Where the range comes from.
Where the range comes from.
A Likely range is a central estimate built from food composition, portion uncertainty, and the details you actually know. It is where the value most likely lands — not a fence around every possibility and not a guarantee.
Start with measured food data.
Likely's food records carry a typical value and natural variation for each nutrient. A generic food covers the common versions that match what you entered; a branded record covers one specific product.
Include what the portion could be.
An exact weight is narrow. “A handful,” “one bowl,” or “some chicken” carries honest portion uncertainty, including a personal hand calibration when you have set one.
Tighten only with real information.
Preparation, cut, brand, weight basis, and other supported details remove possibilities the estimate no longer needs to cover. Missing details keep the band wider; they never make the log invalid.
Uncertainty combines; it does not simply stack.
For a meal or day, Likely combines the central estimates and their uncertainty as distributions. It does not naively add every low end to every other low end and every high end to every other high end, which would make ordinary multi-item days look implausibly uncertain.
An estimate remains an estimate.
Food data can be incomplete, recipes and portions vary, and nobody measured your exact plate. Likely exposes that uncertainty and turns it into plain guidance; it does not claim laboratory precision or provide medical advice.
Questions about the method?
Ask plainly. A human reads it.
Email Likely