A six-cycle calendar from your last period — start and end dates, plus the ovulation and premenstrual windows for each.
Cycle length is counted from the first day of one period to the first day of the next — not from when bleeding stops.
| Cycle | Period starts | Period ends | Ovulation (approx.) | PMS window from |
|---|
Each predicted period is the previous start date plus your average cycle length. Cycle length is measured from the first day of bleeding to the first day of the next — not from the day bleeding stops, which is the most common way people end up entering a number roughly a week too short.
Ovulation is estimated by counting back 14 days from each predicted period, not forward from the last one. That direction matters. The luteal phase — ovulation to period — is fairly consistent at around 12 to 14 days in most people. The follicular phase, before ovulation, is the part that stretches and contracts, and it is what makes cycles vary in length. Counting forward from the last period assumes the variable part is fixed, which is exactly backwards.
The textbook 28-day cycle is a convention, not a norm. Anything from 21 to 35 days in adults is considered within the normal range, and only around one in eight cycles is actually 28 days. Variation of a few days between cycles in the same person is entirely ordinary.
Cycles tend to be longer and more irregular in the first few years after periods begin, settle through the twenties and thirties, and become irregular again in the years approaching menopause. Bleeding typically lasts 2 to 7 days.
| Phase | Roughly when | What is happening |
|---|---|---|
| Menstrual | Days 1–5 | The uterine lining sheds. Oestrogen and progesterone are at their lowest. |
| Follicular | Days 1–13 | Follicles develop and oestrogen rises, rebuilding the lining. This is the phase whose length varies. |
| Ovulation | Around day 14 | A surge in luteinising hormone releases an egg, which survives 12 to 24 hours. |
| Luteal | Days 15–28 | Progesterone rises to maintain the lining; if there is no implantation it falls, and the cycle restarts. Fairly fixed at 12–14 days. |
Pregnancy is the reason people think of first, and worth ruling out with a test if a period is more than a week late and there is any possibility. But it is far from the only cause. Significant stress suppresses the hypothalamic signalling that drives the cycle. Illness, travel across time zones, sharp changes in weight, heavy training loads, thyroid disorders, and polycystic ovary syndrome all delay or skip periods. Hormonal contraception changes the picture entirely — bleeding on the pill is a withdrawal bleed rather than a true period.
Worth raising: cycles consistently shorter than 21 days or longer than 35; bleeding heavy enough to soak through a pad or tampon hourly for several hours, or lasting more than seven days; bleeding between periods or after sex; pain severe enough to interrupt normal activity; periods stopping for three months or more without pregnancy; or any new bleeding after menopause, which should always be checked promptly.
Tracking helps here in a way a single forecast does not — a few months of recorded dates gives a doctor far more to work with than a description from memory.
Count from the first day of bleeding in one period to the day before bleeding starts in the next. If your period began on the 3rd of one month and the 31st, that is a 28-day cycle. The frequent mistake is counting from the day bleeding stops, which produces a number about five to seven days too short and throws every subsequent prediction off. Averaging over three to six recent cycles gives a more reliable figure than any single one.
Because the luteal phase — ovulation to the next period — is the consistent part, at roughly 12 to 14 days for most people. The follicular phase before ovulation is what varies and what makes some cycles longer than others. Counting forward from the last period assumes the variable part is fixed, which gets ovulation wrong for anyone whose cycle is not exactly 28 days. Counting back from the predicted period is the more reliable direction.
It is normal, but so is a good deal else. Anything from 21 to 35 days in adults falls within the normal range, and in practice only about one cycle in eight is exactly 28 days. The 28-day figure is a convention that came from averaging, not a target. A few days of variation between your own cycles is also entirely ordinary and not a sign of anything wrong.
No. Calendar-based prediction is among the least reliable approaches to avoiding pregnancy, because ovulation timing shifts with stress, illness, travel and weight change, and sperm can survive up to five days in the reproductive tract. Even carefully practised fertility awareness methods, which combine temperature and cervical mucus tracking rather than dates alone, have meaningfully higher failure rates than hormonal or barrier contraception.
Many reasons. Significant stress suppresses the hormonal signalling that drives the cycle. Illness, long-distance travel, sharp weight change in either direction, heavy training, thyroid problems and polycystic ovary syndrome all delay or skip periods. Some medications do too. A single late or missed period is common and usually resolves on its own; three or more missed cycles without pregnancy is worth investigating.
Related: Ovulation Calculator · Pregnancy Calculator · Conception Calculator · Pregnancy Week Calculator