The Birthday Paradox: Why 23 People Give You 50/50 Odds of a Shared Birthday
October 8, 2026
With 23 people in a room, there's about a 50.7% chance that at least two of them share a birthday — a result that surprises most people, given there are 365 possible birthdays. The math works because the question isn't about matching one specific date; it's about any match at all among many possible pairs of people.
Skip the math and get your answer instantly:
Open the Permutation & Combination Calculator →The Math Behind It
The easiest path is calculating the probability of no match, then subtracting from 1. The probability the 2nd person doesn't share the 1st's birthday is 364/365. The probability the 3rd person doesn't share either of the first two is 363/365. Multiplying this pattern out for all 23 people and subtracting the result from 1 gives the probability of at least one shared birthday.
Probability of a Shared Birthday by Group Size
| People | Probability of a Match |
|---|---|
| 10 | ~11.7% |
| 23 | ~50.7% |
| 30 | ~70.6% |
| 50 | ~97.0% |
| 70 | ~99.9% |
The Permutation Connection
The 'no match' probability is really a permutations problem: it's the number of ways to assign 23 people distinct birthdays out of 365 days (a permutation, since order/identity matters — which person has which birthday), divided by the total number of ways to assign any 23 birthdays, including repeats (365^23). That ratio is exactly P(365,23) / 365^23, where P(365,23) is the permutation count — connecting this classic probability puzzle directly to permutation counting.
Why This Matters Beyond Trivia
The same mathematical structure — the 'birthday problem' — underlies hash collision analysis in computer science and cryptography, where it's used to estimate how many random values need to be generated before two collide (a 'birthday attack'). Any time you're asking 'what's the chance any two of N things match' rather than 'what's the chance a specific thing matches,' the surprisingly fast-growing birthday paradox math applies.
Frequently Asked Questions
Why does it only take 23 people for a 50% chance of a shared birthday?
Because the question is about any pair matching among all possible pairs of people, not one specific date matching — with 23 people there are 253 possible pairs to check, and even though each individual pair has low odds of matching, the combined probability across all those pairs adds up quickly.
How is the birthday paradox calculated?
Calculate the probability of no shared birthday first — multiplying (364/365) × (363/365) × ... down to (343/365) for 23 people — then subtract that result from 1 to get the probability of at least one match.
Is this really a 'paradox'?
Not a logical paradox — it's called that because the result feels counterintuitive given 365 possible birthdays, even though the underlying math is straightforward once you consider it's about matching among all pairs, not one specific date.
How does the birthday paradox relate to permutations?
The 'no match' probability is the number of permutations of 23 unique birthdays out of 365 possible days, divided by the total number of possible birthday assignments (365^23) — directly tying the puzzle to permutation counting.
Where is this math used outside of trivia?
In computer science and cryptography, the same structure is used to estimate hash collision likelihood — how many random values need to be generated before two happen to match — known as a 'birthday attack' in security contexts.