What the birthday paradox is
The birthday paradox (more precisely, the birthday problem) asks: in a group of n people, what is the probability that at least two of them share the same birthday? It is called a paradox because the answer defies intuition — you only need 23 people for the chance to pass 50%, and just 57 people for it to reach 99%. There is nothing self-contradictory about it; the surprise comes from confusing "someone shares my birthday" with "some pair among everyone shares a birthday."
The formula
It is far easier to compute the probability that everyone has a different birthday and subtract from 1. Assuming 365 equally likely, independent birthdays (ignoring February 29 and any seasonal clustering), the chance that all n people are unique is a product of shrinking fractions:
P(no shared birthday) = (365/365) × (364/365) × (363/365) × … × ((365 − n + 1)/365)
P(at least one shared birthday) = 1 − P(no shared birthday)
Equivalently, this is P = 1 − 365! / ((365 − n)! · 365ⁿ). Each new person must dodge every birthday already taken, so the "unique" probability multiplies by one more term that is slightly less than 1. Once n exceeds 365 the probability is exactly 1 by the pigeonhole principle — there simply are not enough distinct days to go around.
Why the number is so small
The key is that matches happen between pairs, not individuals. A group of n people contains n(n−1)/2 distinct pairs. With 23 people that is 253 pairs, each with roughly a 1-in-365 chance of matching — enough opportunities that a hit becomes more likely than not. The count of pairs grows with the square of the group size, which is why the probability climbs so steeply.
Common reference points
- 10 people → about 11.7% chance of a shared birthday
- 23 people → about 50.7% — the classic "just past even odds" threshold
- 30 people → about 70.6% (a typical classroom)
- 50 people → about 97.0%
- 57 people → about 99.0%
- 70 people → about 99.9%