Birthday Paradox Calculator

Calculate the probability that at least two people in a group share a birthday.

Shared birthday probability
Enter a group size from 1 to 365 to evaluate the classic birthday problem.

About the birthday paradox

The birthday paradox asks how likely it is that at least two people in a group share the same birthday. The result feels surprising because people often compare one chosen person with everyone else. The actual question compares every possible pair. A group of 23 contains 253 pairs, so there are many opportunities for a match even though the calendar has 365 possible birthdays. Under the standard model, a group of 23 already has a probability slightly above 50 percent of containing a shared birthday. The cleanest calculation starts with the complementary event: every person has a different birthday. The first person may have any birthday. The second must avoid that date, giving a factor of 364/365. The third must avoid two occupied dates, giving 363/365, and the multiplication continues for the whole group. Subtracting that product from one gives the probability of at least one match. In symbols, P(match) = 1 - (365/365)(364/365)...((365-n+1)/365), where n is the group size. This calculator uses the conventional simplified birthday model. It treats all 365 calendar dates as equally likely, ignores leap day, and assumes births are independent. Real birth records are not perfectly uniform because seasonality, holidays, scheduled deliveries, and demographic patterns affect daily birth rates. Those differences generally make collisions a little more likely, but the standard model is the appropriate version for probability lessons, interview questions, classroom demonstrations, and rough planning. The probability rises quickly. With 10 people it is about 11.7 percent, with 23 it is about 50.7 percent, with 30 it is about 70.6 percent, and with 50 it exceeds 97 percent. A group needs 57 people to pass 99 percent. This rapid increase illustrates a broader collision principle used in hashing, cryptography, database identifiers, and quality control: when many items are assigned to a limited number of categories, duplicates become likely much sooner than intuition suggests. Use the result as a probability, not a prediction that identifies who will match or which date will repeat. A 50.7 percent probability means that across many comparable groups of 23, just over half would contain at least one shared birthday. One particular group may have no match, one matching pair, or several repeated dates. The calculator answers the at-least-one-match question exactly for the assumptions above.

Birthday paradox examples

Group sizeMatch probabilityInterpretation
10 people11.6948%A match is possible but still relatively uncommon.
23 people50.7297%The classic threshold where a match becomes more likely than not.
30 people70.6316%About seven comparable groups in ten contain a shared birthday.
50 people97.0374%A shared birthday is overwhelmingly likely.

How to calculate shared birthday probability

  1. Enter the total number of people in the group.
  2. Click Calculate Probability to evaluate all possible birthday pairs.
  3. Read the shared-birthday percentage and the complementary all-different percentage.
  4. Change the group size to compare how quickly collision probability increases.

Birthday paradox FAQ

Why is it called a paradox?

It is not a logical contradiction; it is called a paradox because the result conflicts with common intuition. People tend to count dates rather than the rapidly growing number of pairs.

Why is the probability over 50% with only 23 people?

There are 253 distinct pairs among 23 people, and each pair is another chance for a match. Those overlapping chances combine to push the probability above one half.

Does the calculation include February 29?

The standard model uses 365 equally likely birthdays and omits leap day. Including February 29 changes the result only slightly and requires assumptions about leap-year births.

Are birthdays really equally distributed?

No, real birth frequencies vary by season and day of the week. The equal-distribution assumption creates the familiar textbook result and usually gives a useful approximation.

Can this formula find the chance of three matching birthdays?

No, this calculation finds at least one pair sharing a date. Triple matches and the expected number of matches require different occupancy calculations.