How Pregnancy Due Dates Are Calculated
A due date is one date produced by a short piece of calendar arithmetic. This guide takes that arithmetic apart: where the 280 days in Naegele's rule come from, why dating from a conception date uses 266 days instead, what "gestational age" is actually counting, how a cycle length other than 28 days shifts every date in the estimate, and why the estimate is routinely revised — clinicians commonly revise dating after an ultrasound measurement. It is written for anyone who has been handed a date — by a clinic, an app, or the due date calculator on this site — and wants to know how the number was produced.
Naegele's rule is addition
The classical obstetric convention takes the first day of the last menstrual period — the LMP — and adds 280 days. That is the whole calculation. 280 days is exactly 40 seven-day weeks — the week count and the day count are the same measurement expressed two ways.
Worked through with the site's engine: an LMP of February 9, 2026 and the default 28-day cycle give an estimated due date of November 16, 2026. Nothing else was supplied, because nothing else is needed. The rule needs one input and one addition, which is why it can be reproduced by anyone with a calendar — it does not model anything about a particular pregnancy.
The output is a single fixed reference date. Every other figure the calculator reports — the dating start, the trimester boundaries, the gestational age on a given day — is measured from it.
Why 280 days from the LMP but 266 from conception
The calculator also accepts a known conception date, and in that mode it adds 266 days rather than 280. The difference is exactly 14 days, and it is not a competing estimate — it is the same estimate measured from a starting line 14 days further along.
The reason is a convention about counting. Gestational age is counted from the LMP, not from conception. Because the 28-day model places ovulation near cycle day 14, a conception date implies an LMP 14 days earlier, and the engine sets day zero there. So 280 − 14 = 266.
The two modes agree when the inputs agree. A conception date of February 23, 2026 is 14 days after the LMP used above, and the conception-mode estimate returns November 16, 2026 — the same date, from the same day zero. Two different-looking rules, one piece of arithmetic.
This is also the cleanest way to see why gestational age runs ahead of the age of the embryo itself. The count starts roughly two weeks before conception is assumed to have happened, so the number of weeks reported is roughly two weeks larger than the time since conception. The 280-versus-266 gap is that statement written as arithmetic. It is a labeling convention, consistently applied, and not a claim that anything begins at the LMP.
The 28-day assumption, and what a different cycle does
Naegele's rule has one hidden input: it assumes a 28-day cycle with ovulation near day 14. The calculator's optional cycle-length field carries that assumption forward rather than replacing it — the model places ovulation one luteal phase, 14 days, before the next expected period — so entering a longer cycle moves day zero later by (cycle length − 28) days, and a shorter one moves it earlier by the same amount. That one offset is applied to both the dating start and the due date.
Same LMP of February 9, 2026, three cycle lengths, three due dates from the same engine:
- 28-day cycle (the default): November 16, 2026
- 32-day cycle: November 20, 2026 — 4 days later
- 24-day cycle: November 12, 2026 — 4 days earlier
The shift is linear and symmetric because the adjustment is a single addition. That is worth noticing for its own sake: the cycle-length field does not make the estimate cleverer, it moves the whole estimate sideways. Everything downstream — the dating start, the trimester boundaries, the gestational age reported on any given day — moves with it by the same number of days.
The tool accepts whole-number cycle lengths from 20 to 45 days and returns an explanation instead of a number outside that range, because beyond it the offset stops describing anything the rule can represent.
What the adjustment cannot do
It cannot help with cycles that vary. The field takes one length and applies one offset; a cycle that runs short one month and long the next has no single length to enter, and an average of several is not a fact about the particular cycle in question. It also cannot repair an uncertain LMP. Every date downstream is measured from that first date, so an LMP recalled as "sometime in the middle of the month" carries its uncertainty into the due date one day for one day. If cycle arithmetic in general is the part you want to understand, the companion guide on the fertile window and cycle estimates covers the same 14-day convention as it appears in the ovulation calculator.
Gestational age and the trimester boundaries this tool implements
Once day zero is fixed, gestational age is just the number of days elapsed since it, reported as whole weeks plus leftover days. On June 4, 2026, the 28-day example reads 16 weeks 3 days, and the 32-day example reads 15 weeks 6 days on the very same calendar day. Neither is wrong. They are the same date measured from two different day zeros, 4 days apart, and the gap never closes.
Trimester boundaries are drawn from that same day zero. This site implements the boundaries as 13w0d for the second trimester and 27w0d for the third — 91 and 189 days respectively. For the default example those fall on May 11, 2026 and August 17, 2026, which is why the estimate on June 4, 2026 is reported as trimester 2. Other week-based schemes are in use, so these are stated as the convention the tool implements rather than as a boundary in nature. If you see a different trimester on a different site, check which convention it labels before assuming either is broken.
Two edge cases are worth knowing because the tool handles them by refusing rather than inventing. A reference date before day zero produces no gestational age at all — there is no elapsed time to report. And beyond 44 weeks the engine stops reporting a running age — that constant is the reportable ceiling the tool implements, past which it returns no gestational age rather than an implausible one.
Why the estimate gets revised
A due date is an estimate throughout. Clinicians routinely revise dating, commonly after an ultrasound measurement, and a date supplied by your care team supersedes anything this or any other calendar rule produces. The two methods start from different information: the calendar rule works from a remembered date and an assumed cycle, and a clinical measurement works from something else entirely.
Which is the honest summary of what a due-date calculator is for: it reproduces a documented convention transparently, so the assumptions behind the number are visible. It is not prenatal care, it cannot confirm a pregnancy or say anything about how one is progressing, and it does not replace a conversation with a qualified healthcare provider.
Four common misreadings
Reading the due date as a forecast of the birthday
The rule adds a constant to a date. It observes nothing, so it anticipates nothing. Its use is as a shared reference point for scheduling.
Reading gestational weeks as the age of the embryo
Gestational age counts from the LMP by convention, roughly two weeks before conception is assumed to occur. The 280-versus-266 day difference is the entire explanation, and it applies to every gestational-age figure you will encounter.
Expecting the cycle-length field to handle irregularity
One field, one offset, applied uniformly. It corrects a consistent deviation from 28 days and does nothing else — including nothing at all for cycles whose length is the thing that varies.
Treating two different due dates as proof one is wrong
Two dates can differ because the inputs differ, because the cycle-length assumption differs, or because one came from a clinical measurement and one from a calendar. Before comparing any two due dates, it is worth asking what day zero each one used.
A note on the site's other calculators
BMI is documented as not meaningful during pregnancy, a limit the guide on what BMI does and does not measure sets out in full. The other calculators on this site estimate body composition and energy needs from formulas built for general adult populations, and each is documented with its own limitations rather than adapted to pregnancy. Every number printed above was produced at build time by the same tested engine that runs behind the due date calculator, from the fixed dates quoted in the text.
Frequently asked questions
What is Naegele's rule, exactly?
Add 280 days to the first day of the last menstrual period. That is the entire calculation — no other input is required and no other input is used. The rule assumes a 28-day cycle with ovulation near day 14, which is why the calculator on this site offers a cycle-length field that shifts the result by (your cycle length − 28) days.
Why does the conception-date mode add 266 days instead of 280?
Because the two starting points sit 14 days apart by convention. Gestational age is counted from the last menstrual period rather than from conception, so dating from a conception date drops the assumed 14 days between them: 280 − 14 = 266. Enter a conception date exactly 14 days after a last menstrual period and both routes land on the same due date.
Does the cycle-length field help if my cycles are irregular?
Not meaningfully. The adjustment applies one fixed offset — (cycle length − 28) days — to the whole estimate, which is a sensible correction for a cycle that is consistently longer or shorter than 28 days. When the length varies from month to month there is no single number that describes it, so there is nothing for the arithmetic to work with. The tool accepts whole numbers from 20 to 45 days and returns an explanation rather than a result outside that range.
Why is the date my clinic uses different from this one?
Different inputs and different methods produce different dates. A due date is an estimate, and clinicians routinely revise dating, commonly after an ultrasound measurement. The date your care team is working from supersedes anything computed from a calendar here. A calculator disagreeing with a clinic is the expected case, not a malfunction.
Which trimester boundaries does this site use?
Second trimester from 13w0d and third from 27w0d, counted from the dating start rather than from the calendar month. Other week-based schemes are in circulation, so the pages label the convention they implement instead of presenting it as a fact of biology.
What happens to the dates I type in?
Nothing leaves your device. Every calculation on this site runs in your browser, and no input is transmitted or stored. The worked examples printed on this page are computed at build time by the same engine the calculator uses, so the prose and the code cannot drift apart.
Not medical advice: this guide provides general educational information about published dating conventions and the arithmetic behind them. A due date is an estimate; clinicians routinely revise dating, commonly after an ultrasound measurement. This page cannot account for your individual health context, must not be used to diagnose or treat any condition, and does not replace prenatal care — consult a qualified healthcare professional. Dates entered into the calculators are processed locally in your browser and never transmitted. See the methodology page.