Birthday Paradox Explained: Why Just 23 People?

Quick answer: The birthday paradox says that in a group of just 23 people, there is about a 50.7% chance that at least two share a birthday (ignoring leap years and assuming all 365 days are equally likely). With 70 people the chance is 99.9%. It feels surprising because we think about matching our own birthday, but 23 people form 253 different pairs, and any pair can match.

How many people do you need in a room before it is more likely than not that two of them share a birthday? Most people guess around 180, half of 365. The real answer is 23. It is not really a paradox, just a place where intuition and probability part ways.

The Trick: Compute the Opposite

Assumptions: a match means the same month and day; February 29 is ignored; all 365 dates are equally likely; birthdays are independent; and “at least two” includes groups with several matches. It is easier to find the probability that nobody shares a birthday, then subtract from 1. Line people up one at a time:

  • Person 1 can have any birthday: \(\frac{365}{365}\).
  • Person 2 must avoid person 1’s day: \(\frac{364}{365}\).
  • Person 3 must avoid both: \(\frac{363}{365}\), and so on.

For \(n\) people,

\[ P(\text{no match}) = \frac{365}{365} \cdot \frac{364}{365} \cdots \frac{365 – n + 1}{365}, \qquad P(\text{match}) = 1 – P(\text{no match}). \]

For \(n = 23\), \(P(\text{no match}) \approx 0.4927\), so \(P(\text{match}) \approx 0.5073\). Review the basics in our guide to calculating probability.

Probability Table

PeopleChance of a shared birthday
1011.7%
2041.1%
2350.7%
3070.6%
4089.1%
5097.0%
5799.0%
7099.9%
366100% (pigeonhole principle)

Why 23 Is Enough: Count the Pairs

The key is that any two people can match, not just you and someone else. With 23 people there are

\[ \binom{23}{2} = \frac{23 \times 22}{2} = 253 \]

pairs (try our combination calculator). Each pair matches with probability \(1/365\). With 253 chances, a match becomes likely. A quick approximation: \(1 – (364/365)^{253} \approx 50.05\%\), close to the exact 50.73%. It is only approximate, because the 253 comparisons are not fully independent trials: if Alice matches Bob and Bob matches Carol, Alice must match Carol. (Note that \(253/365 \approx 0.69\) is the expected number of matching pairs, not a probability.)

By contrast, the chance that someone among 22 others shares your birthday is only \(1 – (364/365)^{22} \approx 5.9\%\). To pass 50% for your own birthday you need 253 other people (254 including you). Confusing these two questions is what makes the result feel paradoxical.

The General Rule: Square Roots

With \(N\) equally likely values, a repeat becomes likely after about \(\sqrt{N}\) picks. A good approximation for a 50% chance is \(\sqrt{2N \ln 2} \approx 1.1774\sqrt{N}\). For birthdays that gives about 22.5; the exact calculation shows 23 is the first whole number that works.

Where It Matters: Hash Collisions

Computers use “hash functions” to give data short fingerprints. For an ideal \(b\)-bit hash, a generic collision search (finding any two inputs with the same fingerprint) takes about \(2^{b/2}\) tries, not \(2^b\). This “birthday attack” is why SHA-256, a 256-bit hash, is rated at 128-bit collision resistance. Finding an input that matches one given fingerprint is a different, much harder task.

Real Birthdays

Real birthdays are not perfectly uniform: some dates (like mid-September in the United States) are more common. The uniform distribution actually gives the lowest chance of a match; uneven frequencies raise it (by an amount that depends on the distribution), so 23 remains a safe answer.

Common Misconceptions

“You need about 183 people, half of 365.” No. Even for a 50% chance that someone shares your birthday you would need about 253 other people. A match between any pair is far easier.

“With 23 people a match is guaranteed.” No. It is about a coin flip; a guarantee needs 366 people (367 with February 29).

Watch: The Birthday Paradox in Video

If you learn better by watching, these videos are worth your time.

The Birthday Paradox (Vsauce2). An entertaining explanation of why 23 is enough.
Check your intuition: The birthday problem (TED-Ed). A short animated walkthrough of the calculation.
Closer Look at the Birthday Paradox – Numberphile (Numberphile). A deeper look at the mathematics.

Check Your Understanding

Five quick questions to see how much of this has stuck. Pick an answer and press Submit; you will see a short explanation either way.

Answer each question and press Submit to see your result right away.

1. About how many people give a 50% chance of a shared birthday?
Found a bug? Let us know!
2. How many pairs can 23 people form?
Found a bug? Let us know!
3. Why do we compute the probability of no match first?
Found a bug? Let us know!
4. How many people guarantee a shared birthday (ignoring February 29)?
Found a bug? Let us know!
5. What computer attack is named after this paradox?
Found a bug? Let us know!

Frequently Asked Questions

What is the birthday paradox?

It is the surprising fact that in a group of 23 people, the probability that at least two share a birthday is about 50.7%.

Why is it called a paradox?

It is not a true paradox. It only contradicts intuition, because people think about matching their own birthday rather than any of the 253 possible pairs.

How do you calculate the birthday probability?

Multiply 365/365 × 364/365 × … × (365 − n + 1)/365 to get the chance of no match, then subtract from 1.

How many people are needed for a 99% chance?

57 people give about a 99% chance of a shared birthday, and 70 people give 99.9%.

Keep Exploring

Sources and Further Reading

Related to This Article

What people say about "Birthday Paradox Explained: Why Just 23 People? - Effortless Math"?

No one replied yet.

Leave a Reply