data leave my phone?
- No. What you enter stays in your browser, and the tracker screen sends no audience measurement; only views of this presentation page are counted, anonymously.
Sources
- Mifflin et al. 1990 · Am J Clin Nutr 51(2):241-247
- Cunningham 1991 · Am J Clin Nutr 54:963-969
- FAO/WHO/UNU 2004 · Human energy requirements, chapter 5
- EFSA 2013 · Dietary Reference Values for energy
- Hall 2008 · Int J Obes 32:573-576
- Hall and Chow 2011 · Am J Clin Nutr 94:66-74
- Helms, Aragon and Fitschen 2014 · J Int Soc Sports Nutr 11:20
- Helms, Zinn, Rowlands and Brown 2014 · Int J Sport Nutr Exerc Metab 24:127-138
- Garthe et al. 2011 · Int J Sport Nutr Exerc Metab 21:97-104
- Aragon et al. 2017 · ISSN position stand: diets and body composition, J Int Soc Sports Nutr 14:16
- Jensen et al. 2013 · AHA/ACC/TOS guideline, Circulation 129(25 Suppl 2):S102-S138
- NICE NG246 · Overweight and obesity management (recommendations and Box 1)
- NICE NG247 · Recommendations
- Jäger et al. 2017 · ISSN position stand: protein and exercise, J Int Soc Sports Nutr 14:20
- KDIGO 2024 · CKD guideline, Kidney Int 105(4S):S117-S314
- Mountjoy et al. 2023 · IOC consensus statement on REDs, Br J Sports Med 57:1073-1097
- Loucks and Thuma 2003 · J Clin Endocrinol Metab 88:297-311
- Foulis et al. 2023 · Front Physiol
- Potter and Friedl 2025 · Mil Med
- IOM 1998 · Assessing Readiness in Military Women, Table 2-1
- Deurenberg, Weststrate and Seidell 1991 · Br J Nutr 65(2):105-114
- Kyle et al. 2004 · ESPEN, bioelectrical impedance analysis part II, Clin Nutr 23:1430-1453
- Frija-Masson et al. 2021 · JMIR mHealth uHealth 9:e22487
- Slinde and Rossander-Hulthén 2001 · Am J Clin Nutr 74:474-478
- Léger and Mercier 1984 · Sports Med 1:270-277
- Kaur et al. 2017 · Eur J Appl Physiol 117:225-235
- Engeroff et al. 2017 · J Sports Med Phys Fitness 57:1111-1118
- Docter et al. 2025 · Scand J Med Sci Sports 35:e70050
- Watson et al. 2015 · AASM and SRS consensus, Sleep 38(6):843-844
- Nedeltcheva et al. 2010 · Ann Intern Med 153(7):435-441
- Moshfegh et al. 2008 · Am J Clin Nutr
- Madigan et al. 2015 · IJBNPA
- Levinson et al. 2017 · Eat Behav 27:14-16
- FFAB · Anorexie Boulimie Info écoute (permanence téléphonique)
- WHO Global Health Observatory · indicator 4804, adult underweight (BMI < 18.5)
- Regulation (EU) 2017/745 (MDR) · recital 19
- Regulation (EU) 2016/679 (GDPR) · art. 9
Food data and icons
Food values: Anses (2025), Ciqual 2025 food composition table, files of 3 Nov 2025, doi:10.57745/RDMHWY, Licence Ouverte 2.0; Max Rubner-Institut (2025), Bundeslebensmittelschlüssel (BLS) 4.0, doi:10.25826/Data20251217-134202-0, CC BY 4.0; U.S. Department of Agriculture, Agricultural Research Service, FoodData Central (SR Legacy 2018, FNDDS 2021-2023), public domain (CC0 1.0). Usual portions: Anses, INCA3 survey 2014-2015 (Dubuisson et al. 2019, doi:10.1017/S1368980018002896; data.gouv.fr files updated 28 Jan 2021), adult medians computed by Sitraka Labs. Luxembourgish names: Zenter fir d'Lëtzebuerger Sprooch, LOD (CC0). Values selected, rounded and combined by Sitraka Labs; foods marked plate estimate add up the foods listed on them. These bodies do not endorse this tool.
- Anses, Ciqual 2025 food composition table · Licence Ouverte 2.0
- Max Rubner-Institut, Bundeslebensmittelschlüssel (BLS) 4.0 · CC BY 4.0
- U.S. Department of Agriculture, Agricultural Research Service, FoodData Central · CC0 1.0
- Anses, INCA3 survey data (data.gouv.fr) · CRPA L322-1
- Dubuisson et al. 2019, Public Health Nutrition 22(4):584-600 (INCA3) · INCA3 user notice, 2.3
- Zenter fir d'Lëtzebuerger Sprooch, Lëtzebuerger Online Dictionnaire (LOD) · CC0 1.0
Icons: Lucide, ISC licence.
A lifestyle and training information tool, not a medical device. Estimates (body fat, energy use, calories) carry an error of several points and do not replace advice from a doctor or dietitian. Do not use fat-loss mode if you are pregnant or breastfeeding, under 18, or have (or have had) an eating disorder. If you take regular medicines or have a long-term condition (diabetes, kidney or heart disease), talk to your doctor before eating less. The tool never suggests less than 1,200 kcal a day.