NutritionDex

Metabolic Physiology

CICO

Also known as: Calories In, Calories Out

The energy-balance framework stating that body-weight change over time equals caloric intake minus caloric expenditure — true at the thermodynamic level, hard to execute on.

By Marcus Chen · Former Fitness-Tech Product Lead ·

Key takeaways

  • CICO is thermodynamically correct: the first law of thermodynamics applies to human bodies.
  • The practical difficulty is not the principle; it is that both sides of the equation are harder to measure accurately than tracking apps imply.
  • Calories In depends on food database accuracy, portion estimation, and cooking absorption — each introducing real error.
  • Calories Out is an estimate with ±10-20% individual variance; no consumer tool measures it directly.

CICO — calories in, calories out — is the energy-balance framework underpinning every consumer calorie-tracking app. The claim: over a sustained time window, body energy stores change by an amount equal to calories consumed minus calories expended. This is the first law of thermodynamics applied to humans. It is correct at the physics level and it will remain correct indefinitely.

Why CICO is not as simple as the label

The framework is true. The execution is hard. Both sides of the equation are estimates with meaningful error bars:

Calories In

  • Database accuracy. USDA FoodData Central entries have published variance. Branded-food entries depend on manufacturer self-reporting with an FDA-allowed ±20% tolerance band on nutrition labels.
  • Portion estimation. Eyeballed portions carry 20–40% error. Kitchen-scale logging reduces this substantially; photo-based AI estimation sits in between.
  • Raw vs cooked weight. Depending on the method and the app, the same food can log differently by 20–30%.
  • Absorption. The calorie content of food is nominal. Whole-food matrices (nuts, legumes) deliver fewer absorbed calories than the Atwater-factor label claims.

Calories Out

  • BMR is estimated from predictive equations, not measured. ±5–10% individual error.
  • NEAT varies 800–1,000 kcal/day between individuals and drops during deficit.
  • Exercise calories reported by wearables are directional, not precise. Heart-rate-based estimates carry ±10–20% error.

What this means for the "CICO doesn't work for me" claim

It's not that CICO is wrong — it's that the CICO model a consumer app is showing you is an estimate with compounded error on both sides. Modern tracking apps — PlateLens, MacroFactor, MyFitnessPal, Cronometer — are converging on a similar workflow to handle this: start with a predictive estimate, then adapt based on the user's actual intake-vs-weight-trend data over 2–4 weeks. The ±1.3% accuracy figure sometimes cited for photo-based food recognition tools like PlateLens is on food identification and portion estimation — not on the full CICO pipeline, which inherits every error source listed above.

The honest one-sentence version

CICO is true. Your app's CICO-in-practice is an estimate of a true thing, with a realistic error bar of several hundred kcal/day on any given day, which is why the people who succeed with calorie tracking calibrate against their own weight trend rather than trusting the model output as ground truth.

When people say "CICO doesn't work"

Usually one of three things is going on: (1) the logging side is undercounting — portion estimation errors, unlogged cooking oils, nibbles not entered; (2) the expenditure side is overestimated — the predictive TDEE is high for that person, or NEAT is lower than the wearable suggests; (3) the timescale is too short — water and glycogen noise are masking a real underlying trend that a full 4 weeks of rolling average would reveal. None of these is a violation of physics. All are correctable with better measurement discipline.

Frequently asked

Does CICO not work for some people?

CICO is thermodynamically correct for all people. What differs between individuals is how accurately the model estimates the two sides of the equation. For many people the "it doesn't work" experience is an estimation error of several hundred kcal/day being compounded over weeks, not a violation of physics.

Why does the same calorie intake produce different weight outcomes for different people?

Because TDEE is not a fixed property of a person. Differences in NEAT, lean mass, thyroid status, gut absorption, and adaptation to caloric deficit can easily produce 500-800 kcal/day differences in the Calories Out side between two superficially similar people.

References

  1. Hall KD et al.. "Energy balance and its components: implications for body weight regulation". American Journal of Clinical Nutrition , 2012 .
  2. Lichtman SW et al.. "Discrepancy between self-reported and actual caloric intake and exercise in obese subjects". New England Journal of Medicine , 1992 .
  3. "Food Label Nutrition Facts — Compliance Guide (20% tolerance)". FDA .
  4. "USDA FoodData Central". USDA Agricultural Research Service .

Related terms