Overview
The hypergeometric distribution gives the odds of drawing a certain number of 'hits' when you draw without putting items back. It's the math behind card-game draws and lotteries.
Enter how many hits are in the population, the population size, how many you draw, and the minimum number of hits you want.
Worked examples
| Inputs | Result |
|---|---|
| Successes in Population (K)12Population Size (N)60Number of Draws (n)7Minimum Hits2 | P(X >= 2): 42.81%Expected hits: 1.4 |
Show 2 more examples
| Successes in Population (K)24Population Size (N)60Number of Draws (n)7Minimum Hits3 | P(X >= 3): 58.79%Expected hits: 2.8 |
| Successes in Population (K)4Population Size (N)52Number of Draws (n)5Minimum Hits1 | P(X >= 1): 34.12%Expected hits: 0.38 |
P(X >= 2) by population size
With the calculator's starting values, the P(X >= 2) is 42.81%. The table shows how it changes with the population size.
| Population Size (N) | P(X >= 2) |
|---|---|
| 30 | 87.49% |
| 40 | 69.4% |
| 50 | 54.2% |
| 60starting value | 42.81% |
| 70 | 34.4% |
| 80 | 28.14% |
| 90 | 23.39% |
Other inputs: Successes in Population: 12 K · Number of Draws: 7 n · Minimum Hits: 2. Change them in the calculator above for your own numbers.
How it works
It counts the ways to choose your hits from the successes and the rest from the other items, divided by all possible draws. The calculator adds up the chances for every count from your minimum upward. Because nothing goes back, each draw changes the odds of the next, unlike the binomial distribution.
The formula
| K | Successes in the population |
| N | Population size |
| n | Number drawn |
| k | Number of successes drawn |
Practical tips
- With 12 hits in 60 and a draw of 7, the chance of at least 2 is 42.81%.
- In a 60-card deck with 24 lands, a 7-card hand has a 58.79% chance of 3 or more lands.
- The chance of at least one ace in a 5-card poker hand is about 34.12%.
Frequently asked questions
When do I use hypergeometric rather than binomial?
When drawing without replacement from a fixed group. Use binomial when each trial is independent.
What is the expected number of hits?
n × K ÷ N. Drawing 7 from 60 with 12 hits gives 1.4.
Can I get the chance of exactly k?
Subtract the result for k + 1 from the result for k.