Why the calculation counts backwards, not forwards
The menstrual cycle has two phases that behave very differently. The follicular phase — from the first day of bleeding to ovulation — is the variable one: it stretches and shortens with stress, illness, travel, weight change and simple individual variation, and it accounts for nearly all the difference between a 24-day cycle and a 35-day one. The luteal phase — ovulation to the next period — is governed by the lifespan of the corpus luteum and sits close to 14 days in most people, with a usual range of roughly 11 to 17.
That asymmetry is why “ovulation happens on day 14” is only true for a 28-day cycle, and why the reliable way to locate it is to subtract the luteal phase from the cycle length rather than to add 14 to the start. In a 32-day cycle with the same 14-day luteal phase, ovulation is on day 18, not day 14 — and someone timing intercourse for days 12 to 16 would miss the window at exactly its most fertile end.
Get the two definitions straight, because they are easy to slip. Cycle day 1 is the first day of full flow. The luteal phase length used here is the number of days after ovulation up to and including the day before the next period, so a 28-day cycle with a 14-day luteal phase ovulates on day 14, bleeds again on day 29, and the days 15 through 28 are the luteal phase — fourteen of them.
What makes the fertile window six days long
Two lifespans set the width of the window, and they are wildly different in length. Sperm can remain viable in the female reproductive tract for up to about five days when fertile cervical mucus is present — that mucus is what makes it possible, which is why the window is a phenomenon of the late follicular phase rather than of any five days at random. The oocyte, once released, is fertilisable for only about 12 to 24 hours.
Put those together and the window is strongly asymmetric: five days of runway before ovulation and less than a day after it. Wilcox, Weinberg and Baird's 1995 analysis of prospectively collected data found that conceptions arose only from intercourse during a six-day interval ending on the day of ovulation itself, with no pregnancies attributed to intercourse on the days that followed. That is the interval this calculator uses by default, and it is why the “days past ovulation” setting starts at zero rather than at one.
The practical consequence runs against intuition. Waiting until you detect ovulation and then having intercourse is the least effective timing available, because by then most of the window has already gone. The productive strategy is to cover the days leading up to it — which is exactly what the window in the results is for.
The width of the window is fixed by biology, but its position is not fixed by the calendar. Wilcox and colleagues later reported in the BMJ that the timing of the fertile window varies far more between and within women than the standard advice assumes, and that only a minority of women had their fertile window fall entirely within the cycle days 10 to 17 that textbook guidance points at — including women who described their cycles as regular. Calendar prediction gives you a place to start, not a certainty.
Worked example: a 32-day cycle starting 1 August 2026
Take someone whose periods start every 32 days and whose charting has shown a 14-day luteal phase, with the most recent period beginning on 1 August 2026.
- Find the ovulation day. 32 − 14 = day 18 of the cycle.
- Convert that to a date. Cycle day d falls on the start date plus (d − 1) days, so day 18 is 1 August + 17 days = 18 August 2026. (The day number and the August date coincide here only because the cycle began on the 1st.)
- Open the window. 18 − 5 = day 13, which is 1 August + 12 = 13 August.
- Close the window. Day 18 + 0 = day 18, 18 August.
- Check the length. 18 − 13 + 1 = 6 days, as expected.
- Next period. Cycle day 33, which is 1 August + 32 = 2 September 2026.
- Cycle after that. Add another 32 days throughout: period 4 October, window 14–19 September, ovulation 19 September.
Compare this with what a naive day-14 assumption would have given: intercourse concentrated around 14 August, which sits on day 14 — inside the window, but at its least fertile end, four days before ovulation. The 32-day cycle has moved the peak four days later than the folklore expects.
How to use the dates you get
Cover the window rather than aim at a single day. Intercourse every day or every other day across the six-day interval is the approach that maximises the chance of sperm being present when the oocyte is released, and it removes the need to detect ovulation precisely. Trying to hit one day exactly is the strategy most sensitive to the calendar being wrong.
Confirm the prediction with a physiological sign rather than trusting the arithmetic alone. Cervical mucus becoming clear, stretchy and slippery is the classic marker that the fertile window has opened; a urinary LH test turns positive roughly a day or so before ovulation, so a positive result means the window is nearly closed rather than just opening; and a sustained rise in basal body temperature confirms ovulation only after it has happened, which makes it useful for learning your own luteal phase length but useless for timing this cycle.
Feed what you learn back into the inputs. The single most valuable improvement you can make to this calculation is replacing the assumed 14-day luteal phase with your own measured one, because that number is what positions the whole window. Chart temperatures for two or three cycles, count the days from the temperature rise to the next period, and use the average.
If a cycle results in a pregnancy, the ovulation date you have here becomes the best dating anchor available — better than the last menstrual period, because it removes the assumption this page had to make. Enter it as a conception date in the due date calculator, which adds 266 days rather than 280.
Ovulation day and fertile window by cycle length
| Cycle length | Ovulation on cycle day | Fertile window (cycle days) | Next period on cycle day |
|---|---|---|---|
| 21 days | 7 | 2–7 | 22 |
| 24 days | 10 | 5–10 | 25 |
| 26 days | 12 | 7–12 | 27 |
| 28 days | 14 | 9–14 | 29 |
| 30 days | 16 | 11–16 | 31 |
| 32 days | 18 | 13–18 | 33 |
| 35 days | 21 | 16–21 | 36 |
| 40 days | 26 | 21–26 | 41 |
| 45 days | 31 | 26–31 | 46 |
Read down the ovulation column and you can see the whole point of the method: the window moves one day later for every day the cycle lengthens, while its width never changes.
What this calculation cannot do
- It cannot detect ovulation. It projects where ovulation would fall if the coming cycle matches the length you entered. Anovulatory cycles happen, and a cycle can produce a bleed without an egg having been released.
- It is not a contraceptive method. Calendar-only prediction is the weakest of the fertility awareness approaches, precisely because the window's position varies. Methods intended for avoiding pregnancy combine mucus and temperature observations with defined rules and require training.
- Irregular cycles break the assumption. If your cycles range from 25 to 38 days, no single figure describes them, and the honest reading of the projection is a wide band rather than specific days. Cycles that are persistently irregular, absent, or shorter than 21 or longer than 35 days are worth investigating with a clinician.
- It assumes a constant luteal phase. That holds well for most people but not for all. A luteal phase under 10 days is worth confirming with charting or a mid-luteal progesterone test rather than assuming.
- It says nothing about fertility itself. Timing intercourse correctly raises the chance of conception in a given cycle; it does not address egg or sperm quality, tubal patency, or any other factor. Standard advice is to seek assessment after twelve months of trying, or after six months if the female partner is over 35.
- Hormonal contraception and recent pregnancy disrupt the pattern. The first several cycles after stopping the pill, or after a birth or miscarriage, are frequently long or anovulatory, so calendar prediction has nothing stable to work from.
Sex ratio, timing and other things the window does not control
A persistent piece of folklore holds that intercourse close to ovulation favours one sex and intercourse several days before favours the other. The prospective data do not support it: Wilcox and colleagues specifically examined the sex ratio against the timing of intercourse within the fertile window and found no relationship. Timing changes the probability that conception occurs at all, not what is conceived.
Where calendar prediction sits among the alternatives
Calendar prediction is the cheapest and least precise member of a family of methods. One step up, cervical mucus observation reads the window as it opens, since the fertile-type mucus that lets sperm survive appears in response to rising oestrogen in the days before ovulation. Urinary LH testing detects the surge that triggers ovulation, which is precise but late — useful for confirming that the window is real, less useful for planning it. Basal body temperature is retrospective: the shift happens after ovulation, so it tells you what happened rather than what is about to.
The symptothermal approach combines mucus and temperature with explicit rules and is the version taught for both achieving and avoiding pregnancy. WHO's Family Planning: A Global Handbook for Providers sets out the calendar-based and symptoms-based methods with their effectiveness figures, and any use of these methods for contraception rather than conception should start there or with a trained instructor rather than with a calculator.
Once a pregnancy is established, the arithmetic on this page hands over directly. The ovulation date becomes the conception anchor for the due date calculator, gestational weight tracking follows in the pregnancy weight gain calculator, and once the baby arrives the child growth percentile calculator picks up the measurement side.