Quick answer

A due date is a midpoint. Most births happen within several weeks either side of it, and ultrasound dating supersedes the LMP calculation when they disagree.

The 280-day model

The conventional estimated due date is the first day of the last menstrual period plus 280 days — forty weeks. Some versions adjust for cycle length, adding or subtracting the difference from 28 days, on the reasoning that a longer cycle means later ovulation.

The model embeds two assumptions: a 28-day cycle and ovulation on day 14. Both are averages rather than facts about any individual pregnancy, which is why the resulting date is an estimate produced by a convention, not a measurement.

An ovulation calculator makes that day-14 assumption explicit: instead of assuming a 28-day cycle, it estimates ovulation from your own cycle length — the same calendar method, dated from your last period rather than from an average.

Use this guide with the right tool: Open the Pregnancy Due Date Calculator for an estimated due date from the last menstrual period and cycle length. If the question shifts, compare it with the Conception Date Calculator or the Pregnancy Week Calculator.

An expectant mother holding a blank calendar
The 280-day model gives an estimate, not a promise — a due date is a midpoint, not a deadline.

What it does and does not predict

The due date marks the centre of a distribution. Births between 37 and 42 weeks are all within the normal range, and only a small minority occur on the estimated date itself. Presented as a single date, it invites a precision the underlying biology does not have.

This matters for planning rather than for care. Arrangements built to start or end exactly on the due date have no tolerance in either direction, and the direction that causes problems — arriving early — is the one people plan for least.

When the date changes

A dating scan, usually in the first trimester, measures the embryo directly. Where that measurement disagrees with the LMP-based date by more than a defined tolerance, the scan date generally supersedes it, because a direct measurement beats an inference from a cycle assumption.

When the due date is revised, every derived figure moves with it: gestational age, days remaining, the trimester position and any milestone dates already noted. Recalculate the whole set from the new date rather than adjusting individual numbers, or the record ends up internally inconsistent with no signal which entry is stale.

Using the calculator honestly

The calculator applies the standard model exactly and shows the assumptions it used, which makes it useful for understanding how a date was derived, checking that two figures you have been given are consistent, or seeing the effect of a cycle-length adjustment.

It does not date a pregnancy. That is done by the midwife or obstetrician managing it, using clinical assessment and ultrasound. Where the calculator and the notes disagree, the notes are right and the difference is worth asking about rather than resolving yourself.

Worked example

An LMP of 21 May 2026 with a 28-day cycle gives an estimated due date of 25 February 2027 — 280 days later. With a 32-day cycle the adjusted estimate moves to 1 March. A first-trimester dating scan may then revise it again, and that revised date is the one care runs on.

Try your own LMP and cycle length, remembering a dating scan can revise it.

The five weeks around the date

The worked example puts the estimated due date for a last menstrual period of 21 May 2026 at 25 February 2027. That single date is easier to use honestly once the window around it has dates of its own, because the window — not the midpoint — is what planning actually has to fit inside.

Counting from the same LMP, the pregnancy reaches 37 weeks 0 days on 4 February 2027, twenty-one days before the due date. Births from that point on are counted as term. At the far end, 42 weeks 0 days falls on 11 March 2027, fourteen days after the due date. Between those two dates lies a span of 35 days in which an arrival is entirely unremarkable, and the estimated due date sits inside it — closer to the early end than the late one, because the model places three of the window's five weeks before the midpoint and two after it.

The same LMP gives the model's other implied dates. Conception is estimated at 4 June 2026, fourteen days after the LMP — not a recorded event, just the assumption the 280-day model is built on. Halfway through, on 8 October 2026, the pregnancy is dated at exactly 20 weeks 0 days, 140 days from the LMP. Every one of these figures is the same calculation seen from a different point: fix the LMP and the cycle assumption, and the conception estimate, the week count on any chosen day, the term window and the due date all follow together.

Which is exactly why a dating scan revision moves all of them at once. Suppose the first-trimester scan re-dates this pregnancy six days earlier, putting the due date at 19 February 2027. The implied LMP shifts back to 15 May 2026. The week count on 8 October 2026 becomes 20 weeks 6 days rather than 20 weeks 0 days. The term window slides with everything else, running from 29 January to 5 March 2027. No figure survives the revision unchanged, because none of them was measured independently — each was derived from the same starting assumption.

That is the practical content of the advice given earlier in this guide: when the date changes, recalculate the whole set. A record that keeps the old term window beside the new due date, or the old week count beside the new LMP, contains two pregnancies' worth of arithmetic and no indication of which set the care is actually running on. The revision is a substitution, not an adjustment — the new dates replace the old ones everywhere, or the inconsistency the revision was meant to remove is preserved in a new form.

Common mistakes to avoid

  • Entering the last day of the period. The model counts from the first day of the last menstrual period. Using the last day shifts the date by several days.
  • Treating the due date as a deadline. It is the centre of a range spanning several weeks. Build plans with tolerance in both directions.
  • Keeping the LMP date after a scan revision. When dating is revised, recalculate everything from the new date rather than adjusting figures individually.

Use the right calculator

Use the Pregnancy Due Date Calculator from LMP and cycle length, the Conception Date Calculator for a conception window, and the Pregnancy Week Calculator for week and day.

Pregnancy Due Date Calculator
Pregnancy & PediatricPregnancy Due Date Calculatoran estimated due date from the last menstrual period and cycle length
Conception Date Calculator
Pregnancy & PediatricConception Date Calculatoran estimated conception window from LMP or due date
Pregnancy Week Calculator
Pregnancy & PediatricPregnancy Week Calculatorpregnancy week and day on a selected date
Private to this page: the LMP and cycle dates you enter are calculated locally in your browser. No account is required, and no dates are sent to a server for your due date estimate.

Frequently asked questions

How many babies arrive on the due date?

Only a small proportion. The date marks the centre of a distribution spanning several weeks, not a scheduled event.

Should I adjust for my cycle length?

The calculator can, but clinical practice usually prefers to re-date from ultrasound, which is more reliable than adjusting the assumption.

Which date should I use if two disagree?

The one your clinical team is using. There are established criteria for when scan dating supersedes LMP dating.