The dating rule behind every calculator on this site, where each number comes from, and what happens when a scan disagrees with the math.
A due date is arithmetic wearing a clinical costume unless the rule behind it is a real, published one. This page names the rule each calculator runs, points to the primary source behind every figure that is not ours to invent, and says plainly where the arithmetic stops and your provider's judgment starts.
The Due Date Calculator, Pregnancy Calculator, Trimester Calculator and Gestational Age Calculator all start from Naegele's rule: take the first day of the last menstrual period, add one year, subtract three months, add seven days. In practice that is the same as adding 280 days. The rule was described by German obstetrician Franz Naegele in the 19th century, and it assumes a 28-day cycle with ovulation on day 14, an assumption the American College of Obstetricians and Gynecologists (ACOG) has been explicit about not holding for every pregnancy. Our calculators let you enter your own average cycle length so the math adjusts around that assumption rather than pretending it is universal.
ACOG Committee Opinion No. 700, Methods for Estimating the Due Date, states that ultrasound measurement of the embryo in the first trimester, up to 13 weeks 6 days, is the most accurate method available for establishing gestational age, more accurate than counting from the last period even when that date is known with confidence. The committee opinion sets out exactly when a scan should move the due date: before 9 weeks, a crown-rump length that differs from the last-period date by more than 5 days is grounds for redating; from 9 weeks through 13 weeks 6 days, the threshold is more than 7 days; from 14 through 15 weeks 6 days it is still 7 days; from 16 through 21 weeks 6 days it widens to 10 days; from 22 through 27 weeks 6 days it is 14 days; and from 28 weeks on, where ultrasound dating is least reliable, it takes a difference of more than 21 days. Once a first-trimester scan sets a due date, ACOG advises against moving it again based on a later scan, because growth differences later in pregnancy usually reflect the size of the baby, not an error in dating. A companion opinion, Committee Opinion No. 688 on suboptimally dated pregnancies, treats any pregnancy without a confirming scan before 22 weeks as imprecisely dated. None of our calculators can see your ultrasound report, so when a scan and the last-period math disagree, the scan is the one to trust, and the person who can tell you that with certainty is whoever read it.
The Trimester Calculator and the full-term guide use the categories ACOG adopted in Committee Opinion No. 579, Definition of Term Pregnancy: early term is 37 weeks 0 days through 38 weeks 6 days, full term is 39 weeks 0 days through 40 weeks 6 days, late term is 41 weeks 0 days through 41 weeks 6 days, and post-term is 42 weeks and beyond. A working group convened by ACOG replaced the older single "term" label with these four bands after data showed that babies born at 37 or 38 weeks have measurably higher rates of breathing and feeding problems than those born at 39 or 40 weeks. Preterm birth, before 37 weeks, is defined the same way by the CDC's preterm birth program, which also tracks how common it is in the United States.
The 280-day count is a median, not a promise, and the clearest evidence for that comes from a study that could measure pregnancy length precisely because it tracked confirmed ovulation dates rather than last-period estimates. Jukic and colleagues followed 125 women trying to conceive naturally and timed ovulation directly; among pregnancies that reached term, the median length was 268 days from ovulation, close to but not exactly what Naegele's rule predicts, and the middle range of pregnancy lengths still spanned about five weeks even after excluding preterm births (Jukic et al., Human Reproduction, 2013). We use that study, not a rounder or more dramatic-sounding figure, to explain on the due date accuracy guide why a calculated date is a midpoint of a real range rather than a delivery deadline.
The hCG Doubling Time Calculator reflects the pattern described in the clinical hCG literature going back to Kadar's original curves and refined by Barnhart's later cohort work, summarized in a 2024 practical review, beta-hCG Dynamics in Early Gestational Events: hCG commonly doubles every 48 to 72 hours at low starting levels, then rises more slowly as levels climb, which is why a single hCG number tells a clinician very little compared with the trend across two draws. The Pregnancy Weight Gain Calculator follows the ranges published by the National Academies (formerly the Institute of Medicine) in their 2009 report, Weight Gain During Pregnancy: Reexamining the Guidelines, which sets a different target range depending on pre-pregnancy BMI category. We did not build a house formula for either calculator; both run the published method.
Every method described above was designed for population-level estimation, not for reading your specific chart. A cycle length field cannot know that you ovulate late, an hCG doubling formula cannot see an ectopic pregnancy, and a BMI-based weight range cannot account for a twin pregnancy or a condition your provider is already managing. None of the calculators on this site diagnose anything, and a result from any of them is a number to bring into a prenatal visit, not a substitute for one.
The ACOG committee opinions cited here are periodically reaffirmed or replaced; we check the live ACOG and CDC pages against what is written here roughly twice a year and whenever we notice a committee opinion has been superseded. The Jukic study and the 2009 weight gain report are fixed, published works and will not change, though newer research could eventually supersede either, in which case we will cite the newer work and note the date of the change here.
If you run a calculator's own stated method by hand and get a different result, or if you find that one of the sources above has been updated since we last checked it, tell us through the contact form. A genuine calculation error gets fixed and the change is dated; a disagreement about which threshold to use where two sources differ is a judgment call we will explain rather than silently overwrite.
This site runs on display ads through Google AdSense and a small number of affiliate links. That revenue has no role in which formula a calculator uses or what number it returns; the fastest way to lose the one thing this site has going for it, that the math matches the source, would be to bend it for a more dramatic result.
For who is actually behind this site and what gets checked before a guide publishes, see the authors page.