Pregnancy Chance by Cycle Day

Explore how cycle length and variation can change estimated pregnancy chance from one unprotected encounter.

Count from the first day of one period to the first day of the next, not the days you bleed.
Enter how many days shorter or longer your cycles can be: 29 with a variation of 3 means 26 to 32 days. The starting values are an example, not an estimate of your cycle.
Use recent cycle lengths (optional)
Enter consecutive completed cycles, separated by commas or spaces. These replace the two fields above; leave blank to use those fields instead.
Did we solve your problem today?

How to use this calculator

  1. Enter your usual cycle length and how many days shorter or longer it tends to be, or enter up to 10 recent completed cycle lengths.
  2. Select Calculate, then choose Allow for cycle variation, Fixed-cycle scenario or Both.
  3. Hover over the chart on a computer, tap it on your phone, or use the day selector to read each day's estimate.

Cycle day 1 is the first day of your period. These are cycle days, not calendar dates, and the percentages describe separate possibilities for one unprotected encounter.

What the two lines mean

Allow for cycle variation averages over possible ovulation days, using your cycle-length range or recent cycles. The peak can move later with longer cycles and become broader when the possible dates are more spread out.

Fixed-cycle scenario places ovulation on one estimated day. It is a comparison of timing assumptions, not a more accurate or safer mode; a cycle's final length and ovulation day cannot be confirmed from a previous period date.

The estimates assume ovulation occurs and do not adjust for age, sperm factors, fertility conditions, breastfeeding, recent hormonal contraception or repeated intercourse. They are not a personalized pregnancy risk assessment or a method of contraception. If a period is late, this chart cannot tell you whether pregnancy has occurred.

Methodology

The chart combines a published day-relative-to-ovulation probability profile with a range of possible ovulation dates. The measured outcome is clinical pregnancy after one act of unprotected intercourse, not fertilization alone or live birth.

Li, Wilcox and Dunson report probabilities of 4%, 13%, 8%, 29%, 27% and 8% for the five days before ovulation through ovulation day. Both lines use those six values. The source model uses zero outside that interval; a zero on this chart is therefore a model assumption, not proof that pregnancy is impossible.

Bull and colleagues report average luteal-phase lengths and their spread in groups of completed cycles. The luteal phase is the time after ovulation until the next period. Its length helps relate cycle length to possible ovulation timing.

For this comparison, luteal means and standard deviations are linearly interpolated between the study's cycle-group means, with the nearest group used at the ends of the supported 15-to-50-day range. For each possible cycle length, ovulation-day weights follow a discrete bell-shaped approximation centered on cycle length minus luteal length; the weights are normalized over days 1 through the day before that cycle ends. This approximation uses population variability, not a measurement of your own ovulation uncertainty.

Without cycle history, each whole-number cycle length in your entered range receives equal weight. With history, each recorded cycle receives equal weight, including repeats. Past cycles cannot establish all possible future cycles. The fixed scenario uses the mean cycle length minus the interpolated luteal mean, rounded to the nearest day; it does not confirm ovulation.

For each displayed day, the variation-aware estimate is the weighted average of the six pregnancy probabilities across possible ovulation days. Both lines stop at the shortest modeled cycle so their points compare the same set of cycle scenarios. Neither line adds the daily percentages together or treats repeated encounters as independent.

This is a study-informed timing comparison, not the published Li calendar-day prediction model or a validated personal forecast. Combining these studies, interpolating their grouped results and choosing the weighting distributions are modeling assumptions. The underlying studies involved different populations; the estimates may not apply to you, particularly with very irregular cycles or cycles without ovulation.

Sources