← Nutrient Scoring Models

NUTRITION ANALYSIS

Custom model

The Custom Exponential Model is designed for practical food comparison, athletic performance, and incomplete Cronometer entries. It uses a soft cap: additional intake continues to improve the result, but with diminishing returns.

Custom Exponential Model

The model first converts every nutrient to an amount per 100 kcal. This makes a calorie-dense and a calorie-sparse food comparable. For each nutrient, it divides that amount by the nutrient's 2,000-kcal daily-value reference, then applies the soft cap:

Value = 1 − e⁻ˣ

Here, x is the amount per 100 kcal divided by the reference amount per 100 kcal. A nutrient at its reference gives about 63.2% of its possible contribution; more intake keeps helping but never adds more than 1. Positive nutrients are averaged, limited nutrients are averaged separately, and the final raw score is:

Raw score = 50 + 50 × positive average − 50 × limited average

The result is rounded to one decimal place and clamped to 1-100. Missing or non-numeric nutrient fields are left out of that group's average; the remaining known nutrients are reweighted to the full group. This is why incomplete data does not automatically reduce the score, while data_coverage_percent tells you how much evidence the result contains. The low-energy fallback returns 50 when energy is unavailable or below the safety threshold.

The custom model's full positive set is Protein, Fiber, Omega-3, Monounsaturated Fat, Vitamins A, C, D, E and K, Thiamine (Vitamin B1), Riboflavin (Vitamin B2), Niacin (Vitamin B3), Pantothenic Acid (Vitamin B5), Vitamin B6, Folate, Vitamin B12, Choline, Calcium, Copper, Iron, Magnesium, Manganese, Phosphorus, Potassium, Selenium, Iodine and Zinc. The nine essential amino acids are also positive, but each has half weight: Histidine, Isoleucine, Leucine, Lysine, Methionine, Phenylalanine, Threonine, Tryptophan and Valine. Limited nutrients are Added Sugars, Saturated Fat, Trans Fat, Alcohol, Cholesterol, Oxalate, Phytate and Sugar Alcohol.

Data completeness and reliability

Every result includes data_coverage_percent. The denominator is the number of nutrients used by the selected model.

  • Missing, null, or untracked Cronometer values are unavailable.
  • A measured value of 0 is known and counts toward coverage.
  • Lower coverage means the score is less reliable for comparison.
  • Coverage is model-specific: the Custom model uses its own set of positive and limited nutrients.

Always interpret a score together with its data coverage.

Open nutrient scoring

Read about the Nutrient Rich Food Index (NRF)

Related guides

Continue with these connected Fuel Lens guides.