An average is only as useful as your understanding of what went into it. In a food record, a blank day means that food is missing from the record. It is not evidence that no food was eaten. A partly logged day can be harder to spot because it already has a calorie total.
Know which days the average includes
In CORPUS History, the calorie Daily Avg uses days with positive logged intake. The calorie-balance Daily Avg uses days with positive intake and an eligible burn calculation. Burn, steps and sleep have their own coverage. None of these counts should be assumed to represent every date in the selected week.
The following values are invented to demonstrate the arithmetic. They are not dietary advice or a suggested intake.
A seven-day period with four food records
Record | Result |
|---|---|
Four logged days | 1,800 + 2,000 + 2,100 + 1,900 = 7,800 kcal |
Three blank days | Intake unknown |
Average across the four logged days | 7,800 ÷ 4 = 1,950 kcal |
Dividing by all seven dates | About 1,114 kcal; this treats missing records as zero and does not describe known intake |
If all four food days have eligible burn, their balance average covers those four days. The three blank days do not tell you whether the whole week was a deficit, surplus or maintenance. If only three of the four food days have eligible burn, the balance coverage is three of four calorie days—not three of seven calendar days.
A partial day can still enter the calculation
Suppose a day’s estimated Total Burn is 2,400 kcal, but only 1,000 kcal of food is logged. The record implies a 1,400 kcal deficit. If a further 1,200 kcal of food was eaten but omitted, adding it changes the recorded deficit to 200 kcal. The burn estimate did not change; the food side became more complete.
CORPUS cannot infer completeness from a positive calorie total. A balance label describes the entries available to its calculation. It does not certify that every meal, drink or ingredient has been captured.
Read the coverage note before comparing weeks
A balance coverage note counts eligible balance days against positive-calorie days. An asterisk can flag positive-calorie days without eligible balance, but blank food days alone do not necessarily trigger that marker. Check the calendar and individual days as well as the summary.
Different metrics can cover different dates. Avoid subtracting the displayed weekly calorie average from the displayed burn average to check the balance average: those two averages may not share a denominator. Read the balance calculation and its own coverage instead.
Missing Health records also need attention. A displayed zero activity value is not, by itself, proof that no activity occurred. Review readable Health data and permissions before interpreting an unexpected result. An unavailable daily burn is different from a confirmed zero-activity day.
Keep a missing record visibly missing
Review the week’s dates. Identify blank days, partial days and any future-dated plans.
Add food you can account for, using the best source and amount available. Do not invent precise entries to make the calendar look complete.
Keep uncertainty in mind when comparing periods with different coverage. Four well-recorded days and seven partly recorded days answer different questions.
Use reusable Recipes, Meal Templates and Day Templates to reduce repeated entry, while checking that the saved food and amounts match this meal.
Future-dated food plans can enter a selected period’s calorie totals and averages, while future days are excluded from balance. Separate plans from completed records before calling a figure “this week’s intake.”
A useful weekly review can end with a narrower conclusion: “These four recorded days averaged 1,950 kcal.” That is more informative than treating an incomplete week as a complete account of what happened.
Sources & further reading
CORPUS — App Store product description and release notes. Accessed 2026-09-30.
Apple: Manage Health data on your iPhone, iPad or Apple Watch. Accessed 2026-09-30.
Drafted with AI assistance and checked in a separate automated editorial review. This article has not received human or medical review. Our editorial standards.