How custody time is counted

Looking for the public explainer? See How courts count parenting time.

The single most common question we get is: "Why does my count look one day short?" It isn't — CustodyProof counts custody time the way family courts commonly do: by overnights. This article explains the math in plain language.

Overnights, not days

When you log a stay from one date to another, CustodyProof counts the nights your child slept at your home — from the start date up to, but not including, the end date. The end date is the morning the stay ends (the return or exchange day), so it isn't counted as a night.

Example A. The children are with you from January 1 to January 11.

They sleep at your home the nights of Jan 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10.
On Jan 11 they go back. No night on the 11th.
10 nights, even though 11 calendar dates were touched.

Example B. Friday after school to Sunday evening.

Friday night: yes.
Saturday night: yes.
Sunday night: no — they are not sleeping at your home.
2 nights, not a "3-day weekend."

If you count "days" you will always come out one higher than a night count on a stay that ends on the return morning. People then think the log is "one short." It is using a different unit.

This is intentional and matches the convention many family courts use when they talk about parenting time, so the numbers in your reports line up. How your order counts time is a question for your attorney.

Timed visits are counted separately

A timed visit (same-day parenting time with start and end times, like Wednesday 3:30–7:30 pm) involves no overnight — so it never enters your overnight counts or percentages. Instead, visits are tracked on their own as:

a count of visits, and
total hours of visit time.

You'll see these separately on the Dashboard ("Visits YTD") and in the Statistics Timed Visits section. Keeping them separate means your overnight percentage stays clean and defensible, while your day-time parenting still shows up in your record.

One night is one night — no matter how many children

If you log the same night for two or three children, your statistics still count that calendar night once. Custody percentages measure your time, not a per-child tally, so nights are never double-counted.

Per-child detail doesn't disappear, though:

The By Child tab in Statistics shows each child's own nights and percentage.
PDF reports can additionally show child-nights — a clearly-labeled cumulative figure where 3 children × 1 night = 3 child-nights (but still 1 night). See Reports and exports.

How percentages work

Your custody percentage is:

unique custody nights in the period ÷ total days in the period

The period is whatever you've selected (Year to Date, This Month, a custom range, and so on).
Percentages are capped at 100% — they can never read higher.
Timed visits are not part of this number.

Why don't my percentage and the other parent's add up to 100%? CustodyProof only knows what you record. Days with nothing logged (or logged only as visits) simply aren't your custody nights — they aren't automatically assigned to anyone else.

A single-day entry

An untimed entry where the start and end date are the same day counts as 1 day. If the parenting time had specific hours, log it as a timed visit instead so the hours are captured.

Dashboard and Statistics always match

The Dashboard and the Statistics page use the same counting engine and the same rules above, so for the same period they always show the same numbers. If you ever see a mismatch, refresh the page — and if it persists, contact Support with both numbers and the period you were viewing.

Why this matters

Getting custody math wrong has real consequences in court. CustodyProof applies these rules identically everywhere — the calendar, the Dashboard, Statistics, and every report — so a number you quote from one screen will match the report you hand your attorney.

Reminder: CustodyProof documents your time; it doesn't interpret your custody order. How your jurisdiction counts custody for legal purposes is a question for your attorney.