Medical disclaimer. This article is general information, not medical advice, and nothing here is a method of contraception. If your cycles are absent or very irregular, or you have been trying to conceive without success, talk to your doctor or midwife about your own situation.
Your ovulation window is shorter than almost everyone expects: six days, ending on the day you ovulate. That is the five days before ovulation plus ovulation day itself. Sperm can survive in the fallopian tubes for up to seven days, so intercourse before ovulation still counts; the window closes on ovulation day, so the day after usually does not. In the study that established this, conception occurred only when intercourse fell inside those six days — and not once from intercourse after the estimated day of ovulation.
The hard part is not the length of the window. It is knowing where those six days sit in your particular cycle — and that is where calendar math starts making promises it cannot keep.
The six-day window, in numbers
In 1995, Wilcox, Weinberg and Baird studied 221 healthy women who were planning to become pregnant. Each collected daily urine specimens and kept a record of when she had intercourse, so the day of ovulation was estimated from hormone metabolites rather than guessed from a calendar. Across 625 menstrual cycles, 192 pregnancies were identified — 129 of them ended in live births — and conception occurred only when intercourse took place during a six-day period ending on the estimated day of ovulation.
The days inside that window are not equal. The paper reports the probability of conception from intercourse on a single day rising from about 0.10 five days before ovulation to about 0.33 on the day of ovulation itself — more than a threefold difference inside one week. The authors' own conclusion is carefully worded: nearly all pregnancies can be attributed to intercourse in those six days.
Dunson and colleagues later followed 782 couples across 5,860 cycles and found the same shape — "nearly all pregnancies occurred within a 6 day fertile window," with no evidence of a shorter fertile window in older men or women. What age changed was the height of the curve, not its width: day-specific probabilities declined for women from the late twenties onward, and were about twice as high at 19–26 as at 35–39.
ASRM's committee opinion uses the same definition — "the 6-day interval ending on the day of ovulation" — and adds the practical version. Intercourse every one to two days during the fertile window "yields the highest pregnancy rates," but the margin is narrow: cycle fecundity was "similar with intercourse that occurred daily, every other day, and even every 3 days in the fertile window." Going more often than that "is not associated with lower fecundity," so there is nothing to ration.
Ovulation is anchored to your next period, not your last one
This is the single most useful thing to understand about cycle timing, and it is the opposite of how most people count.
The NHS puts ovulation at around 10 to 16 days before the next period — measured backwards from the period that has not arrived yet, not forwards from the one that has. Notice that even this anchor is a week wide. It is the steadier end of the cycle to count from; it is not a constant.
So "day 14" is not biology. It is the answer you get if you assume a 28-day cycle and a 14-day second half. Both are conventions, and either can be wrong for you.
How cycle length moves the window
Working from that assumption, the usual estimate takes the middle of the NHS range: ovulation lands about 14 days before your next period, counting cycle day 1 as the first day of bleeding. That is the arithmetic inside ProbaBaby's ovulation calculator too — cycle length minus 14, then five days back from there. Play that out:
| Cycle length | Estimated ovulation | Fertile days |
|---|---|---|
| 24 days | around cycle day 10 | days 5–10 |
| 28 days | around cycle day 14 | days 9–14 |
| 32 days | around cycle day 18 | days 13–18 |
| 35 days | around cycle day 21 | days 16–21 |
Look at the 24-day row. On a short cycle, the fertile window can open while you are still bleeding — which is why "you can't get pregnant during your period" is bad advice rather than a rule. Sperm survival does the rest of the work.
And on a 35-day cycle, the classic advice to "try around day 14" points at a week that is not fertile at all.
Why a calendar estimate stays an estimate
In 2000, the same research group published day-by-day estimates from 696 cycles contributed by 213 women, and the result deserves to be better known. Women had at least a 10% chance of being in their fertile window on every single day between cycle day 6 and cycle day 21. In only about 30% of women did the fertile window sit entirely inside the textbook days 10 to 17. Even women who considered their cycles regular had a 1–6% chance of being in their fertile window on the day their next period was due.
That is not a flaw in the arithmetic. It is what happens when you apply a population average to one person's month. A predicted ovulation day is a reasonable place to start looking — it is not a result.
Which is why the signs your body gives you matter more than the square on the calendar.
Three signs, and what each one is actually for
Cervical mucus — the forward-looking one
As estrogen rises before ovulation, cervical mucus changes from scant and sticky to clear, slippery and stretchy, like raw egg white. That change is a prediction: it shows up while the window is still open.
ASRM cites pregnancy rates of roughly 38% for intercourse on the day of peak mucus, against roughly 15–20% the day before or after, and puts the probability highest "when the mucus is slippery and clear." One large study it summarises found that mucus predicted the day-specific probability of conception as well as or better than basal body temperature or urinary LH monitoring.
LH tests (OPKs) — the "soon" alarm
Ovulation predictor kits detect the surge of luteinising hormone that precedes ovulation, and ovulation may follow within two days of a detected surge. ASRM notes that a randomised controlled trial found these devices decreased time to pregnancy — and that false positives occur in roughly 7% of cycles. One strange result is not the story; the pattern across the cycle is.
Basal body temperature — the receipt
Your resting temperature rises after ovulation, not before it. Progesterone from the corpus luteum acts on the hypothalamus to reset the set point, lifting basal body temperature by roughly 0.5–1 °F through the luteal phase. So the shift is a receipt: by the time you can see it, the window has already closed.
It is also a blunter instrument than charting apps imply. Studies of the method describe temperature as an unreliable technique for determining accurate ovulation timing, and some have put it at only around 22% accurate for detecting ovulation at all — which is why it works best as a months-long picture of your own pattern, and badly as a verdict on today.
Used together, the three cover each other's blind spots. Mucus and LH look forward; temperature looks back and confirms the other two were right.
If your cycles are irregular
When your cycle length changes a lot from month to month, the subtraction has nothing stable to work from — there is no reliable "next period" to count back from. In that situation the body signs stop being a nice extra and become the primary source.
It is also worth saying plainly to a clinician. Absent periods, cycles that change dramatically, bleeding patterns that are new for you, or a stretch of trying without success are all reasons to book an appointment rather than to buy more test strips.
How ProbaBaby helps
ProbaBaby's fertility planning module is free — no credits and no paywall anywhere in it — and it treats the calendar as the starting point rather than the answer. She can log six things: basal body temperature, cervical mucus, period, symptoms, intimacy and ovulation tests, each with its own sheet, its own history and its own chart. Four signals — calendar, a BBT shift, an OPK peak and egg-white mucus — are combined into one ovulation assessment with a confidence level, and a biomarker-confirmed ovulation re-centres the fertile window; the calendar signal on its own never counts as confirmation.
The module draws the published Wilcox day-by-day probabilities as a real curve rather than a decorative arc, running from five days before ovulation to the day after — where the value is zero — with implantation marked at 6 to 10 days past ovulation and the earliest reliable test at 14. The app's standalone ovulation and fertility-window calculators follow the rule the studies give: six days, ending on ovulation, and nothing after it. The app prints its own disclaimer alongside all of it — estimates only, not a contraceptive and not medical advice — and the cycle length it starts from is the one in your profile, so it is worth keeping that accurate.
If the window closes the way you hoped, the arithmetic changes: a due date is counted from the first day of your last period, which is why our due date calculator asks for that rather than for a conception date. We wrote up how due date calculators work if you want the formulas behind the number.
The short version
Six days, ending on ovulation day. Ovulation sits roughly two weeks before your next period, not two weeks after your last one. Mucus and LH tests tell you the window is opening; temperature tells you it closed. And a predicted date is a hypothesis your body is free to ignore.