The Birthday Paradox: Why 23 People Can Share a Birthday

3 min read

Summary

A room of 23 people has about a 50.7% chance that two people share a birthday under a simple model. Learn why the number feels surprising and what the calculation assumes.

The Birthday Paradox: Why 23 People Can Share a Birthday

Quick prediction: in a group of 23 people, what is more likely? That at least two people share a birthday, or that everyone has a different birthday?

Under a simple model, a shared birthday is slightly more likely. The chance is about 50.7%. This result is called the birthday paradox. It is a maths result that feels surprising, not a broken rule.

The calculation assumes 365 calendar days are equally likely for a birthday and ignores leap day. Real birthdays are not spread perfectly evenly, so 50.7% is a model result, not a promise about every classroom.

First, what counts as a match?

We are checking whether any two people in the group were born on the same calendar date. We are not checking whether someone shares your birthday in particular.

With 23 people, there are 253 different pairs to compare. You can find that number with this formula: 23 × 22 ÷ 2 = 253. The formula counts each pair once. It does not mean the pair comparisons are all independent, so we need a careful probability calculation.

Six dots connected by fifteen lines, showing each possible pair.
Six people create 15 pairs to compare. With 23 people, there are 253 possible pairs.

Count the chance of no match

It is easier to first work out the chance that everyone has a different birthday.

  • The first person can have any of the 365 birthdays, so the chance is 365 out of 365.
  • If that birthday is taken, the second person has 364 dates left that would avoid a match.
  • If the first two birthdays differ, the third person has 363 dates left.
  • The same pattern continues. For the 23rd person, 343 of the 365 dates would avoid a match.

Multiply those chances together. The chance that all 23 people have different birthdays is about 0.4927, or 49.27%.

There are only two outcomes in this model: everyone has a different birthday, or at least two people match. So subtract the no-match chance from 100%:

100% - 49.27% = 50.73%

That is why 23 people give a little more than a one-in-two chance of a shared birthday.

Why does the number feel so small?

Many people first imagine checking whether someone has their own birthday. In a group of 23, that gives only 22 other birthdays to compare with yours.

The birthday problem checks every possible pair in the group. With 23 people, there are 253 pairs. More pairs mean more chances for a match. Harvard’s birthday problem explanation shows how different the fixed-birthday question is from the any-pair question.

Try the idea without collecting anyone’s birthday

Draw six dots on paper to represent six people. Connect every dot to each of the other dots without drawing the same line twice. You will get 15 lines, because six people make 15 pairs.

This small picture shows why the number of comparisons grows quickly. With a larger group, there are many more pairs to check. You do not need to ask classmates for their birth dates to try the maths.

The next time someone says two people in a room share a birthday, ask which question they mean: “Does anyone match anyone?” or “Does someone match my birthday?” Probability can give different answers because the questions are different.

Share this post

Spread the word

Comments

Ulearngo
Ulearngo provides study and exam preparation tools that help students learn effectively and prepare confidently for upcoming examinations.

Ulearngo is independent and is not affiliated with or endorsed by any examination board, government agency, university, or admissions body.

Products

Resources

Company

Legal

Copyright © 2026 Ulearngo. All rights reserved.