Period Calculator
Map out when your next periods are due, and the days in between when ovulation is most likely, from the start of your last period and how long your cycles usually run. The calendar scrolls forwards and backwards three months at a time, so you can look ahead or check a month that has already passed.
Important dates for the next 6 cycles:
| Period | Most Probable Ovulation Days |
|---|
Estimates built from an average cycle. Real cycles vary, and this is not a reliable way to avoid pregnancy.
What this is actually predicting
This calculator does one thing: it repeats your cycle forward. Give it the day your last period started and how long your cycles usually run, and it stamps that same interval out across the calendar, six cycles ahead and as far back as you care to scroll.
That is a model, not a forecast in the weather sense. It has no information about you beyond two numbers, so it cannot know that you are about to start a new job, cross five time zones, or come off the back of a virus — all of which can move a period by days. What it gives you is where your period would fall if your body repeated itself exactly, which is a useful default and a bad certainty.
Most people come to a page like this for something practical: booking a holiday, picking a date for an event, planning around an exam, or just wanting to know whether being four days late is unusual. For all of those, an interval model is genuinely enough. It is when the answer needs to be exact that you should stop trusting it.
Two numbers people mix up
The form asks for two lengths and they measure different things. Getting them the wrong way round is the single most common reason a prediction comes out badly wrong.
how many days you actually bleed — usually somewhere between 2 and 7
Cycle length
first day of one period → the day before the next one starts — usually 21 to 35
How the calendar is built
period days = first day + 0 … + (period length − 1)
ovulation days = first day + (cycle − 16) … + (cycle − 12)
and the whole pattern repeats every cycle length, forwards and backwards
Count the cycle from the first day of proper bleeding, not from spotting, and count up to the day before the next period rather than the day it arrives — otherwise you count one day twice. A cycle running the 9th to the 8th of the next month is a 28-day cycle when February is not involved.
Period length is the easier of the two but is worth being honest about. Count days with real bleeding. A day of light spotting at the end is not usually counted, and including it will paint a wider pink band on the calendar than your body will produce.
What counts as regular
Regular does not mean identical. Almost nobody runs the same number of days every single cycle, and a difference of a few days between one and the next is entirely ordinary. The usual working definition is that a cycle is regular when the gap between your shortest and longest cycle across several months stays under about eight days.
That threshold is more useful than it looks, because it tells you what to enter here. If your cycles run 27, 29, 28 and 30 days, your spread is three days and an average of 28 will serve you well. If they run 24, 31, 27 and 35, the spread is eleven and no single average describes them — the calendar will look precise and be wrong.
When the spread is that wide, the spread itself is the answer. Enter your average and then read the printed date as the middle of a band as wide as your spread — eleven days of variation means a date give or take five, and no amount of arithmetic narrows that. Writing the band on the calendar is more honest than believing the day.
Why the later months drift
The six rows in the table are not equally trustworthy, and the reason is worth understanding rather than just accepting.
Only the first row is anchored to something that actually happened — the date you typed in. Row two assumes row one landed exactly. Row three assumes rows one and two both did, and so on. Errors do not cancel out as they compound; they accumulate. If your cycles wobble by two days in either direction, the sixth row is carrying something like a fortnight of slack even though it prints a single date.
The practical rule that follows is simple: recalculate whenever a period actually arrives. One observed date resets the whole chain and is worth more than any number of projections stacked on each other. Treat the near rows as dates and the far ones as rough positions in a month.
Scroll back to test the prediction
The calendar moves backwards as well as forwards, and that is more useful than it first appears. Most people only look ahead, but the months behind are the only way to find out whether the model fits you at all.
Try this. Enter the start date of your most recent period, then page back through the previous three or six months and compare the shaded bands against what actually happened. If your remembered dates sit inside the pink, your cycle length is about right. If the prediction runs steadily earlier or later the further back you go, your true average is a day or two off what you entered, and nudging the cycle length by a day will usually snap it into place.
This is the cheapest accuracy improvement available on this page, and it costs nothing but a few clicks. A cycle length that fits the last six months will predict the next six far better than the number you assume is yours.
Changes worth a doctor's attention
A calculator can tell you that a period is late. It cannot tell you why, and there are patterns where the right response is a medical appointment rather than a recalculation.
Worth raising with a doctor: periods that stop for three months or more when you are not pregnant; bleeding that is heavy enough to soak through a pad or tampon every hour or two, or that passes clots larger than about a coin; bleeding between periods or after sex; periods lasting longer than seven days; cycles that suddenly become much shorter, much longer or much more erratic than your own normal; pain severe enough to stop you doing ordinary things; and any bleeding at all after menopause.
None of that list is a diagnosis and none of it means something is seriously wrong — several have ordinary explanations. But they are the changes that are worth a professional look rather than another month of watching, and a calendar is not equipped to reassure you about any of them.
What it cannot do
Repeating an interval only works while there is an interval to repeat, and plenty of ordinary circumstances remove one. On hormonal contraception the bleeding you get is scheduled by the method rather than by a cycle, so these dates describe a pattern you are not currently running. Postpartum, during breastfeeding, and across the years leading into menopause the spacing shifts month to month by design. Polycystic ovary syndrome, thyroid disorders and significant weight or training changes can stretch cycles out or remove them entirely. In all of those the calendar will still draw neat bands, and the neatness is the misleading part.
The green days deserve a specific warning: they mark where ovulation is statistically likely in an average cycle, not where it happened in yours, and they are far too loose to plan around if pregnancy would be unwelcome. For how the fertile window works and what your body signals about it, see the ovulation calculator; if you are actively trying to conceive, start with the conception calculator.
Frequently asked questions
How do I calculate when my next period is due?
Add your average cycle length to the first day of your last period. If your last period started on 9 August and your cycles run 28 days, the next one is due around 6 September. The calendar on this page repeats that interval forward six cycles and shades the days you would expect to bleed.
What is the difference between cycle length and period length?
Period length is how many days you bleed, commonly two to seven. Cycle length is the gap from the first day of one period to the day before the next begins, commonly 21 to 35 days. The calculator needs both: period length sets how wide the shaded bands are, cycle length sets how far apart they sit.
Is it normal for my period to be a few days late?
Variation of a few days between cycles is very common and is not usually a sign that anything is wrong. Illness, disrupted sleep, travel, stress and changes in weight or exercise can all shift the timing. A cycle is generally still considered regular if your shortest and longest across several months differ by less than about eight days.
How many months should I track before trusting the average?
Three to six gives you something worth using. One cycle tells you almost nothing, since any single cycle can be an outlier. Once you have a few, use the calendar's backward navigation to check your entered cycle length against the months that have already happened and adjust it if the shading has drifted.
Can I use this to avoid getting pregnant?
No, and the shading on this page is not built for that job. It marks where ovulation tends to fall in an average cycle, while the actual release date shifts from month to month for reasons the calendar cannot see. Methods that depend on predicting it have a much higher failure rate than other options, so if pregnancy would be unwelcome, that is a conversation to have with a doctor or clinic rather than a calculator.
Why are the last rows of the table less reliable?
Because each row assumes every cycle before it ran to exactly the length you entered. The first row rests on a date you observed; the sixth rests on five projections stacked on top of each other, so small variations compound. Recalculate each time a period actually starts and the chain resets.
What if my cycles are irregular?
No single average will describe them well. Work out the gap between your shortest and longest recent cycle and carry that number with you: a printed date is only meaningful give or take roughly half that spread. If your cycles are very erratic, have stopped, or have shifted noticeably from what is normal for you, that is worth raising with a doctor rather than modelling around.
Everything here is produced by repeating one interval across a calendar. It records nothing about your body, diagnoses nothing, is not medical advice, and is not adequate as contraception. Timing shifts with illness, sleep, travel, medication, training load, weight and a number of treatable conditions, none of which the arithmetic can detect. Bleeding that stops for months, soaks through protection hourly, runs past a week, arrives between periods or after sex, hurts enough to interrupt your day, or occurs at all after menopause should be looked at by a doctor, midwife or nurse rather than tracked for another month. Learn more about how we build our calculators and read our privacy policy.