Period Calculator

What a menstrual cycle measures

A menstrual cycle is measured from the first day of one period to the first day of the next, not from the last day of one to the first day of the next. Day 1 is the first day of full flow. The cycle ends the day before the next period begins, so a 28 day cycle that starts October 1 has its next start on October 29.

That single definition explains most of the confusion around period tracking. If your period lasts 5 days and your cycle is 28 days, the gap between the end of one bleed and the start of the next is 23 days, not 28. Counting from the last day of the period shifts every prediction about a week early, which is enough to make a working forecast look broken.

Cycle length also differs between people and between cycles in the same person. The Office on Women's Health puts the typical adult range at 21 to 35 days, with wider ranges common in the first years after menarche and again in the years before menopause. A 28 day cycle is an average, not a rule, and a person who runs at 31 days is not irregular.

The calculator on this page takes one date and one number. From those it produces the next start, the three starts after that, the end of the next period, and an ovulation estimate. Everything below explains where those numbers come from and how much weight to put on them.

How the next period date is calculated

The arithmetic is one line: next period start equals the first day of the last period plus your cycle length in days. Every later date is the same addition repeated. If the last period began October 1 and the cycle runs 28 days, the next start is October 29, then November 26, then December 24, then January 21.

Two rules keep the output usable. Use the first day of flow as the anchor rather than any other day, and use an average cycle length taken from real dates rather than a number from a website. If your last six cycles ran 29, 27, 30, 28, 28 and 29 days, the average is 28.5, and rounding to 28 is fine because the underlying data does not support half days.

The end of the next period is the start plus the period length minus one day. A 5 day period that starts October 29 runs through November 2, which is four days after the start, not five. That off by one is the second most common source of bad dates after counting from the wrong day.

The table below runs the same October 1 anchor at several cycle lengths, so you can see how much one number moves the calendar.

A 7 day difference in cycle length moves the third predicted start by 21 days, which is why the average matters more than the last cycle alone.

Worked example: a 28 day cycle from October 1

Take a last period that began October 1, 2026, a 28 day cycle, and a period that lasts 5 days. These are the calculator's defaults, and the output reads like this:

• Next period starts: October 29, 2026
• Next period ends: November 2, 2026
• Following starts: November 26, 2026 and December 24, 2026
• Fourth start: January 21, 2027
• Estimated ovulation: October 15, 2026
• Fertile window: October 10 to October 16

The ovulation date comes from subtracting 14 days from the next start. That 14 day figure is the typical length of the luteal phase, the stretch from ovulation to the next period, and it is more stable across people than the first half of the cycle. The follicular phase, which runs from day 1 to ovulation, is the part that stretches and shrinks, which is why a longer cycle usually means later ovulation rather than a longer luteal phase.

Change the cycle length to 32 days and the next start becomes November 2, the second becomes December 4, and ovulation moves to October 19. Everything shifts together, which is the behavior you want from a forecast built on one assumption.

The forecast is only as good as your history

A period prediction is a projection of past rhythm onto future dates, and it inherits every flaw in the history behind it. One cycle tells you almost nothing. Three cycles give a rough range. Six cycles give a usable average, and twelve let you see seasonal or lifestyle patterns.

The average is the right input, not the shortest or the longest. If your last six cycles ran 26, 31, 28, 30, 27 and 29 days, the average is 28.5 and the spread is 5 days. Entering 26 produces dates that arrive early half the time, and entering 31 produces dates that arrive late, so the average minimizes the miss.

Some history is not worth averaging. A cycle that was 45 days long because of an illness, a travel week or a new medication is an outlier, and folding it into the mean drags every future date late. Drop the outlier, note why, and keep the average built from ordinary cycles.

Freshness matters too. A cycle length from two years ago describes a different body and often a different schedule. Rebuild the average every few months, or after any event that changed your routine for more than a couple of weeks.

How far off can the prediction be

Error grows with the variability of your cycles and with how far ahead you project. A cycle that swings 3 days in either direction can move a single prediction by 3 days and a three-cycle prediction by 9. A cycle that swings 6 days produces an 18 day window by the third cycle, which is wide enough that the date stops being useful for planning anything specific.

That is the honest answer to the question of how far off a prediction can be: for a steady cycle, within 2 to 3 days; for a moderately variable cycle, within a week at the next period and two weeks by the third; for a cycle that varies by more than 8 days, the forecast is a rough guess and should be treated as one.

The pattern is linear, and it is the reason most tracking apps show a range rather than a single date once the data is noisy. A range of 4 days around a predicted start is realistic for a steady cycle. A range of 10 days is not a failure of the tool, it is what the input supports.

Ovulation estimates carry more error than the period date, because they assume a 14 day luteal phase and a cycle length that held in the past. Treat the ovulation day as the middle of a window that runs about 5 days before it and 1 day after, and treat the whole window as an estimate rather than a schedule.

What shifts cycle length, and when to ask

Cycle length changes for ordinary reasons and for medical ones, and the difference usually shows up in how long the change lasts.

• Stress. A few weeks of poor sleep or a hard deadline can delay ovulation by a few days to a week, which pushes the whole cycle late.
• Travel and time zones. Long flights and shift work disturb the hormonal timing that sets ovulation, and one late cycle after a trip is common.
• Weight change. Losing or gaining a noticeable amount of weight, or dropping below a certain body fat level, can shorten, lengthen or stop cycles.
• Intense training. Heavy endurance training is a known cause of longer cycles and missed periods.
• Perimenopause. Cycles shorten first, then become erratic, often with a gap of several months before periods stop for good.
• Thyroid problems and PCOS. Both change cycle length, and both are diagnosed with blood tests rather than by tracking alone.

The ACOG guidance on abnormal uterine bleeding treats a cycle shorter than 21 days or longer than 35 days as worth a conversation with a clinician, along with bleeding that lasts more than 7 days, bleeding between periods, or a sudden change from a stable pattern. The NICHD menstruation pages cover the research side of the same questions.

A single odd cycle is usually nothing. Three months of a new pattern is a reason to write the dates down and take them to an appointment, because a log of six dates is far more useful to a clinician than a memory of feeling off.

Planning dates, not birth control

The dates from this tool are useful for planning and useless as contraception. Calendar based methods fail because ovulation can happen earlier or later than the 14 day assumption, sperm can survive several days, and a short cycle can place the fertile window close to the end of bleeding.

Where the forecast does earn its place: scheduling a trip, a swim meet or a wedding, packing supplies, timing a refill, or knowing when to expect a test result to be meaningful. For a pregnancy test, the useful date is the first day of a missed period, which is the predicted start plus a day or two. Testing earlier than that raises the chance of a false negative because the hormone has not built up yet.

For an appointment, give the scheduler the predicted week rather than a single day, and mention that the date is an estimate. For a vacation, book the flexible option if the trip lands within 3 days of a predicted start.

This page is an estimate for planning, not medical advice, and it cannot confirm ovulation or diagnose a cycle disorder. Dates assume a regular cycle length that you supply. Talk to a clinician about anything that changes your cycle for more than a few months.